DEL 04: Publication of the updated module handbook

20337 – EIT Health Master in Health and Medical Data Analytics

EIT Health

Erlangen | 30 March 2021

Contents

Contents ...... 0 Study track at the FAU ...... 2 Alignment ...... 2 1st year, 1st semester (autumn/winter semester) ...... 2 Innovation and Entrepreneurship 1 (10 ECTS) ...... 2 HMDA Common core (20 ECTS) ...... 6 1st year, 2nd semester (spring/summer semester): ...... 11 HMDA Common Core continuation: ...... 11 Innovation and Entrepreneurship 2 (10 ECTS) ...... 12 Master’s electives (20 ECTS) 5 ECTS to be chosen from the seminar catalogue. 5 ECTS to be chosen from the modules M5 and M8. The following modules represent an example of selectable option...... 15 2nd year, 1st semester (autumn/winter semester) ...... 19 Innovation and Entrepreneurship 3 (10 ECTS) ...... 19 HMDA Specialization (20 ECTS) ...... 23 2nd year, 2nd semester (spring/summer semester): ...... 27 Master’s thesis in collaboration with industry partner or hospital ...... 27 Study track at the UPM ...... 28 Alignment ...... 28 1st year, 1st semester (autumn/winter semester): ...... 28 Common Core UPM ...... 28 Electives UPM ...... 33 I&E UPM ...... 37 1st year, 2nd semester (spring/summer semester) ...... 40 Common Core UPM ...... 40 Electives UPM ...... 42 I&E UPM ...... 44 2nd year, 1st semester (autumn/winter semester): ...... 50 I&E...... 50 Health and Medical Data Analytics specialization (20 ECTS) ...... 51

EIT Health is supported by the EIT, a body of the European Union

2nd year, 2nd semester (spring/summer semester): ...... 57 Study track at the UGA ...... 58 1st year, 1st semester (autumn/winter semester) ...... 58 Innovation and Entrepreneurship 1 (9 ECTS) ...... 58 HMDA Common core (21 ECTS) ...... 59 1st year, 2nd semester (spring/summer semester) ...... 63 Innovation and Entrepreneurship 2 (9 ECTS) ...... 63 Master ́s electives (21 ECTS) ...... 63 2nd year, 1st semester (autumn/winter semester) ...... 68 Innovation and Entrepreneurship 3 (9 ECTS) ...... 68 HMDA specialization (21 ECTS) ...... 73 2nd year, 2nd semester (spring/summer semester) ...... 85 Master’s thesis in collaboration with industry partner or hospital 30 ECTS ...... 85 Study track at the UL ...... 88 1st year, 1st semester (autumn/winter semester): ...... 91 Innovation and Entrepreneurship 1 (10 ECTS) ...... 91 HMDA Common Core (20 ECTS) ...... 95 1st year, 2nd semester (spring/summer semester): ...... 99 Innovation and Entrepreneurship 2 (10 ECTS) –Medical specialization ...... 99 Master ́s Electives (20 ECTS) ...... 101 2nd year, 1st semester (autumn/winter semester) ...... 110 Innovation and Entrepreneurship 3 (10 ECTS) ...... 110 HMDA specialization (20 ECTS) ...... 114 2nd year, 2nd semester (spring/summer semester) ...... 119 Master’s thesis in collaboration with industry partner or hospital (30 ECTS) ...... 119

Study track at the FAU

Alignment 1st year

E-learning/on- E-learning/on- MOOC site site

HMDA Master I&E 1 I&E 1 core 20 elective 10 ECTS ECTS 10 ECTS 20 ECTS

2nd year

Practical modules (10 ECTS), incl. I&E summer school (5 ECTS)

I&E 3 speciali HMDA and I&E 10 zation Master’s thesis

ECTS 20 ECTS 30 ECTS

Study abroad min. 15 ECTS Internship(abroad) 15 ECTS

1st year, 1st semester (autumn/winter semester)

Innovation and Entrepreneurship 1 (10 ECTS) INNOVATION AND LEADERSHIP Responsible lecturer: Prof. Dr. Kathrin Möslein ECTS: 5

Course type and weekly hours: lecture (4 hours) Exam type: presentation, project report

Contents:

Creating a sustainable innovative environment is a leadership task. In order to succeed at this task, leaders must develop innovative abilities to deal with the challenges inherent in a business environment characterized by fluid, unstructured and changing information. The aim of this course is thereby twofold. First, the course delineates and describes different yet emerging innovation tools, organizing them into a coherent set of classes. Each class of tools is described using a set of up-to-date business cases that depict the current status of the information systems. The second aim of this course is to get an overview of how to structure leadership systems towards innovation, how leaders can motivate to foster innovative thinking and what new forms of innovation (e.g., open innovation) mean for the definition of leadership. In doing so, this lecture represents an Idea Transformation Class as students are encouraged not only to merely develop, but to actively deploy specifically developed concepts.

Learning outcomes and competencies: The students

• will understand and explore the theories and practicalities of leadership in open innovation contexts. • will gain knowledge on leading and communicating innovation and translate it in leadership behavior in real case contexts. • learn to assess, reflect, and feedback the impact of practical leadership for innovation. • can independently define new application-oriented problem solving in e-business in relation to the economic impact for businesses, along with solving problems using the appropriate methods. • discuss possible solutions in groups and present their research results.

Literature: Huff, Möslein & Reichwald: Leading Open Innovation; 2013 MIT Press, ISBN-13: 978- 0262018494

Keywords: innovation, leadership

DESIGNING TECHNOLOGY Responsible lecturer: Prof. Dr. Kathrin Möslein ECTS: 5 Course type and weekly hours: lecture (4 hours) Exam type: Research project and written assignments

Contents: The course covers the process of designing innovative artifacts to extend human as well as organizational capabilities, and to solve problems within organizations and industries. The course requires analytical thinking, where students can identify and clearly articulate problems that they would like to solve and the process of solution finding. While existing technical knowledge from students is

welcome, it is not a prerequisite for the course. Students can also contribute by conducting theoretical/empirical research, along with developing IT artifacts. To conclude, the course offers a balance between creativity and scientific thinking, which can be of immense help to students seeking to learn either skill or both.

Theoretical approaches which will be covered in the course:

• Design theory, systems theory, communication theories • Design science research and piloting • Agile innovation and interactive value creation

Application domains for the design projects:

• Recommendation Systems • Voice & Emotion & Pattern Recognition • Internet of Things and associated technologies

Learning outcomes and competencies: The students:

• develop a sound understanding of both social and technological aspects of various innovative technologies. • apply the design science research method, build artifacts, and evaluate them, around a given theme. • understand the design science paradigm and apply it to develop knowledge on the management and use information technology for managerial purposes. • can effectively communicate this knowledge. • are familiar with innovation technologies in the context of artificial intelligence and augmented reality that can link and enable different types of innovation technologies across the boundaries of sociotechnical systems. • adopt this knowledge in practical work on design problems, which will be related to the usage of humanoid robots for man-machine interaction.

Literature:

• Baldwin, C. Y., & Clark, K. B. (2004). Modularity in the Design of Complex Engineering Systems. In Complex Engineered Systems Understanding Complex Systems, 175–205. • Kroes, P. (2010). Engineering and the dual nature of technical artefacts. Cambridge Journal of Economics, 34 (1), 51–62. Hevner, A. R., March, S. T., Park, J. & Ram, S. (2004). Design Science in Information Systems Research. MIS Quarterly: Management Information Systems, 28 (1), 75- 106. • Fichman, R., Dos Santos, B., & Zheng, Z. (2014). Digital Innovation as a Fundamental and Powerful Concept in the Information Systems Curriculum. MIS Quarterly: Management Information Systems, 38, 329–353.

Keywords: innovation, design thinking

SERVICE INNOVATION Responsible lecturer: Prof. Dr. Kathrin Möslein ECTS: 5 Course type and weekly hours: seminar (contact hours: 30h, independent study: 120h) Exam type: Written assignments

Contents: Services now account for over 80% of all transactions in developed economies, but typically receive much less R&D attention than products. Developing service innovations demands a clear strategy from businesses with four interlocking core elements: search, selection, implementation, and evaluation of innovative concepts. If even one of these phases is not clearly thought through, the entire innovation process is likely to collapse. This course focuses on successful approaches, methods, tools, and efforts to develop service innovations.

Learning outcomes and competencies: The students can:

• learn about items, notions, characteristics, and special features in innovation management for services, service design methods and cases. • learn to judge and discuss innovation management tasks and alternative solutions with respect to the specialties of services. • experience methods of service design by themselves in interactive lectures, gain a feeling for suitable methods and learn to reflect different effects. • apply their knowledge and competences in solving cases and thereby analyze selected issues of managing, developing and innovating services. • work together in international small work groups, present their results in English, give feedback to other students’ work, and discuss different solution approaches.

Literature: Specific literature will be listed in the course

FOUNDATIONS OF INNOVATION AND ENTREPRENEURSHIP (COMING SOON) Responsible lecturer: Prof. John Bessant ECTS: 5 Course type and weekly hours: online course (4 hours) Exam type: t.b.d.

Contents: Learning outcomes and competencies: Literature: Keywords: innovation, entrepreneurship

STRATEGIC INNOVATION MANAGEMENT (COMING SOON) Responsible lecturer: Prof. Dr. Kai-Ingo Voigt ECTS: 5

Course type and weekly hours: seminar (t.b.d.) Exam type: t.b.d.

Contents: Learning outcomes and competencies: Literature: Keywords: innovation, entrepreneurship

TECHNOLOGY AND INNOVATION MANAGEMENT Responsible lecturer: Prof. Dr. Kai-Ingo Voigt ECTS: 5 Course type and weekly hours: seminar (attendance: 45h, self-study: 105h) Exam type: case studies

Contents: Technologies and innovations are of fundamental importance to companies’ growth and success. Therefore, future engineers and business executives need to be familiar with theory, concepts, and tools of technology as well as innovation management. In this regard, this course places special emphasis on economic decisions in the context of technology management, considering also disruptive changes in the business environment. Moreover, this course is about understanding success factors of innovations and the organization of innovation processes. We will additionally discuss options of timing strategies as well as special innovation concepts like open innovation and cross-industry innovation. Furthermore, the challenges and possibilities of business model innovations will be emphasized. Transforming a business idea into a structured and well-developed business plan will be the final topic of this course. All topics will be illustrated by incorporating numerous practical examples.

Learning outcomes and competencies: In this course, the students will acquire deep and comprehensive knowledge on the current state of research in the field of technology and innovation management. After finishing this course, the students will be able to assess and evaluate the crucial role of technologies and innovations as basis of competitive advantages for industry and service companies. Moreover, the students will learn to successfully transfer their acquired theoretical knowledge to practical real-world topics and to structure and solve related problems. The gained analytical and conceptual skills will enable the students to independently handle complex economic problems and to apply “the right” methods and concepts to deal with the challenges of technology and innovation management. They will also learn how to holistically reflect and present technology-or innovation-driven business ideas.

Literature: /

HMDA Common core (20 ECTS) PATTERN RECOGNITION Responsible lecturer: Prof. Dr.-Ing. Andreas Maier ECTS: 5 Course type and weekly hours: lecture (3 hours), exercise class (1 hour) Exam type: oral exam (30 min.)

Contents: Mathematical foundations of machine learning based on the following classification methods:

• Bayesian classifier • Logistic Regression • Naïve Bayes classifier • Discriminant analysis • Norms and norm dependent linear regression • Rosenblatt’s Perceptron • Unconstraint and constraint optimization • Support Vector Machines (SVM) • Kernel methods • Expectation Maximization (EM) Algorithm and Gaussian Mixture Models (GMMs) • Independent Component Analysis (ICA) • Model Assessment • AdaBoost

Learning outcomes and competencies: Students:

• understand the structure of machine learning systems for simple patterns. • explain the mathematical foundations of selected machine learning techniques explain the mathematical foundations of selected machine learning techniques. • apply classification techniques in order to solve given classification tasks. • evaluate various classifiers with respect to their suitability to solve the given problem. • understand solutions of classification problems and implementations of classifiers written in the programming language Python.

Literature:

• Richard O. Duda, Peter E. Hart, David G. Stock: Pattern Classification, 2nd edition, John Wiley & Sons, New York, 2001 • Trevor Hastie, Robert Tobshirani, Jerome Friedman: The Elements of Statistical Learning -Data Mining, Inference, and Prediction, 2nd edition, Springer, New York, 2009 • Christopher M. Bishop: Pattern Recognition and Machine Learning, Springer, New York, 2006

Keywords: pattern recognition, classification, machine learning, python programming

MACHINE LEARNING FOR TIME SERIES Responsible lecturer: Prof. Dr. Björn Eskofier, Prof. Dr. Oliver Amft ECTS: 5 Course type and weekly hours: lecture (2 hours), exercise class (2 hours) Exam type: oral exam (30 min.)

Contents: Aim of the lecture is to teach Machine learning (ML) methods for a variety of time series applications. The following topics will be covered:

• An overview of applications of time series analysis • Fundamentals of Machine learning (ML) methods, such as Gaussian processes, Monte Carlo sampling methods and deep learning, for time series analysis. • Design, implementation, and evaluation of ML methods in order to address time series problems. • Working with widely used toolboxes that can be used for implementation of ML methods, such as TensorFlow or Keras.

Learning outcomes and competencies:

• Students develop an understanding of concepts of time series problems and their wide applications in industry, medicine, finance, etc. • Students learn concepts of machine learning (ML) methods in general and tackling time series problems in particular. • Students understand the characteristics of time series data and will be capable of developing and implementing ML methods to model, predict and manipulate such data in concrete problems.

Literature:

• Machine Learning: A Probabilistic Perspective, Kevin Murphy, MIT press, 2012 • The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Trevor Hastie, Robert Tibshirani, Jerome Friedman, Springer, 2009. • Deep Learning, Ian Goodfellow and Yoshua Bengio and Aaron Courville, MIT Press, 2016 • Reinforcement Learning: An Introduction, Richard S. Sutton, and Andrew G. Barto, MIT press, 1998

Keywords: machine learning, data mining

DEEP LEARNING Responsible lecturer: Prof. Dr.-Ing. Andreas Maier ECTS: 5 Course type and weekly hours: lecture (2 hours), exercise class (2 hours) Exam type: oral exam (30 min.)

Contents: Deep Learning (DL) has attracted much interest in a wide range of applications such as image recognition, speech recognition and artificial intelligence, both from academia and industry. This lecture introduces the core elements of neural networks and deep learning, it comprises:

• (multilayer) perceptron, backpropagation, fully connected neural networks. • loss functions and optimization strategies • convolutional neural networks (CNNs) • activation functions • regularization strategies • common practices for training and evaluating neural networks. • visualization of networks and results

• common architectures, such as LeNet, Alexnet, VGG, GoogleNet • recurrent neural networks (RNN, TBPTT, LSTM, GRU) • deep reinforcement learning • unsupervised learning (autoencoder, RBM, DBM, VAE) • generative adversarial networks (GANs) • weekly supervised learning • applications of deep learning (segmentation, object detection, speech recognition, ...)

The accompanying exercises will provide a deeper understanding of the workings and architecture of neural networks.

Learning outcomes and competencies: The students:

• explain the different neural network components, • compare and analyze methods for optimization and regularization of neural networks, • compare and analyze different CNN architectures, • explain deep learning techniques for unsupervised / semi-supervised and weakly supervised learning, • explain deep reinforcement learning, • explain different deep learning applications, • implement the presented methods in Python, • autonomously design deep learning techniques and prototypically implement them, • effectively investigate raw data, intermediate results, and results of Deep Learning techniques on a computer, • autonomously supplement the mathematical foundations of the presented methods by self- guided study of the literature, • discuss the social impact of applications of deep learning applications.

Literature:

• Ian Goodfellow, Yoshua Bengio, Aaron Courville: Deep Learning. MIT Press, 2016. • Christopher Bishop: Pattern Recognition and Machine Learning, Springer Verlag, Heidelberg, 2006 • Yann LeCun, Yoshua Bengio, Geoffrey Hinton: Deep learning. Nature 521, 436–444 (28 May 2015)

Keywords: deep learning, neural networks, pattern recognition, signal processing

MULTI-OMICS DATA ANALYSIS Responsible lecturer: Prof. Mário J. Gaspar da Silva ECTS: 5 Course type and weekly hours: MOOC → Online course offered for all HMDA-students Exam type: e-exam (30 min. -on-site at the partner institutions)

Contents: 1: Introduction to epigenetics

• The genetic code • Code and personalized medicine • Genome-wide association studies • Limitations of genome-centric studies • The role of epigenetics • The central dogma • Multi-omicdata collection • The need for multi-omic data analysis • Epigenetics and personalized medicine • Epigenetics in our daily life • Case studies on identical twins • Case studies on ancestral influence

2: Essential background on (biomedical) data analysis

• Sample omic datasets • Data exploration • Data preprocessing • Clustering • Biclustering and pattern mining • Classification • Regression

3: Integrative multi-omics for personalized medicine

• Integrating multiple sources of omic data • Essentials on heterogeneous data analysis • The role of exposomics in personalized medicine • Combining multi-omic and medical data for personalized medicine • Unsupervised analysis of multi-omic data • Enrichment analysis as the way of increasing current knowledge on epigenetics. • Comprehensive study of epigenetics from integrative patterns of disease • Supervised analysis of multi-omic data • Discovery of multi-source epigenetic markers for personalized medicine

Learning outcomes and competencies:

• structured view on epigenetics and its role in personalised medicine • be familiar with current findings, opportunities, and challenges in personalized medicine (along its prevention, diagnostic and treatment components) • understand the relevance of genomic, proteomic, metabolomic, clinomic and exposomic data in epigenetics. • master essentials of supervised and unsupervised data analysis • be able to analyze multiple sources of omicdata and master principles on how to learn from heterogeneous multi-omic data.

Literature: /

Keywords: epigenetics, multi-omic data analysis, personalised medicine

1st year, 2nd semester (spring/summer semester):

HMDA Common Core continuation: PATTERN ANALYSIS Responsible lecturer: Prof. Dr.-Ing. Andreas Maier ECTS: 5 Course type and weekly hours: lecture (3 hours), exercise class (1 hour) Exam type: oral exam (30 min.)

Contents: Based on the lecture Pattern Recognition, this lecture introduces the design of pattern analysis systems as well as the corresponding fundamental mathematical methods. The lecture comprises:

• an overview over regression and classification, in particular the method of least squares and the Bayes classifier • clustering methods: soft and hard clustering • classification and regression trees and forests • parametric and non-parametric density estimation: maximum-likelihood (ML) estimation, maximum-a-posteriori (MAP) estimation, histograms, Parzen estimation, relationship between folded histograms and Parzen estimation, adaptive binning with regression trees • mean shift algorithm: local maximization using gradient ascent for non-parametric probability density functions, application of the mean shift algorithm for clustering, color quantization, object tracking. • linear and non-linear manifold learning: curse of dimensionality, various dimensionality reduction methods: principal component analysis (PCA), local linear embedding (LLE), multidimensional scaling (MDS), isomaps, Laplacian eigenmaps • Gaussian mixture models (GMM) and hidden Markov models (HMM): expectation maximization algorithm, parameter estimation, computation of the optimal sequence of states/Viterbi algorithm, forward-backward algorithm, scaling • Bayesian networks • Markov random fields (MRF): definition, probabilities on undirected graphs, Hammersley- Clifford theorem, cliques, clique potentials, examples for MRF-based image pre-processing and processing of image sequences • Markov random fields and graph cuts: sub-modular functions, global optimization with graph cut algorithms, application examples.

Learning outcomes and competencies: The students

• explain the discussed methods for classification, prediction, and analysis of patterns, • define regression and classification tasks as optimization problems, • compare and analyze methods for manifold learning and select a suited method for a given set of features and a given problem,

• compare and analyze methods for probability density estimation and select a suited method for a given set of features and a given problem, • apply non-parametric probability density estimation to pattern analysis problems, • apply dimensionality reduction techniques to high-dimensional feature spaces, • explain statistic modeling of feature sets and sequences of features, • explain statistic modeling of statistical dependencies, • implement presented methods in MATLAB or Python, • supplement autonomously the mathematical foundations of the presented methods by self- guided study of the literature, • discuss the social impact of applications of pattern analysis solutions.

Literature:

• Richard O. Duda, Peter E. Hart und David G. Stork: Pattern Classification, Second Edition, 2004 • Christopher Bishop: Pattern Recognition and Machine Learning, Springer Verlag, Heidelberg, 2006 • Antonio Criminisi and J. Shotton: Decision Forests for Computer Vision and Medical Image Analysis, Springer, 2013 • Kevin P. Murphy: Machine Learning: A Probabilistic Perspective, MIT Press, 2012 • papers referenced in the lecture.

Keywords: pattern recognition, pattern analysis

Innovation and Entrepreneurship 2 (10 ECTS) SAVE MEDICAL DEVICES Responsible lecturer: Björn-Erik Erlandsson, Ph.D., Prof. ECTS: 5 -7,5 Course type: online Exam type: written examination and project work (online)

Learning outcomes and competencies: The main objective with this course is to give the student substantial understanding about the regulatory framework for medical devices and how personal protection and intended product performance can be assured by the medical device industry and the health care sector. After passing the course, the student should be able to:

• Describe, explain, and apply in practical use the regulatory demands for medical devices. • Describe the difference between regulatory demands in different countries. • Explain the interaction between authorities, regulatory bodies, standardization organizations and industry when placing a medical device on the market. • Define quality and explain different methods for assuring quality in an organization or for products or services. • Enlarge upon the essential role of risk analysis and quality assurance for the medical device industry.

• Explain and discuss how standardization development enhances the work in the medical device industry and the healthcare sector.

Literature: /

FUNDAMENTALS OF ANATOMY AND PHYSIOLOGY Responsible lecturer: Prof. Dr. Clemens Forster ECTS: 5 Course type and weekly hours: lecture (2 hours) Exam type: written exam (90 min.)

Contents:

• fundamentals of anatomy, physiology, and pathophysiology • important medical terms • relevant and frequent clinical pictures • relevant methods in biological and technical vision • discussion of methods and theoretical approaches to recognize relevant medical questions. • critical consideration of the most important imaging techniques in important clinical pictures • presentation of the organizational structures of diagnostic processes

Learning objectives and competencies: The students:

• understand and can apply the most important and common medical terms. • are familiar with the basics of anatomy and physiology. • can interpret important clinical pictures. • understand and explain medical questions in diagnostics and therapy using examples.

MEDICAL DEVICE REGULATION Responsible lecturer: Heike Leutheuser ECTS: 2.5/5 - 2.5 ECTS are received for taking the course one semester. If taken two semesters, as a student, it is possible to receive 5 ECTS Course type and weekly hours: seminar, 5 x 8h Exam type: written exam (60 minutes)

Contents: Students must attend 5 seminar dates. The first two dates, which are offered every semester, are compulsory. You can choose the remaining 3 course days. You can also attend a course day in the following semester, if you like a topic better there, but it is advisable to complete the seminar within a semester. Seminar days are held from 9:00 am to 5:00 pm.

Learning objectives and competencies: The certificate course Medical Device Regulation offers the combination of gaining knowledge in the university environment with a seminar character and the opportunity to make contacts with students as future specialist. You get to know the legal framework for medical devices. They understand the conditions, relationships and dependencies between corresponding directives, laws, and standards. You will be able to successfully and timely take necessary measures to comply with legal requirements.

Literature: /

MEDICAL COMMUNICATIONS Responsible lecturer: Dr. Miyuki Tauchi-Brück ECTS: 2.5 Course type and weekly hours: lecture (3.5 hours) Exam type: written exam (60 min.)

Contents: Advancement in medicine is a huge collaborative work involving physicians, patients, medical professionals, engineers, scientists, and authorities to name a few. To promote and ease the development, there are rules and regulations to follow that enable interdisciplinary groups work together. Skills and knowledge for the entire structure in medical development belong to “medical communications”. This lecture is to introduce “medical communications” to undergraduate and graduate students with medicine-related majors. The contents include physicians-patients and researchers-authorities communications in relation to pre-clinical and clinical studies. The focus of the lecture is on clinical studies. Published articles in medical journals, regulatory documents, and/ or websites from different organizations will be used as study materials and active participation of students is expected.

1. Clinical studies a. Phase 0-IV clinical studies for a new drug Study designs/ terminologies Objective of studies in each phase Different study designs for different objectives Subjects Ethical issues in clinical studies Key statistics often used in clinical studies. b. Clinical study for medical devices Classification of medical devices 2. Communications a. Formality Guidelines from International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) Regulations in studies with animal subjects (preclinical studies) European legislation Regulations in studies with human subjects (clinical studies) Arzneimittelgesetz (AMG)Sechster Abschnitt: Schutz des Menschen bei der klinischen Prüfung Declaration of Helsinki Good Clinical Practice Requirement for drug approval Requirement for CE marking of medical device. b. Publication Journals: Manuscript writing/ reading Guidelines: CONSORT, STROBE, CARE, ARRIVE, etc. Terminologies: MedDRA Conferences: Oral/ poster presentation c. Patients and publication ethics Patients’ information/ informed consent. Who are patients? What patients want to know: Information source for patients

Learning objectives and competencies: The aim is to let the students:

• Understand the structures and designs of clinical studies, including drugs and medical devices. • Be aware of ethical issues in clinical studies. • Find problems and solutions in patient-physician communications. • Practice soft skills used in medical communications, including “skimming and scanning” journal articles in unfamiliar fields, summarizing, writing, and presenting data.

Literature: /

ETHICS OF (MEDICAL) ENGINEERING Responsible lecturers: Dr. rer. Nat. Jens Kirchner, Christoph Merdes ECTS: 2,5 Course type and weekly hours: lecture (2 hours) Exam type: 2 written reports (30 essays)

Contents: This course introduces the ethical reflection of engineering, with a particular focus on the area of medical technology. It offers both an elementary introduction to normative ethics and the discussion of a variety of specific ethical problems including the engineer's responsibility, over the ethics of robotics and questions of justice and allocation in the context of the deployment of high-end medical technology. The course addresses:

• the basics of utilitarianism, deontological ethics, and virtue ethics • the ethical challenges in the construction of semi-autonomous machines • the ethical role and efficacy of professional codes • the just allocation of resources in society from the vantage point of medical technology • the responsibility of engineers and the ethical aspects of whistleblowing • dealing with test subjects and personal data

Literature:

• Kraemer, F., Van Overveld, K., & Peterson, M. (2011). Is there an ethics of algorithms? Ethics and Information Technology, 13(3), 251-260. • Kant, I. (1996[1785]). Groundworks for the metaphysics of morals. Kant's Practical Philosophy, Wood Allen & Gregor, Mary (ed.), Cambridge University Press, pp. 37-108.

Keywords: ethics, ethics of engineering, ethics of medicine

Master’s electives (20 ECTS) 5 ECTS to be chosen from the seminar catalogue. 5 ECTS to be chosen from the modules M5 and M8. The following modules represent an example of selectable option. KNOWLEDGE DISCOVERY IN DATABASES Responsible lecturer: Klaus Meyer-Wegener ECTS:2.5 Course type and weekly hours: lecture (2 hours) Exam type: oral exam (30 minutes)

Contents:

• what is data mining and why do we do it? • a multi-dimensional view of data mining • what kinds of data or patterns can be mined? • what technologies are used?

• what kinds of applications are targeted? • major issues in data mining • a brief history of data mining

Learning objectives and competencies:

• know the typical KDD process. • know procedures for the preparation of data for data mining. • know the definition of distance or similarity functions for the different kinds of attributes. • define distance and similarity functions for a particular dataset. • check attributes of a dataset for their meaning with reference to an analysis and transform attribute values accordingly, if required. • know how a typical data warehouse is structured. • are familiar with the principle of the Apriori algorithm for the identification of frequent itemset. • know the FP-growth algorithm for a faster identification of frequent itemset: • present the definitions of support and confidence for association rules. • describe the construction of association rules based on frequent itemset. • are capable of describing the course of action in classification tasks. • present the construction of a decision tree based on a training dataset. • present the principle of Bayes' classification. • enumerate different clustering procedures. • describe the steps of k-means clustering. • know the different kinds of outliers.

Literature:

• Jiawei Han, Micheline Kamber, and Jian Pei: Data Mining –Concepts and Technologies, 3rd ed. Waltham, MA: Morgan Kaufmann, 2012 (The Morgan Kaufmann Series in Data Management Systems). -ISBN 978-0-12-381479-1

ARCHITECTURE FOR DIGITAL SIGNAL PROCESSING Responsible lecturer: Dr. rer. nat Dr. phil. Jens Kirchner ECTS: 5 Course type and weekly hours: lecture (2 hours) Exam type: E-exam

Contents:

• Basic algorithms of signal processing (FFT, windowing, digital FIR, and IIR-filters) • Non-idealities of digital filters (quantization of filter coefficients, fixed-point arithmetic) • CORDIC-architectures • Architectures of systems with multiple sampling rates (conversion between different sampling rates) • Digital signal generation • Measures of performance improvement (pipelining) • Architecture of digital signal processors • Applications

Learning objectives and competencies: Students

• can obtain fundamentals of signal theory and can define as well time-continuous and value- continuous as time-discrete and value-discrete signals in time and frequency domain. • can construct a real time digital signal processing system and dimension its components according to requirements. • can review pros and cons of analogue versus digital signal processing. • can apply Fourier transformation and illustrate the advantages of fast Fourier transformation in the context of digital signal processing. • can dimension digital filters and evaluate their performance.

Literature: /

MEDICAL IMAGING SYSTEM TECHNOLOGY Responsible lecturer: Dr.-Ing. Wilhelm Dürr ECTS: 5 Course type and weekly hours: lecture with exercise (4 hours) Exam type: written exam

Contents: Röntgen ́s discovery of "a new kind of ray" about 100 years ago was the beginning of the partially spectacular development of imaging systems for medical diagnosis. New knowledge and developments, especially in physics, led to consequent applications in the area of medicine. Overtime, there developed the following (most significant) medical imaging techniques: roentgenography, nuclear medical imaging, sonography, x-ray computer tomography and magnetic resonance tomography. After an overview of the historical development and the basic principles of the theory of linear systems involved, the individual techniques will be discussed in detail. Following the description of the functional principles, the point of concentration will lie in the technical realization. Biological, physical, and technical limits are to be described. What is possible today, is to be shown through examples in application.

Learning objectives and competencies:

• know the basics of physics and technology of X-ray systems, nuclear medical imaging, sonography, X-ray computer tomography and magnetic resonance technology. • can describe and explain the functioning of medical imaging systems. • are familiar with the application spectrum and can discuss advantages and disadvantages of the various modalities. Literature:

• Fercher, A.F.: Medizinische Physik. Springer-Verlag, 1992 • Morneburg, H. (Hrsg.): Bildgebende Systeme für die medizinische Diagnostik. Publicis-MCD- Verlag, 1995 • Rosenbusch, G., Oudkerk, M., Amman, E.: Radiologie in der medizinischen Diagnostik. Blackwell

ADVANCED MEDICAL IMAGING FOR CLINICAL NAVIGATION USING SMART DEVICES Responsible lecturer: Dr. Björn Heismann (Siemens Healthineers)

ECTS: 5 Course type and weekly hours: advanced seminar (2 hours), project work (2 hours)

Contents: You will be part of an interdisciplinary project team challenged to prototype a novel (wearable) device for the emerging field of digital indoor navigation. You and your team will have the chance to develop your ideas and shape patient care of the future. Experts from industry, healthcare and research will give you valuable feedback and support you to bring your idea to the next level.

Learning outcomes and competencies: Stage 1:

• Challenge introduction and lectures on state-of-the-art methodology, algorithms and sensors used for indoor navigation. • Project management workshop (e.g., design thinking approach, prioritizing exercise & introduction to useful PM tools) • Clinical needs screening and interviews with healthcare experts • Ideation process • Development of concrete hardware/software design specifications. You will be guided in your team and project to set realistic milestones and iteratively prioritize your technical implementation.

Stage 2:

• Data capturing • Rapid prototyping • Feedback round with potential users and customers

Stage 3:

• Embracing user feedback and final hack sprint • Prototype pitch with experts from healthcare and industry

Literature: /

HUMAN COMPUTER INTERACTION Responsible lecturer: Prof. Dr. Björn Eskofier ECTS: 5 Course type and weekly hours: lecture (3 hours), exercise class (1 hour) Exam type: written exam (90 min.)

Contents: Aim of the lecture is to teach basic knowledge of concepts, principles, models, methods, and techniques for developing highly user-friendly Human Computer Interfaces. Beyond traditional computer system the topic of modern user interfaces is also discussed in the context of automobile and intelligent environments, mobile devices, and embedded systems. This lecture addresses the following topics:

• Introduction to the basics of Human Computer Interaction • Design principles and models for modern user interfaces and interactive systems • Information processing of humans, perception, motor skills, properties, and skills of the users • Interaction concepts, metaphors, standards, norms, and style guides • In-and output devices, design space for interactive systems • Analysis-, design-and development methodologies and tools for easy-to-use user interfaces • Prototypic implementation of interactive systems • Architectures for interactive systems, User Interface Toolkits, and components • Acceptance, evaluation methods and quality assurance

Learning outcomes and competencies:

• Students develop an understanding for models, methods, and concepts in the field of Human- Computer Interaction. • They learn different approaches for designing, developing, and evaluating User Interfaces and their advantages and disadvantages. • Joining the course enables students to understand and execute a development process in the area of Human-Computer Interaction. • Student will be able to do an UI evaluation by learning basics about Information processing, perception, and motoric skills of the user. • Additionally, appropriate evaluation method as well as acceptance and quality assurance aspects will be learned.

Literature: /

Keywords: human-computer interaction, human machine interface, mobile human computer interaction, ubiquitous and embedded interactive systems

2nd year, 1st semester (autumn/winter semester)

Innovation and Entrepreneurship 3 (10 ECTS) THE HMDA Ś SCHOOL ON LEARNING FROM HEALTH DATA Responsible lecturer: various ECTS: 5 Course type: workshop/laboratory course (150 hours in total, including self-study and preparation of tasks in teams) Exam type: /

Contents: With its yet unfulfilled promise to revolutionize the healthcare economy and save billions of euros in the process, Artificial Intelligence (AI) and health data management in general are exploding in popularity. Indeed, the growth of the global AI health market is expected to reach US$6.6 billion by 2021.

But can AI and data-driven technologies truly live up to expectations in the field of health?

Over 5 demanding days at this exciting bioHealth Computing school, graduate students (Master & PhD) and early career professionals in science, informatics and healthcare are immersed in a challenging mix of theoretical and practical sessions on AI technology and innovation and coached to develop business models of market-acceptable products and services using AI technologies.

Learning from Health Data is an accelerated learning programme proposed by a consortium of EIT- Health partner universities and co-organised by the Université Grenoble-Alpes and ESI-Archamps.

The school is fully in line with EU goals to deliver innovation-led solutions enabling European citizens to live longer, healthier lives. The school adheres to the 2030 Agenda for Sustainable Development of the UN, and in particular to the objectives of the UHC2030 programme whose mission is to create a movement for accelerating equitable and sustainable progress towards universal health coverage (UHC).

The application form includes a section where candidates should provide a 50 to 200-word outline of an innovative idea or project related to health and medical data analytics. This might be expressed in terms of:

• An unmet need in healthcare which could benefit from the development of data-driven products or services. • The (re)deployment of an existing technology to provide an innovative product or service for healthcare. • A currently unavailable but potentially marketable product or service involving data-driven technology for healthcare.

The best ideas may serve as the basis for a group project in the Business Development & Innovation component of the school.

Learning outcomes and competencies: Students can

• apply industry standard techniques and team management. • develop, independently and within a team, their problem-solving and creative skills. • implement their ideas as prototypes by applying agile software development methods. • use the results of their projects for the creation of start-ups. • develop world-class solutions in the field of IT and health technologies, address societal challenges, contribute to the competitiveness of Europe, • focus on unmet needs in healthcare, AI candidate technologies, experienced-based co-design, business creation, health assessment and regulatory affairs. • participate a series of advanced courses and hands-on activities on IT proposed by experts from partner universities, hospitals, and industries. One breakout session on advanced application in health • will examine several uses of machine learning, big data and internet of things presented by leaders in Health Research and Development, reviewing the latest techniques. • develop innovative ideas in multidisciplinary teams translating them into value creation through a business model and under the health regulation framework.

INNOVATION LAB FOR WEARABLE AND UBIQUITOUS COMPUTING Responsible lecturer: Prof. Dr. Eskofier

ECTS: 10 Course type and weekly hours: internship (4 hours) Exam type: internship, graded certificate

Contents: Mini-computers documenting our rhythm of life, EKG-Sensors tracing every detail or glasses, that transfer us into another reality are amongst the technologies we are meanwhile facing in our everyday lives. At the Innovation Lab for Wearable and Ubiquitous Computing students develop such technologies and learn about the possibilities and requirements to build a start-up. The Lab is funded by the Center of Digitalization Bavaria (ZD.B). By applying agile development methods (Scrum), teams of 5 to 8 students develop prototypes of products within the wearable and ubiquitous computing field. Participating students have open access to the Innovation Lab, which provides them with everything they need to develop their prototypes. The project ideas originate from cooperating companies or the students themselves. Besides the great practical experience gained during development, students also learn about entrepreneurship. There will be tutorials covering design thinking, market analysis, management of development processes, securing intellectual property, and business plan creation.

Learning objectives and competencies:

• Ideation, Design Thinking • Patent Research, Markt Analysis • Agile Development Methods (Scrum) • Prototyping • Securing Intellectual Property • Introduction to Entrepreneurship, Startup Financing

Literature: /

MEDTECH ENTREPRENEURSHIP LAB (EITCAMPUS PROJECT) Responsible lecturer: Heike Leutheuser ECTS: 10 Course type and weekly hours: lab course (hours t.b.d.) Exam type:

The MedTech E-Lab participants will be assessed weekly and learners will receive useful, documented feedback throughout the semester. At the end of each semester, a jury of healthcare and business professionals will evaluate each team's prototype and presentation at a demo day and pitch event. Every participant will be graded. The overall grade consists of four parts:

• Mid-semester presentation (30%) • Report (20%) • Code, Scrum Meeting, Practical work (40%) • Team performance (10%)

10 ECTS will be awarded after the successful completion of the semester long course. All MedTech E-Lab learners will receive a certificate after they have successfully completed all of the program components.

Contents: The Central Institute of Medical Engineering (ZiMT) is offering, in collaboration with the University Hospital Erlangen, Siemens Healthcare GmbH, Medical Valley EMN and Universidad de Navarra/IESE a new course in entrepreneurship and medical engineering: The MedTech Entrepreneurship Lab. This practical course is geared towards students who are keen to solve practical real-world challenges from hospital partners, whilst using a hands-on approach and acquiring relevant tools to build and launch a business.

The semester long MedTech Entrepreneurship Lab will work with teams of learners from different disciplines to solve practical questions in medical technology and digital health. Students will work on pre-selected projects from the University Hospital of Erlangen -for instance in the field of Neurology, Anesthesiology, Gastroenterology, and Palliative Medicine. Participants will have open access to the Innovation Lab, which provides them with an infrastructure to develop their project specific prototype. Through a series of seminars in medical device regulations, workshops in business and finance, regular mentoring meetings, and individual coaching from industry and healthcare professionals, learners will implement their ideas as prototypes by applying lean startup methodology and develop their business plan. The course will be offered in the summer and winter semester and will culminate in a pitch competition in front of investors, end-users, and challenge-providers and the general public. Learning goals and skills:

• Ideation, Design Thinking • Market Analysis • Patent Research / Securing Intellectual Property • Business planning • Prototyping • Introduction to Entrepreneurship, Startup Financing • Medical device / software law • Clinical evaluation

Scrum will be used as an agile development tool in order to support the students in their prototyping process. Besides the great practical experience gained during development, students will gain knowledge in entrepreneurship and business creation, subsequently empowering them to turn an idea into a startup venture. The MedTech Entrepreneurship Lab is funded by EIT Health.

Learning outcomes and competencies: Students

• gain knowledge in entrepreneurship and business creation. • acquire competencies and skills in design thinking, business models, value proposition, prototyping, financing, and pitching. • learn about medical device and medical software regulations in Europe and abroad. • work in interdisciplinary teams addressing real-world challenges in healthcare. • gain knowledge about decision-making related to product development in the healthcare sector. • develop innovative, needs-based products together with hospital and industry partners be empowered to turn an idea into a startup venture.

Prerequisites: Organizational information At the beginning of each term, clinical projects from the University Hospital of Erlangen will be presented and subsequently assigned to multidisciplinary student groups (minimum of 5). Timing for the meetings will be scheduled in the first week. Registration via StudOn from the 16.03.2020 till the 21.04.2020.

Application procedure Seminar places will be assigned by a first come - first serve basis. In the case that we receive too many applications, there will be a waiting list. However, we reserve the right to choose applicants that best fit the target group description. The course is open to students of the before mentioned study fields (Bachelor 5th semester or higher and all Master students).

For other study programs or information to ECTS distributions, please contact us: mailto:zimt- [email protected]

HMDA Specialization (20 ECTS) WEARABLE AND IMPLANTABLE COMPUTING Responsible lecturer: Prof. Dr. Oliver Amft ECTS: 5 Course type and weekly hours: lecture (2 hours), exercise class (2 hours) Exam type: written exam (60 min.)

Contents: The course provides an overview on the system design of wearable computing systems and implantable systems. Electronic design topics will be addressed, including bioelectronics, flexible electronics, electronics textile integration, multiprocess additive manufacturing. On the system functional level, frequent sensor and actuators and their designs for on-body and implantable systems are discussed. Powering and energy management concepts will be detailed, including processing and task scheduling, sparse sampling, and sparse sample signal processing. Energy harvesting methods for wearable and implantable systems are analyzed. Principles of biocompatibility and system validation for remote health monitoring are covered. Concrete design problems related to context awareness, energy-efficient context recognition, and mechanical design in medical applications are demonstrated, prototypes realized and discussed in mini projects.

Learning outcomes and competencies: Students:

• gain overview on context awareness, sensors, and actuators for context management in digital health. • understand design concepts and apply/analyse wearable and implantable system design methods for accessories, smart textiles, skin-attachable using soft substrates, and encapsulation. • analyse the electrical and physical principles, select and optimize on-body energy harvesting and power management techniques.

• apply system evaluation methods, assess, and design for biocompatibility. • create continuous context recognition and energy-efficient processing using sparse sampling, related signal, and pattern processing methods. • create digital models of wearable systems.

Literature: /

Keywords: wearables, digital health

BIOMEDICAL SIGNAL ANALYSIS Responsible lecturer: Prof. Dr. Björn Eskofier ECTS: 5 Course type and weekly hours: lecture (2 hours), exercise class (2 hours) Exam type: written exam (90 min.)

Contents: The lecture content explains and outlines (a) basics for the generation of important bio signals of the human body, (b) measurement of bio signals, and (c) methods for bio signals analysis. Considered bio signals are among others action potential (AP), electrocardiogram (ECG), electromyogram (EMG), electroencephalogram (EEG), or mechanomyogram (MMG). The focus during the measurement part is for example the measurement technology or the correct sensor and electrode placement. The main part of the lecture is the analysis part. In this part, concepts like filtering for artifact reduction, wavelet analysis, event detection or waveform analysis are covered. In the end, an insight into pattern recognition methods is gained.

Learning outcomes and competencies: Students

• reproduce the generation and measurement of important bio signals of the human body. • recognize relations between the generation of bio signals and the measured signal. • understand the importance of bio signal analysis for medical engineering. • analyze and provide solutions to the key causes for artifacts in bio signals. • apply gained knowledge independently to interdisciplinary research questions of medicine and engineering science. • acquire competences between medicine and engineering science. • learn how to reproduce and argumentatively present subject-related content. • understand the structure of systems for automatic classification of simple patterns. • work cooperatively and act responsibly in groups. • implement bio signal processing algorithms in MATLAB. • solve classification problems in MATLAB.

Literature:

• R.M. Rangayyan, Biomedical Signal Analysis: A case-study approach. 1st ed., 2002, New York, NY: John Wiley & Sons.

• E.N. Bruce, Biomedical Signal Processing and Signal Modeling. 1st ed., 2001, New York, NY: John Wiley& Sons.

DIAGNOSTIC MEDICAL IMAGE PROCESSING Responsible Lecturer: Prof. Dr.-Ing. Andreas Maier ECTS: 5 Course type and weekly hours: online course (2 hours) and online exercises (2 hours) Exam type: written exam (60 min.)

Contents: The contents of the lecture comprise basics about medical imaging modalities and acquisition hardware. Furthermore, details on acquisition-dependent preprocessing are covered for image intensifiers, flat- panel detectors, and MR. The fundamentals of 3D reconstruction from parallel-beam to cone-beam reconstruction are also covered. In the last chapter, rigid registration for image fusion is explained. In the exercises, algorithms that were presented in the lecture are implemented in Java.

Learning outcomes and competencies: The participants

• understand the challenges in interdisciplinary work between engineers and medical practitioners. • develop understanding of algorithms and math for diagnostic medical image processing. • learn that creative adaptation of known algorithms to new problems is key for their future career. • develop the ability to adapt algorithms to different problems. • are able to explain algorithms and concepts of the lecture to other engineers.

Literature: /

INTERVENTIONAL MEDICAL IMAGE PROCESSING Responsible Lecturer: Prof. Dr.-Ing. Andreas Maier ECTS: 5 Course type and weekly hours: online course (4 hours) Exam type: written exam (60 min.)

Contents: This lecture focuses on recent developments in image processing driven by medical applications. All algorithms are motivated by practical problems. The mathematical tools required to solve the considered image processing tasks will be introduced. The lecture starts with an overview on preprocessing algorithms such as scatter correction for x-ray images, edge detection, super-resolution, and edge-preserving noise reduction. The second chapter describes automatic image analysis using feature descriptors, key point detection, and segmentation using bottom-up algorithms such as the random walker or top-down approaches such as active shape models. Furthermore, the lecture covers geometric calibration algorithms for single view calibration,

epipolar geometry, and factorization. The last part of the lecture covers non-rigid registration based on variational methods and motion-compensated image reconstruction.

Learning outcomes and competencies: The participants

• summarize the contents of the lecture. • apply pre-processing algorithms such as scatter correction and edge-preserving filtering. • extract information from images automatically by image analysis methods such as key point detectors and segmentation algorithms. • calibrate projection geometries for single images and image sequences using the described methods. • develop non-rigid registration methods using variational calculus and different regularizes. • adopt algorithms to new domains by appropriate modifications.

Literature: /

VISUAL COMPUTING IN MEDICINE Responsible Lecturer: Peter Hastreiter ECTS: 5 Course type and weekly hours: online course (4 hours) Exam type: oral exam (30 min.)

Contents: The flood and complexity of medical image data as well as the clinical need for accuracy and efficiency require powerful and robust concepts of medical data processing. Due to the diversity of image information and their clinical relevance the transition from imaging to medical analysis and interpretation plays an important role. The visual representation of abstract data allows understanding both technical and medical aspects in a comprehensive and intuitive way.

Learning outcomes and competencies: Based on a processing pipeline for medical image data an overview of the characteristics of medical image data as well as fundamental methods and procedures for medical image analysis and visualization is given. Examples of clinical practice show the relation to the medical application:

• Overview of imaging techniques in medicine • Grid structures, data types, image formats • Preprocessing, filtering and interpolation. • Fundamental approaches of segmentation • Explicit and implicit methods of registration • Medical visualization (2D, 3D, 4D) of scalar-, vector-, tensor data • Practical demonstrations in the labs and in the clinics

Literature:

• B. Preim, C. Botha: Visual Computing for Medicine, Morgan Kaufmann Verlag, 2013

• B. Preim, D. Bartz: Visualization in Medicine - Theory, Algorithms, and Applications, Morgan KaufmannVerlag, 2007 • H. Handels: Medizinische Bildverarbeitung, Bildanalyse, Mustererkennung und Visualisierung für diecomputergestützte ärztliche Diagnostik und Therapie, Vieweg und Teubner Verlag, 2009 • P.M. Schlag, S. Eulenstein, Th. Lange: Computerassistierte Chirurgie, Elsevier Verlag, 2010 • E. Neri, D. Caramella, C. Bartolozzi: Image Processing in Radiology, Springer Verlag, 2008

2nd year, 2nd semester (spring/summer semester):

Master’s thesis in collaboration with industry partner or hospital Supervisors: 1 academic supervisor, 1 technical supervisor (industry partner), 1 medical supervisor (physician or member of the Faculty of Medicine) ECTS: 30 Exam type: written thesis and oral presentation

By writing the Master ́s thesis, students should learn to execute scientific research in the field of medical engineering. To get familiar with the concept of scientific work, the FAU offers the course “Nailing your Thesis” in summer semester. The topic is issued and supervised by a cooperation of a fulltime university teacher and a non-academic business partner. The thesis results out of a previous internship to combine the technical research with study tract distinctive business and entrepreneurship components. It is recommended to start looking for a final thesis subject in the second year of study. This way, students have time to take matching study modules to acquire specialized knowledge. Registration is possible for all students, that have completed all mandatory modules and have acquired at least 75 ECTS credits. From there on, the thesis is limited to 6 months, which correspond to 30 ECTS or 900 hours of work. Within the deadline, the thesis paper must be handed in at the supervising lab. The thesis presentation and the grading can take place after the deadline.

Study track at the UPM

Alignment 1st year

E-learning/on- MOOC site

HMDA I&E 1 Master core 20 elective 10 ECTS ECTS 20 ECTS

2nd year

Practical modules (10 ECTS), incl. I&E summer school (5 ECTS)

I&E 3 speciali HMDA and I&E 10 zation Master’s thesis

ECTS 20 ECTS 30 ECTS

Study abroad min. 15 ECTS Internship(abroad) 15 ECTS

1st year, 1st semester (autumn/winter semester):

Common Core UPM STATISTICAL DATA ANALYSIS Responsible lecturer: Arminda Moreno ECTS: 4.5

Course type and weekly hours: 1hour of lecture + 2 hour of laboratory Exam type: written exam (3 hours) + group presentation (3 hours)

Contents: The course is intended to be a non-exhaustive survey of techniques to convert multivariate data into useful information so that good decisions can be made. The perspective is twofold, theorical and applied, covering topics such as: exploratory data analysis, statistical summaries and graphical representations, dimensionality reduction, regression techniques, time series analysis, decision theory and probabilistic graphical models. There will be an emphasis on hands-on application of the theory and methods throughout, with extensive use of R.

Syllabus:

1. Descriptive statistics and statistical modelling. 1.1. Aspects of multivariate data. Descriptive statistics. Introduction to R. 1.2. Dimensionality reduction: Principal Component Analysis and biplots. 1.3. Regression models. 1.4. Discrimination analysis and clustering. 2. Time Series. 2.1. Definitions, Applications and Techniques. 2.2. Stationarity and Seasonality. 2.3. Common approaches. 2.4. Box-Jenkins model identification, estimation, and validation. 2.5. Forecasting 3. Introduction to Decision Analysis. 3.1. Structure and representation of a decision problem. 3.2. Decision making under certainty and uncertainty. 3.3. Preferences and beliefs modelling. 3.4. Collective decision making. 4. Graphical Models for Decision Making. 4.1. Decision Trees and Influence Diagrams for optimal decisions. 4.2. Bayesian networks for diagnosis and prognosis. 4.3. Sensitivity Analysis for explanation of reasoning

Learning outcomes and competencies:

• To perform a time series analysis using the proper statistical methodology. • To build, interpret and conduct diagnostics analysis of regression models. • To apply the expected utility paradigm to solve decision problems. • To use multivariate data representation and dimensionality reduction techniques. • To build, estimate and interpret probabilistic graphical models.

Literature:

• Johnson, R.A., Whichern, D.W. (2007) Applied Multivariate Statistical Analysis. Pearson Education • Rencher, A.C. Methods of Multivariate Analysis.

• Everitt, B.S. and Dunn G. (1997) Applied Multivariate Data Analysis. Arnold. • Hair, J.F., Black, W.C., Babin, B.J., Anderson R.E. Multivariate Data Analysis. • Sharma, S (1996). Applied Multivariate Techniques. Wiley.

Keywords: Statistics, regression models, temporal series, PCA descriptive statistics

MACHINE LEARNING Responsible lecturer: Prof. Pedro Larrañaga / Prof. Concha Bielza ECTC: 4.5 ECTS Course type and weekly hours: lecture, assessment activity Exam type: written exam + practical project.

Contents: The amount of data generated in the healthcare sector is growing exponentially, and intelligent systems able to transform this huge quantity of data into knowledge, as represented by mathematical and statistical models, are more than necessary. Machine learning is a part of artificial intelligence that allows to build those models. Machine learning comprises several methods enabling this transformation in such a way that the resulting software systems can provide actionable insights towards optimal decisions. This course covers four groups of techniques: supervised classification, unsupervised classification, probabilistic graphical models and spatial statistics. The course includes theoretical and applied lessons, with specialized software tools used to solve practical problems.

Syllabus:

1. Introduction to machine learning 2. Supervised classification 2.1. Performance evaluation 2.2. Feature subset selection 2.3. Non-probabilistic classifiers: k-nearest neighbors, classification trees, rule induction, support vector machines 2.4. Probabilistic classifiers: discriminant analysis, logistic regression, Bayesian network classifiers 2.5. Metaclassifiers 2.6. Multi-dimensional classifiers 3. Unsupervised classification 3.1. Non-probabilistic clustering: hierarchical, partitional, subspace clustering, cluster ensembles, evaluation criteria 3.2. Probabilistic clustering: the EM algorithm, finite mixture models, clustering with Bayesian networks 4. Probabilistic graphical models 4.1. Bayesian networks: basics, inference, learning, dynamic Bayesian networks. 4.2. Markov networks: basics, inference, learning, conditional random fields. 5. Spatial statistics 5.1. Spatial point processes 5.2. Complete spatial randomness

5.3. Goodness-of -fit test via simulation. 5.4. Common models: cluster, regular, Gibbs

Learning outcomes and competencies:

•To be able to identify the appropriate (supervised and unsupervised) classification techniques to solve a given real-world problem •To learn probabilistic graphical models, perform inferences and interpret the structure, parameters and conditional independences •To perform a spatial analysis using the proper spatial point process methodology •To be able to apply machine learning software tools for practical problems

Literature:

•A. Baddeley, E. Rubak, R. Turner (2015) Spatial Point Patterns: Methodology and Applications with R, Chapman, and Hall/CRC •C. Bielza, P. Larrañaga (2019) Data-Driven Computational Neuroscience, Cambridge University Press, to appear •R. Duda, P.E. Hart, D.G. Stork (2001) Pattern Classification. Wiley •D. Koller and N. Friedman (2009) Probabilistic Graphical Models: Principles and Techniques. The MIT Press •K.P. Murphy (2012) Machine Learning: A Probabilistic Perspective. The MIT PressKeywords: Supervised classification, unsupervised classification, probabilistic graphical models, spatial statistic

Keywords: Supervised classification, unsupervised classification, probabilistic graphical models, spatial statistics

DATA PROCESSES Responsible lecturer: Ernestina Menasalvas Ruíz ECTS: 4.5 Course type and weekly hours: 2 hours: 1h lecture + 1 laboratory Exam type: 2 hour written test + 2 assignments.

Contents: In this course we will deepen on the importance of data for an organization. In fact, the course is centered on the process of extraction of knowledge from databases as a support for decision making. Data Science project development will be central to the course. This course will be adapted depending on the students’ profile, but main goal will be to deepen on the importance of data for an organization and deepen on the development of data science projects. The course will start with the definition of data science projects and will analyze on the one hand the importance to map business needs to data mining problems and on the other the importance of understanding the data sources in the organization. We will continue the course with understanding the potential of data analysis in the health domain. Later students will understand the data value chain and will go deep into the process of knowledge extraction. At this stage CRISP-Dm methodology will be used. The course will follow on going deeper into the different phases of the process: in) business

understanding, ii) data understanding, iii) data preparation, iv) modeling v) evaluation and vi) deployment. Through all the phases the main emphasis will be on students getting hands on the different steps, techniques, algorithms and tools. Before finishing the course will cover basic aspects of the GDPR and the implications on the process of knowledge extraction in a company. The course will complement all the lectures with use cases in the health domain

Syllabus:

1. Introduction 1.1. Course description. 1.2. Data Science and Data Scientist Skills. 1.3. The Value hidden in data. 2. Data Science in the Health domain 2.1. Data Sources in the health domain. 2.2. Main challenges 3. The process of Knowledge Discovery in Databases 3.1. CRISP-DM 4. Business Understanding 4.1. Goal of BU. 4.2. Planning of a Data Science project. 5. Data Understanding 5.1. Understanding data. 5.2. Nulls and outlier’s detection. 5.3. Correlation analysis 6. Data Preparation 6.1. Preparing data for mining: dealing with problems encountered in understanding, transforming data, discretization, data reduction, aggregation. 7. Data mining/data modeling 7.1. Type of problems. Data nature, data problems and possible algorithms. 7.2. Classification, association, and clustering 8. Evaluation and Deployment 8.1. Evaluation of the models. 8.2. Deployment of the models 9. Ethics 9.1. GDPR and implications in Data Science

Learning outcomes and competencies:

• The ability to propose a well-founded approach in any domain where big data can play a role. • The capacity to identify and link the key issues related to the use of big data in the main economic, industrial, societal, and scientific domains.

Literature:

• Ian Witten, Eibe Frank, Mark Hall, Data Mining: Practical Machine Learning Tools and Techniques, 3rd Edition, Morgan Kaufmann, ISBN 978-0-12-374856-0, 2011.

• Smart Machines: IBM's Watson and the Era of Cognitive Computing. Columbia University Press (October 15, 2013) • Database Systems: The Complete Book (DS:CB), by Hector Garcia-Molina, Jeff Ullman, and Jennifer Widom • Healthcare Data Analytics (Chapman & Hall/CRC Data Mining and Knowledge Discovery Series). Chandan K. Reddy, Charu C. Agg

Keywords: CRISP-DM workflow, data mining, data science, medical data analysis, data lifecycle, preprocessing, evaluation

Electives UPM BIG DATA Responsible lecturer: Antonio Latorre de la Fuente ECTS: 6 Course type and weekly hours: 3 hours Exam type:1 hour written test + 2 practical assignment.

Contents: This course will allow the student to gain the fundamentals for the analytical visualization of large volumes of data. With an eminently practical approach, the technologies, and fundamentals necessary to successfully accomplish the whole data analysis process will be presented in the context of Big Data, from the raw data to its visualization, through the models derived from them.

Syllabus

1. Introduction to Big Data 1.1. Architectures and applications 1.2. Data types 1.3. Visual analytics 2. Big Data Ecosystem 3. Big Data Technologies 3.1. Technological Challenges 3.2. Basic solution: gfs + MapReduce 3.3. Hadoop (hdfs + yarn) 3.4. Pig 3.5. Hive 3.6. Beyond MapReduce 3.6.1. Tez 3.6.2. Spark 3.6.3. Flink 4. Spark 4.1. Spark Basics 4.2. Brief Introduction to Scala 4.3. Spark Applications 4.4. Spark SQL 5. Machine Learning with Spark

5.1. Brief review of Machine Learning basics 5.2. Spark MLlib 6. Information Visualization 6.1. Information Visualization Fundamentals 6.2. Data Abstractions 6.3. Tasks Abstractions 6.4. Interaction Techniques and Visual Encoding 6.5. Design Methods 6.6. Visualization Examples Analysis 6.7. Lessons Learnt

Learning outcomes and competencies:

• •To learn how scientific computing techniques are applied in a specific field of science or engineering • •To know techniques of visualization and processes of data analysis, and of programming, design and debugging of algorithms, for high performance computing. • •To be able to process massive data

Literature:

• •Jiawei Han, Micheline Kamber, Data Mining: Concepts and Techniques, 2nd edition, Morgan Kaufmann, ISBN 1558609016, 2006 • •Pang-Ning Tan, Michael Steinbach, Vipin Kumar, Introduction to Data Mining, Pearson Addison Wesley, ISBN: 0321321367, 2005 • •Ian Witten and Eibe Frank, Data Mining: Practical Machine Learning Tools and Techniques, 2nd Edition, Morgan Kaufmann, ISBN: 0120884070, 2005. • •Keim, D., Kohlhammer, J., Ellis, G., Mansmann, F. Mastering the information age. Solving problems with visual analytics 2010 Eurographics Association

Keywords: Big Data architecture, visualization, spark

INTELLIGENT SYSTEMS Responsible lecturer: Martín Molina ECTs: 4,5 Course type and weekly hours: 2h Exam type: Practical coursework

Contents: In a wide sense, intelligent systems can be considered as a type of computer systems that implement and integrate artificial intelligence methods to acquire and use knowledge for solving problems with limited resources. This course starts presenting machine learning techniques with a general overview followed by a detailed description of machine learning algorithms for symbolic representations (e.g., decision trees and rules). Then, the course explains basic concepts of knowledge representation and reasoning together with specific methods (e.g., representations for ontologies). Next, the course presents language technologies, including solutions for natural language understanding and

natural language generation. Finally, the course presents a unit related to ethics for artificial intelligence. The course combines both a theoretical and a practical presentation and the students have to develop practical exercises related to the main presented concepts and techniques.

Learning Outcomes:

• To know what the main challenges and achievements in the area are of intelligent systems. • To be able to use and apply methods for knowledge acquisition to create manually and automatically knowledge bases using other sources of information (e.g., data sets or text documents). • To be able to identify areas of application where the techniques of intelligent systems can be used. • To know the existing techniques about intelligent systems (knowledge acquisition, knowledge representation and reasoning) understanding their scope and limitations.

Syllabus:

1. 1.Introduction 1.1. Introduction to the course 2. 2.Machine learning 2.1. Overview of machine learning 2.2. Evaluating learned models 2.3. Learning decision trees 2.4. Learning classification rules: Prism, Ripper 2.5. Learning association rules: Apriori 2.6. Learning rules from relational data 2.7. Learning rules with evolutionary algorithms 3. 3.Knowledge representation 3.1. Overview of knowledge representation 3.2. Ontologies 4. 4.Language technologies 4.1. Natural language understanding 4.2. Natural language generation 4.3. Linguistic resources 5. Ethics for artificial intelligence

Keywords: Knowledge discovery, knowledge engineering, ontologies, computational vision

CLOUD COMPUTING AND BIG DATA ECOSYSTEMS Responsible lecturer: Marta Patiño Martínez ECTS: 4.5 Course type and weekly hours: 3 hours Exam type: written test + 2 practical assignment

Contents: This course presents architecture for scalable distributed systems and data management systems: map-reduce, big table, data streaming, persistent queues.

Syllabus

1. Introduction 2. Big Table 3. Dynamo 4. Data Streaming 5. Persistent Queues 6. Containers. AWS

Learning outcomes and competencies:

• To know the applications and systems based on distributed computing. • To be able to process massive data. • To design and implement highly parallel and / or distributed systems. • To know and design information extraction systems

Literature:

•NoSQL Distilled: A Brief Guide to the Emerging World of Polyglot Persistence. P.Sadalage, M. Fowler. 2012 •Big Data Now: Current Perspectives from O'Reilly Radar. O’Reilly. 2011 •Graph Databases. I. Robinson, J. Webber, E. Eifrem. O’Really. 2013

Keywords: Big Data, Internet of Things (IoT), non-structured data, data management technologies, data streaming, big table, big data & cloud ecosystems management, performance evaluation

OPEN DATA AND KNOWLEDGE GRAPHS Responsible lecturer: Oscar Corcho ECTs: 4,5 Course type and weekly hours: Exam type:

Contents: During this course you will learn about the main foundations for the Web of Linked Data and the Semantic Web, including the W3C recommendations that are applicable in this area (RDF, RDF Schema, SPARQL, OWL, CSV on the Web) and methodologies for the generation and publication of Linked Data from multiple heterogeneous data sources and formats. You will also learn about how to create shared agreed vocabularies and ontologies that can give support to such Linked Data generation and publication and will understand how these principles and approaches have been applied to several domains. You will work on a practical hands-on exercise that will be the basis for your participation in hackathons and other similar events.

Learning Outcomes:

• Develop ontologies that serve as vocabularies for the data available on the Semantic Web and for the Linked Data • Manage bibliographic sources in the domain, including manuals, online documentation and scientific papers. • Identify and solve those types of real-world problems in which Linked Data and Semantic Web technologies can be successfully applied. • Use different languages, techniques, methods, and methodologies that enable the development of ontologies and data for the Semantic Web. • Generate data in the format used in the Semantic Web and in the Web of Linked Data, and to publish them for the use of third parties. • Develop applications that exploit the data available on the Semantic Web and on the Web of Linked Data

Syllabus:

1. Introduction to the Web of Linked (Open) Data 2. Data and knowledge representation and access in the Web of Linked Data 2.1. RDF and RDF Schema 2.2. SPARQL 2.3. OWL 3. Linked Data generation, linking and publication from heterogeneous data sources. 3.1. Methodological guidelines for Linked Data generation 3.2. RDF generation from relational databases 3.3. RDF generation from semi-structured data sources 3.4. RDF generation for statistical data 3.5. Data linking 3.6. Linked Data publication 4. Vocabulary selection and development for the Web of Linked Data 4.1. Methodologies for building vocabularies 4.2. Existing vocabularies 5. Linked Data applications 5.1. Linked/Open/Big Data in Government 5.2. Linked/Open/Big Data in Science 5.3. Linked/Open/Big Data in Journalism

Keywords: Semantic Web, Linked Data, Ontologies, applications

I&E UPM Introduction to innovation and entrepreneurship management Responsible lecturer: Alberto Tejero ([email protected]) ECTS: 6 Course type and weekly hours: 2-3 hours Exam type: activities during lectures and presentation of group ́s work

Contents:

Idea generation, technology-based entrepreneurship, marketing and markets, organization and project management, new product and process development, entrepreneurial finance, human resource development.

Syllabus:

• Innovation processes □ From the idea to the market: a long and risky way towards innovation □ Relationship of innovation to research and development: ▪ Integrated view within the knowledge triangle paradigm. □ Nature of knowledge and value of R&D and innovation ▪ Agents, Process, Results ▪ Strategic Planning of R&D and innovation ▪ Evaluation of innovation projects ▪ Implicit project management issues • Organizational structures to support innovation. □ Organization of R&D and innovation ▪ Approaches for private and public entities. □ Organizational models to accommodate innovation processes. □ Large, SMEs and spin-off cases □ Dynamic evolution and growth of start-ups • Innovation models □ Types of innovation ▪ Technology, organizational, commercial innovation ▪ Product, process, service innovation ▪ Evolutionary, disruptive innovation ▪ Open, closed innovation □ Technology maturity levels (TRL) ▪ Innovation dependence on maturity level □ Open innovation approaches ▪ Rationale ▪ Open innovation platforms, services, and products ▪ Open software □ IC open innovation model • Protection of technology □ Protection schemes ▪ Patents vs Industrial secreta ▪ Other schemes (e.g., semiconductor layout, biotech) ▪ The case of software patents (legislation approaches) □ Filing processes ▪ Patent offices (e.g., EPO) ▪ The UPM regulation • Management of innovation projects □ Life cycles models Identification of milestones □ Human resources ▪ Skills and profiles

▪ Management of international teams □ Type of results ▪ Prototypes ▪ Proof of concept ▪ Pilots and demonstrators • Understanding market environment □ Industrial sector analysis □ SWOT • Financial support for innovation □ How much and when money is needed? ▪ Rounds (from seed capital to expansion) □ Sources of funding ▪ F&F ▪ Risk capital ▪ Public funds □ EU public policies for innovation ▪ Innovation in H2020 ▪ Innovation in the regional policy • Substantial part of the contents will be based on the interest of students based on case studies, examples, and geographical/sectorial specificities of this DS master course. □ Set of case studies (both successful and not) to discuss them with students. Case studies should cover a number of approaches, countries, etc. ▪ Discussions with ICT entrepreneurs. ▪ Discussions with IT companies with intrapreneurship programs. □ Visit to technology-based incubators: UPM (actúaupm in the CAIT), Telefónica (Wayra) or the Business Incubator Centre of the ESA (BIC-ESA) in Madrid.

Learning outcomes and competencies:

• To know the main concepts, terminology and main issues related to entrepreneurship and innovation management with focus in the IT sector. • The capacity to identify and link the key issues related to project innovation management and, specifically on the data science field in open, international and cooperative innovation contexts. • The ability to propose the right management structure and activities of an innovation project from its conception to the deployment to the outcomes to the market by using a specific management model adapted to the type of project. • The ability to select the best approach to protect his/her technology depending on the type, maturity level and geographical constraints (through patents, industrial secret, etc.) and to understand their consequences in accessing or commercializing it. • The capacity to understand the basis for entrepreneurship and the rationale for launching a technology-based company creation from previous R&D activities. • The capacity to identify different sources for innovation funding and to select the most appropriate one according to the business model and involved technology. • The knowledge of main European Union (EU) policies and programs to support research and innovation. The role played by the EIT in the EU landscape.

Literature:

• •Henry Chesbrough. Open Innovation: The New Imperative for Creating and Profiting from Technology (HBS Press, 2003). • •Henry Chesbrough. Open Services Innovation. Rethinking your business to grow and compete in a new era. Ed. Jossey-Bass. 2011. ISBN 978-0-470-90574-6 • •Hippel, Eric von (2013): Open User Innovation. In: Soegaard, Mads and Dam, Rikke Friis (eds.). "The Encyclopedia of Human-Computer Interaction, 2nd Ed.". Aarhus, Denmark: The Interaction Design Foundation. Available online at http://www.interaction- design.org/encyclopedia/open_user_innovation.html • •Osterwalder, A. and Pigneur, Y.: Business model generation. John Wiley & Sons 2010. • •The Innovative and Entrepreneurial University: Higher Education, Innovation & Entrepreneurship in Focus. US Dept. of Commerce. Oct. 2013 • •J.P. Murmann. The co-development of industrial sectors and academic disciplines. Science and public policy. Vol. 40 No.2 Apr. 2013. • •Hugo Hollanders and Nordine Es-Sadki.Innovation Union Scoreboard. Maastricht Economic and Social Research Institute on Innovation and Technology (UNU-MERIT 2017)

Keywords: Innovation, management, organizational structures, market environments, financial support, entrepreneurs, business

1st year, 2nd semester (spring/summer semester)

Common Core UPM DEEP LEARNING Responsible lecturer: Prof. Martin Molina Gonzalez ECTS: 3 Course type and weekly hours: lecture, assessment activity Exam type: written exam + practical project

Contents: Deep learning has emerged from the connectionist branch of machine learning, aided by the arrival of big data and increased computational power (e. g., parallelization using graphics processing units - GPUs). Deep learning has proved to be significantly better than other approaches to solve problems that cope with large amounts of data as it is required, for example, in computer vision (image or video processing) or speech understanding. This course presents a theoretical and practical view of deep learning. The course presents first the foundations of artificial neural networks with both supervised and unsupervised learning. Then, the course presents different types of deep architectures (e.g., convolutional neural networks) and application domains (e.g., computer vision and natural language processing). To complement the practical view, the course also presents specialized software tools for deep learning and describes how to use them in practical problems.

Syllabus:

1. Introduction to deep learning 2. Artificial neural networks 2.1. Foundations 2.2. Learning in artificial neural networks 2.3. Tools 3. Deep learning for computer vision 3.1. Foundations of computer vision 3.2. Convolutional neural networks 3.3. Sample projects 4. Deep learning for natural language processing

Learning outcomes and competencies:

• To know the main challenges and achievements of deep learning • To be able to identify areas of application where the techniques of deep learning can be used. • To know the existing techniques and software tools about deep learning, understanding their scope and limitations • To be able to apply machine learning software tools for practical problems related to deep learning.

Keywords: Neural Networks, computer vision, NLP, applications

INFORMATION RETRIEVAL, EXTRACTION, AND INTEGRATION Responsible lecturer: David Pérez del Rey ECTS: 4,5 Course type and weekly hours: 2 Exam type: practical coursework

Contents: The amount of available data in any area has grown dramatically during the las years. However, this increment did not have a proportional impact in the knowledge available for decision making. There is a need of automatic models to manage the data, taking into account that the majority of the data will never be used by a human being. The course Information Retrieval, Extraction and Integration is focused on the necessary tasks to extract information, models to efficiently retrieve data for further integration. These are critical tasks to provide relevant information for decision making, which complexity increases with the amount of data available. As application areas, we focus on biomedicine, due to the complexity and to the specific requirements.

Syllabus:

1. Basic Concepts 1.1. Introduction 1.2. Data, Information and Knowledge 1.3. Data types 2. Extraction and Information Retrieval 2.1. Information Extraction

2.2. Information Retrieval Models 2.3. Natural Language Processing 2.4. Web Search Engines 2.5. Non-textual Data 3. Data Integration 3.1. Integration Architectures 3.2. Semantic Interoperability 3.3. Data Provenance 4. Applications in Biomedicine 4.1. Biomedical Information Systems 4.2. Clinical interoperability Standards 4.3. Medical terminology 4.4. Scientific literature retrieval system

Learning outcomes and competencies:

• The ability to analyze information needs to build an information system. • To understand database and data representation language foundations • To understand and interact with information retrieval systems. • To understand and interact with data extraction systems. • To understand and interact with integration systems. • To apply information retrieval, data extraction and integration to the biomedical field • To be able to develop simple information retrieval architectures. • To be able to identify applications of interest where the information retrieval methods can be used.

Literature:

1. Baeza-Yates, Ricardo, and Berthier Ribeiro-Neto. Modern information retrieval. New York: ACM press, 1999. 2. Kimball, Ralph, and Margy Ross. The data warehouse toolkit: the complete guide to dimensional modeling. John Wiley & Sons, 2011. 3. Doan, AnHai, Alon Halevy, and Zachary Ives. Principles of data integration. Elsevier, 2012. 4. Manning CD, Raghavan P, Schütze H. Introduction to Information Retrieval. Cambridge University Press. 2008. 5. Witten IH, Moffat A, Bell TC. Managing Gigabytes: Compressing and Indexing Documents and Images, 2nd Edition. Morgan Kaufmann. 1999. 6. Korfhage, R. Information Storage and Retrieval. Wiley. 1997. 7. Bird S, Klein E, Loper E. Natural Language Processing with Python. O'Reilly 2009.

Keywords: Interoperability, Data Integration, Information retrieval, Text Mining, Non-structured data

Electives UPM GRAPH ANALYSIS AND SOCIAL NETWORKS Responsible lecturer: Prof. Javier Bajo Pérez ECTS: 3

Course type and weekly hours: lecture, assessment activity Exam type: written exam + practical project

Contents: Social computing is a general term for an area of computer science that is concerned with the intersection of social behavior and computational systems. During recent years the Internet introduced a social element where users could network, share interests, publish personal insights, and use their computers for more than just doing a job faster, and this has led to the development of social machines where both humans and machines collaborate to solve social problems. This course presents the principal of social computing and focuses on graph and network analysis as well as on the design of social machines.

Syllabus:

1. Introduction to social computing 2. Graph Analysis 3. Network Analysis 4. Intelligent Agents and Multi-agent Systems 5. Design of Social Machines

Learning outcomes and competencies:

• To know the main challenges and achievements of social computing • To be able to identify areas of application where the techniques of social computing can be used. • To know the existing techniques and software tools for graph and network analysis, understanding their scope and limitations • To be able to design social machines for practical problems related to deep learning.

Keywords: social computing, graph analysis, network analysis, multiagent systems

AI AND LEGAL/SOCIETAL/ETHICAL ASPECTS Responsible lecturer: Víctor Rodríguez Doncel ECTC: 3 Course type and weekly hours: Exam type: Keywords: Data protection, intellectual property, SW licensing, IA, and data processing ethics

Contents: Virtually every data science and AI professional will have to cope with legal and ethical issues during his or her professional career. This course provides the student with practical and theoretical tools to address these issues. In particular, the student will be given some general notions on the legal framework in Europe of AI and medical data processing, necessary to avoid breaching the law and necessary to exercise their rights. Specific focus will be made both on ‘privacy and data protection’, including practical cases around the EU GDPR (General Data Protection Regulation) and Electronic Health Records, and on intellectual property and copyright regulation, including practical examples on data licenses and open data policies.

Syllabus:

1. Introduction. Overview of issues raised by medical data analytics. a. Case studies 2. European legal framework medical data analytics a. Privacy and Data Protection: general ideas b. Privacy and Data Protection: the GDPR applied to HER. c. Intellectual property: general ideas d. Intellectual property: data licensing, open data e. Medical 3. Ethics of medical data analytics a. Autonomy, System Design, Agency, and Liability. Ownership, control, access. b. Algorithm bias. Governance, Explain ability, and Accountability. c. Ethical guidelines in EU research programs

Learning outcomes:

• Ability to produce documents assessing the impact of medical data analytics, with consideration of legal, regulatory, privacy, ethics, and human behavior topics. • Identify recurrently appearing legal issues in the exercise of the medical data analytics profession. • Enumerate the ethical issues considered by the European Commission in their research programmes and Demonstrate familiarity with relevant examples of medical data analytics systems arising ethical problems. • Demonstrate knowledge of philosophical issues involved in ethics of AI in practical contexts.

Keywords: Privacy, Data Protection, GDPR, Ethics

I&E UPM ENTREPRENEURSHIP & BUSINESS MODELLING Responsible lecturer: Arístides Senra Díaz ([email protected]) ECTs: 6 Course type and weekly hours: 4 Exam type: continuous evaluation, presentation of a business project and development of a business plan

Contents: Business modelling and development in phases – (a) idea recognition – (b) concept design – (c) Business modelling and planning – (d) Business plan presentation.

Application of subjects from the I&E Basics course or introduced in the Bus Dev Lab:

• Opportunity recognition • Creativity techniques, Design thinking • User-centred product/service design

• Product development, project management • Business Model Canvas (9 boxes), value creation, value proposition, revenue models • Alternate business modelling methods • Methods and tools for customer discovery, customer validation, evidence-based decision making, lean process. • Market analysis, competitors’ analysis • Business ethics, sustainability • Finance (cash flow management, financial scenarios) • Other business development planning concepts methods and tools (strategy, organization, marketing, market entry / go-to-market, risk analysis) • Financing, fund raising. • IP and intellectual assets, IP strategies, patent management • Technology maturity, TRL, prototyping, Technology transfer • Pitching and oral communication

Learning Outcomes:

• The ability to identify innovative business ideas inside (intrapreneurship) or outside a pre- existent firm (entrepreneurship). • The ability to evaluate from professional point of view business opportunities with high growth potential. • The ability to develop business projects with customer orientation. • The ability to explain his/her technology-based business ideas to external investors. • The ability to write non-technical information about the business idea for investors or customers. • The ability to develop a complete business model by using pre-defined methodologies. • The ability to understand the necessary steps to create a sustainable technology-based company. • The ability to get external funds to finance the development. • The capacity to present media material on Internet about their idea.

Syllabus:

1. Introduction to entrepreneurship fundamentals a. Attitude required and analysis methodology from risk investors. b. Four circles analysis methodology c. Idea analysis d. Team analysis e. Resources from founders f. First approach to business model 2. Eleven steps to analyse a business model proposal a. Customer analysis b. Elevator pitch template c. Competitors analysis d. Competitive advantage analysis e. Team matrix analysis 3. Fast analysis methodology a. Environment analysis

b. Sector analysis. Porter methodology for entrepreneurs

c. Life cycle d. Value chain analysis e. DAFO matrix analysis and conclusions for start-ups and new projects launching. 4. Marketing and operations a. Positioning b. MVP and Marketing MIX c. MIS and monitoring d. Sales projection methodology/ Sales forecast 5. Financial fundamentals a. Investment plan/ resources needed. b. Financing expectations c. Negotiation with investors d. Profit and loss e. Treasury f. Balance g. Profitability 6. Writing the business plan a. Business model generation b. Business model canvas generation c. Index d. Executive summary e. Presentation of the business plan 7. Demo Day a. How to prepare a presentation for investors b. Demo day practice

Keywords: Entrepreneurship, Business planning, Financing

INTRODUCTION TO TECHNOLOGY WATCH AND COMPETITIVE INTELLIGENCE Responsible lecturer: Javier Segovia ECTS: 1 Course type and weekly hours: Seminar Exam type: individual group work

Contents: This Seminar constitutes the second part of the Basics course. It will be offered during the month of January 2019 by using a blended learning approach with the use of some on-line modules prepared by the UPM.

After presenting the basic elements of innovation management, students will receive detailed information on tools and procedures related to the identification, selection and eventually absorption/adaptation of technologies, which could be useful for the selection of the technologies required to implement their own business projects in the master’s degree.

Syllabus:

• Technology evolution □ Technology maturation □ Technology roadmaps □ Technology forecasting □ Introduction to quantitative approaches in forecasting: econometrics, exponential- smoothing techniques, s-curves, other. • Technology watch □ Processes used. □ Internal and external □ Scouting networks □ Tools for technology watch • Technology intelligence □ Use in decision making. □ Trend-charts □ Connection to the maintenance of IP portfolio □ Road mapping of products/services • Technology transfer □ Technology absorption □ Technology transition • Work on a case study (group activity) □ Big data in some sectors (e.g., health) • Visual analytics

Learning outcomes and competencies:

• To provide students with some conceptual and practical tools to understand the evolution of technologies for specific purposes. • To know how the develop and interpret a technology roadmap in specific technical areas. • To understand the relationship of technology intelligence to decision making in innovation management. • The knowledge of the rationale and basic concepts related to technology watch and competitive intelligence. • The ability to identify the way that information and communications technologies evolve over time. • The ability to identify the maturity level of a technology. • The ability to develop and interpret a technology roadmap. • The ability to use some techniques applied for technology watch and forecasting. • The ability to understand the relationship of technology watch and competitive intelligence to support decision making in innovation management. • The ability to identify and to describe the specificities of the market and its technology dependencies. • The capacity to incorporate technology watch units in a given organization. • The ability to understand common barriers for technology transfer and absorption in the field of digital services.

Literature:

• René Rohrbeck: Harnessing a Network of Experts for Competitive Advantage: Technology Scouting in the ICT Industry. R&D Management, Vol. 40, No. 2 pp. 169-180 http://www3.interscience.wiley.com/journal/123275929/abstract • Ramona-Mihaela MATEI, Ioan RADU Conceptual Relationship between Information and Communication Technologies and Competitive Intelligence Activities • Gestión de la I+D+i: Sistema de vigilancia tecnológica e inteligencia competitiva. UNE 166006:2011 • HAMDI Hassen - RAMRAJSINGH Athissingh: Veille et intelligence économique au sein des TPE :vers l’appropriation des outils gratuits • Big Data: Big today, normal tomorrow ITU-T Technology Watch Report. November 2013 http://www.itu.int/dms_pub/itu-t/oth/23/01/T23010000220001PDFE.pdf

Keywords: technology watch, maturation, roadmaps, technology transfer, visual analytics, big data in health, scouting networks

LAUNCHING OF DIGITAL-BASED PRODUCTS AND SERVICES Responsible lecturer: Francisco Jariego ([email protected]) ECTS: 2 Course type and weekly hours: Seminar, 4 hours week Exam type: activities during lectures + individual work

Contents: Business modelling and development in phases – (a) idea recognition – (b) concept design – (c) Business modelling and planning – (d) Business plan presentation.

Application of subjects from the I&E Basics course or introduced in the Bus Dev Lab:

• Opportunity recognition • Creativity techniques, Design thinking • User-centred product/service design • Product development, project management • Business Model Canvas (9 boxes), value creation, value proposition, revenue models • Alternate business modelling methods • Methods and tools for customer discovery, customer validation, evidence-based decision making, lean process • Market analysis, competitors’ analysis • Business ethics, sustainability • Finance (cash flow management, financial scenarios) • Other business development planning concepts methods and tools (strategy, organization, marketing, market entry / go-to-market, risk analysis) • Financing, fund raising • IP and intellectual assets, IP strategies, patent management • Technology maturity, TRL, prototyping, Technology transfer

• Pitching and oral communication

Syllabus:

• Understanding the sector: towards digital economy o Global market o Global organisations o Mergers and acquisitions: dynamic restructuring • Structure of the ICT sector o Main stakeholders ▪ Incumbents and new entrants ▪ Actors in value chains o Regulation (in the EU) for new products ▪ EC competences (EU Treaties) ▪ Industrial policies ▪ Global markets ▪ Indicators from the EU Digital Agenda • Introduction to the specificities of marketed digital products and services o Digital services evolution (versioning, service life cycle). o Technology integration and evolution ▪ HW/SW embeddedness, open platforms, etc. o Services components • Digital marketing o Channels o Market surveys o The new role of users • What happens when the product/service is already in the market? o Lean approach o Metrics and dashboard o Different routes to success • Personal assignments to students E.g., practical case on launching an application based on data analysis.

Learning outcomes and competencies:

• The ability to define a marketing plan and its international expansion • The ability to negotiate with other actors in the ICT field the participation in the value chain • The ability to create a commercial structure • The criteria to define a pricing strategy • The knowledge to obtain information on initial users

Keywords: Digital marketing, digital economy, policies

I&E SEMINARS Responsible lecturer: Javier Segovia

ECTS: 5 Course type and weekly hours: 3 hours

Contents: An elective I&E course will be offered covering advanced topics on any of the following: business development, business finance, marketing, innovation management, intellectual property, and market research.

Literature:

• Thomke, Stefan, and Jim Manzi. "The Discipline of Business Experimentation." Harvard Business Review 92, no. 12 (December 2014): 70–79. • Chapter 7: Testing and Experimenting in Markets in H. M. Neck, C. P. Neck, and E. L Murray, Entrepreneurship: The Practice & Mindset (2017), Sage Publications: Thousand Oaks, CA.

Keywords: Business Finance, Marketing, Business Finance, Innovation management, IP

2nd year, 1st semester (autumn/winter semester):

I&E I&E STUDY Responsible lecturer: Raúl Gutiérrez Sanchis ECTS: 6 Course type and weekly hours: 3 hours Exam type: 2 practical assignments (individual and group)

Contents: Two topics – with related concepts, methods and/or tools – will be covered in the context of a selected innovation or entrepreneurial case:

− One fixed and common topic: Assessing the impact of a technology on an industry, market and/or organization, the support and barriers to its deployment, the influence on a specific goal/agenda (technology transfer, existing industry, new company, etc.). − One case-dependent topic: pertaining to market / business environment analysis (main forces affecting the business, suppliers, partners, competition, environmental issues), sustainability and social issues, business modeling, go-to-market strategies, etc.

The innovation or entrepreneurial project may be originating from:

− Cases issued from EIT Digital Innovation Action Lines: within Activities, Partners / Business Community projects − Cases based on the continuation of students EIT Digital Summer School (or BDLab) project − Cases within other innovation or entrepreneurial projects rooted in a real-life environment as may be collected in the university ecosystem.

Learning outcomes and competencies:

• the ability to use knowledge, ideas, and technology to create new or significantly improved products, services, processes, policies, new business models or jobs (Innovation skills and competencies) • the ability of decision-making and leadership, based on a holistic understanding of the contributions of Higher Education, research and business to value creation, in limited sized teams and contexts (Leadership skills and competencies)

Literature:

• Thomke, Stefan, and Jim Manzi. "The Discipline of Business Experimentation." Harvard Business Review 92, no. 12 (December 2014): 70–79. • Chapter 7: Testing and Experimenting in Markets in H. M. Neck, C. P. Neck, and E. L Murray, Entrepreneurship: The Practice & Mindset (2017), Sage Publications: Thousand Oaks, CA.

Keywords: Technology impact, barriers, market analysis, business modelling

I&E SUMMER SCHOOL Responsible lecturer: Various ECTS: 5 Course type and weekly hours: workshop/laboratory course (150 hours in total, including self- study and preparation of tasks in teams) Exam type: /

Contents: The I&E practical training includes the preparation, execution and documentation and pitch of a practical project, which is based on a real use case in the healthcare context. The practical work is carried out in teams a laboratory setting. The infrastructure of the lab gives students the chance to execute their own product ideas or to work on topics from industrial partners.

Learning outcomes and competencies: Students are able to:

• apply industry standard techniques and team management • develop, independently and within a team, their problem-solving and creative skills • implement their ideas as prototypes by applying agile software development methods • use the results of their projects for the creation of start-ups

Health and Medical Data Analytics specialization (20 ECTS) Specialization offered by UPM: Analysis of Clinical Data

DATA MANAGEMENT AND KNOWLEDGE IN HEALTH Responsible lecturer: Victor Maojo ECTC: 4 Course type and weekly hours: lecture (2 hours) Exam type: 2 individual work group

Contents: This course will address the foundational topics of the area of data and knowledge management for health. The topics will include: (1) Introduction. (2) Data, Information & Knowledge in Biomedicine. (3) Research design for health data management. (4) Decision Making in biomedicine. (5) Biomedical terminologies and vocabularies. (6) Electronic Health Records (EHRs) and Hospital Information Systems (HIS). Surveying Information systems that can be found in Health environment and studying most common technologies and standards used in the area. (7) Integration and interoperability for health data and knowledge sources. Classical and semantic methods to integrate heterogeneous clinical information stored in different formats. (8) Bioinformatics applications in biomedicine. Techniques and applications for data sequence processing and analysis. Sequence alignment. At the intersection between biology and medicine, we will focus on new areas such as precision and personalized medicine, including new topics and applications. There will be an emphasis on basic and research concepts, with examples from recent international projects in the area.

Syllabus:

• Clinical interoperability and integration • Hospital information systems and electronic health records • Artificial Intelligence in biomedicine • Bioinformatics • A Brief Introduction to Biomedical Text Mining

Learning outcomes and competencies:

• Ability to assess the importance of documentary sources, manage them and find information for the development of any research work. • To be able to use the terminology appropriately and perform public presentations on the topics of the module. • To be able to analyse the state of the art in a given subject, understanding what the main achievements and challenges are, and draw conclusions for one’s own work. • Once the aforementioned points have been understood, students should be able to successfully apply them to the analysis and solution of problems with a complexity proportional to their level of experience.

Literature:

1. Shortliffe, E. H., & Cimino, J. J. (Eds.). (2013). Biomedical informatics: computer applications in health care and biomedicine. Springer Science & Business Media. 2. Benson, T., & Grieve, G. (2016). Principles of Health Interoperability. Springer.

Keywords: Hospital Information Systems, Clinical Terminologies, Standards, Bioinformatics, Interoperability

E-HEALTH: PROMOTING HEALTHY AGING Responsible lecturer: Elena Villalba ECTC: 4,5 Course type and weekly hours: lecture (3 hours) Exam type: individual assignments and teamwork.

Contents: This course focuses on understanding the necessary models, techniques and architectures that allow the development of interactive systems in the E-health domain. This course covers eHealth, e Inclusion, co- production of health, empowerment, social innovation, social networks, serious games, and participation in society. A final team project closes the course.

Syllabus:

1. Active and Health Ageing 1.1. Course introduction 1.2. Definition and frameworks 2. Clinical perspective 3. Political perspective 4. Social and personal perspective 5. Intrinsic capacity and frailty 6. Cognitive Decline and Mild Cognitive Impairment 7. Mobile Health 8. Active and Healthy Ageing Project

Competences:

• Ability to make connections between the wishes and needs of the consumer or client and what technology can offer • Ability to analyse the information’s needs that arise in an environment and carry out the user- centred design process in all its stages

Learning outcomes:

• To evaluate the usability and accessibility of prototypes • To apply techniques for modelling the context of use • To understand techniques, technologies and processes that allow to prototype, to develop and to improve digital interactive systems based on various user interface technology platforms • To understand the needs of specific contexts involving ageing population and its derived disabilities

Literature:

• Bousquet, Jean, et al. "Operational definition of Active and Healthy Ageing (AHA): A conceptual framework." The journal of nutrition, health & aging 19.9 (2015): 955-960.

• Beard et al. (2016). The World Report on ageing and health: a policy framework for healthy ageing. Lancet 2016; 387: 2145-54. • OMS. Global age-friendly cities: a guide (2017). Disponible en: http://www.who.int/ageing/publications/age_friendly_cities_guide/en/ • Mapping mHealth research: a decade of evolution. Fiordelli, Maddalena, Nicola Diviani, and Peter J. Schulz. Journal of medical Internet research 15.5 (2013). • From Personal to Mobile Healthcare: Challenges and Opportunities Villalba-Mora, Elena, Ignacio Peinado, and Leocadio Rodriguez-Mañas. (2016). Emerging Perspectives on the Mobile Content Evolution. IGI Global, 2016. 124-137.

Keywords: Healthcare data and information systems, Internet of Things, Medical Sensors, Mobile Health, Devices, Ethical and Legal Issues, HIS, elderly, frailty, unhealth

MEDICAL TIME SERIES DATA MINING Responsible lecturer: Juan Pedro Caraça-Valente ECTS: 3 Course type and weekly hours: lecture (2 hours) Exam type: Project and 2-3 small assignments during the course

Contents: In this course we will focus on Data Mining techniques suited for Time Series data. Time Series are present is almost every area of science and industry and has a great presence in Medicine. Many medical tests provide time series, like electrocardiograms, electroencephalograms, auditory brainstem responses, isokinetic curves, etc. Data Mining techniques can help the physicians analyze these results, as in many other fields. But Time Series data are a great challenge for traditional data Mining Techniques as attributes are no longer single valued. In this subject we will address the specific problems of Data Mining in Time Series, how some classic techniques have to be adapted for time series data, specific techniques that arise, etc.

Syllabus

1. Introduction to Time Series 1.1. Problems of Data Mining in Time Series 1.2. Medical Time Series 2. Basic Techniques 2.1. Fourier Transform 2.2. Euclidean Distance 2.3. Segmentation 3. Time Series Distances 4. Dimensionality Reduction on Time Series 5. Time Series Data Mining Techniques 5.1. Comparing Time Series 5.2. Searching for subseries 5.3. Pattern Identification 5.4. Event Detection 5.5. Temporal Abstraction

6. Evaluation of Data Mining projects in Medicine 7. Applications to Medicine Learning outcomes and competencies:

• To know how to deal with Time Series data in general and in Medical domains in particular • To be able to deal with the specific problems of data Mining in Times Series and be able to propose a plan to overcome then. • To know and be able to apply Data Mining Techniques in Time Series data

Keywords: Data Mining, Healthcare data, Time series comparison and search, pattern identification, event detection, Temporal abstraction

COMPLEX DATA IN HEALTH Responsible lecturer: Alejandro Rodríguez ECTC: 4,5 Course type and weekly hours: lecture (2 hours) Exam type:

Contents:

1. Complex networks 1.1. Basics of complex networks. Cytoscape for plotting. 1.2. Physical networks. For instance, connectome, calculate basic topological features, and network randomisation. 1.3. Functional networks. Reconstructing brain networks with correlation. Difference with causality. 1.4. Data mining and networks. Optimisation of networks. Using networks in classification tasks. 1.5. Other topics. MST. Link filtering and prediction. Multi-layer and time evolving. 2. Disease Networks 2.1. Human disease networks 2.2. Human symptom disease networks 2.3. Disease understanding 2.4. Approaches and utilities of disease networks 3. Managing complex data 3.1. Textual health information and its mining 3.2. Biological databases

Syllabus:

• Complex networks and network-based analyses o Reconstruction of physical vs. functional networks o Analysis of networks and evaluation of topological metrics o Integrating networks in a data mining process o Links prediction and filtering • Human disease networks o Understanding of basic disease components and features

o Understanding of disease similarities and metrics o New paradigms for disease understanding o Utility of disease networks • Managing complex data o Retrieving information from textual sources o Dealing with complex biomedical data: sources and databases o Managing complex biomedical information

Learning outcomes and competencies:

• Understanding the basis of complex networks and its use in health • Understanding human disease networks: types, uses, challenges • Dealing with complex biomedical data, including unstructured sources

Keywords: Complex Networks, Human Disease Network, Human Symptoms Disease Network, Complex databases, Biomedical sources, Disease understanding, Textual information, Data structuration

DEVICES AND BIOMETRIC APPLICATIONS FOR E-HEALTH Responsible lecturer: Agustín Álvarez ECTs: 3 Course type and weekly hours: lecture (2 hours) Exam type: Practical assignment

Contents: This course will introduce students into the basics of e-health applications from the point of view of device built-in sensors and focused on biometric data analysis useful for patient monitoring. Basic procedures for sensor managing for synchronous and/or asynchronous mode of operation in mobile, wearable and IoT devices will be presented. Finally, application development for different scenarios in the health domain will be reviewed.

Syllabus:

1. Introduction: from devices to medical related applications. 2. Biometric signals and e-Health. 2.1. Signals form common devices (e.g., mobile phones, smart). 2.1.1. Voice/speech. 2.1.2. Gyroscope/Accelerometer data analysis 2.2. Medical oriented devices. 2.2.1. Glucose scan. 2.2.2. ABG reading. 2.2.3. Holter monitor. 3. IoT & edge computing for e-Health. 3.1. Edge computing resources. 3.2. Cloud platforms for IoT. 4. Mobile application development.

4.1. Review of main applications. 4.2. APIs for sensor & dedicated devices. 4.3. Sensor data acquisition. 4.4. Data filtering and preparation. 4.5. User interfaces.

Learning outcomes and competencies:

• To know the applications and systems based on biometric data. • To be able to deal with unstructured sources as device raw data. • To know the fundamentals of data extraction and analysis in mobile and wearable computing devices. • To discover potential new health monitoring applications dealing with IoT and edge computing systems.

Literature

• Mobile Health: Sensors, Analytic Methods, and Applications, James M. Rehg (Editor), Susan A. Murphy (Editor), Springer, 2017, ISBN: 9783319513935 • Handbook of Multisensory Data Fusion: Theory and Practice (2nd Edition), Martin Liggins II, David Hall, James Llinas, CRC Press, 2008, ISBN 9781420053081. • Data Science for Healthcare. Methodologies and Applications, Consoli, Sergio, Reforgiato Recupero, Diego, Petkovic, Milan (Eds.), Springer, 2019, ISBN: 9783030052485.

Keywords: sensors, IoT, mobile & wearable devices, e-Health applications, patient monitoring

2nd year, 2nd semester (spring/summer semester): MASTER THESIS

Supervisor: At least 1 academic supervisor ECTC: 30 Exam type: written thesis and oral presentation

It is recommended to start looking for a final thesis subject in the second year of study. This way, students have time to take matching study modules to acquire specialized knowledge. Registration is possible for all students, that have completed all mandatory modules and have acquired at least 75 ECTS credits. From there on, the thesis is limited to 6 months, which correspond to 30 ECTS or 900 hours of work.

Study track at the UGA

1st year, 1st semester (autumn/winter semester)

Innovation and Entrepreneurship 1 (9 ECTS) INITIATION TO PROJECT MANAGEMENT Responsible Lecturer: Jean Breton ECTS: 3 Course type and weekly hours: 16h Exam type: scientific report, presentation

Content: This course provides a sketch of the traditional method of project management. The model that is discussed here forms the basis for all methods of project management. Later courses go into more depth regarding a model that is particularly appropriate for IT-related projects. Dividing a project into phases makes it possible to lead it in the best possible direction. Through this organisation into phases, the total workload of a project is divided into smaller components, thus making it easier to monitor. The following paragraphs describe a phasing model that has been useful in practice. It includes six phases:

1. Initiation phase 2. Definition phase 3. Design phase 4. Development phase 5. Implementation phase 6. Follow-up phase

Prerequisites: The plans must not necessarily have a technological characterization. They must economically be viable, but not compulsorily lucrative (partnerships, fair trade, ...).

SCIENTIFIC READING AND WRITING Responsible Lecturer: Don Martin ECTS: 3 Course type and weekly hours: 16h Exam type: scientific report, presentation.

This course will introduce the basis to perform good oral and written presentations with several sequences of combined lecture, tutorial, workshop to allow the master students to deal with any situation of their Master 1 year or their future work.

Content: A first sequence will help the students to find an internship. A tutorial and a workshop will be organized in order to improve their CV and practice an interview. A second teaching sequence will address the way to write a scientific report, from an internship report towards a scientific publication. This part will be followed by a lecture concerning the different modes of scientific communication, written or oral (report, meeting, lab-book, etc.) within a laboratory. A third sequence will deal with oral scientific presentation. A tutorial will explain how to prepare the presentation as well as how to present it. A workshop will give the opportunity to practice in front of an audience. It will also be the basis for an evaluation. Another lecture will provide an opening towards electronic communication tools in order to be able to succeed in a skype interview and to be visible on the web using tools like LinkedIn or Viadeo.

A closing lecture will offer a transversal view of the different points highlighted throughout the course.

Learning outcomes: Students will improve their skills in written and oral presentation. The course will provide competences required for the Master 1. Students will be trained in order to be able to write their CV and internship report, achieve interviews and be able to provide convincing scientific presentation. These specific points will also be extended to other oral and written communication tools as well as electronic communication that can be needed during and after a Master 2 program.

HMDA Common core (21 ECTS) MOLECULAR BASES OF HUMAN DISEASES Responsible lecturer: Arnaud Seigneurin ECTS: 3 Course type: 22h Exam type: t.b.d.

Knowledge of the language of human pathology to facilitate dialogue between scientists and medical practitioners. Better integrate in the medical field the scientific and technical knowledge acquired in the IS Master. To show how, from the knowledge of molecular and cellular mechanisms, biotechnologies make it possible to improve the treatment or diagnosis of diseases. Course examples:

• Pathology Basics • Mechanisms of HIV and EBV infections • Molecular and cellular bases of carcinogenesis • Diabetes and its complications • Atheroma and cardiovascular diseases • Place of inflammation in human pathologies • Cell bases of autoimmune diseases • Mechanisms of host-pathogen relations: malaria

MOLECULAR TOOLS TO HEALTH Responsible lecturer: Arnaud Seigneurin

ECTS: 6 Course type: 36h

Presentation of the methods of preparation and transformation of macromolecules (nucleic acids, polysaccharides, and analogues): in vitro and in vivo synthesis of DNA, RNA, modified DNA, principles of extraction, purification, manipulation of macromolecules by enzymes, PCR and RTPCR in real time, analytical methods, and strategy in molecular biology. Knowledge of vectors: main elements constituting a vector, fast cloning methods, inducible systems, promoter activity control, genomic insertion method, mass spectrometry analysis.

The goals are to acquire:

• a knowledge of the main techniques in molecular engineering • a knowledge of the chemical and biochemical tools used in biotechnologies using biological polymers (DNA, polysaccharides, etc.).

Examples of interventions:

• Synthesis of normal and modified nucleic acids. • Chiral selectors applied to nucleic acids and biological polymers. • NAPs and their use • Polysaccharides: Preparations, Biotechnological Changes and Sugar • Protein Interactions. • PCR and RTPCR in real time • Vectors • DNA analysis: mass spectrometry, sequencing methods, capillary electrophoresis, dHPLC • Sequencing and sequencing methods for eukaryotic genomes • Méthodes de séquençage et séquençage des génomes eucaryote

PROTEOMICS FOR HEALTH RESEARCH Responsible lecturer: Sandrine Bourgoin ECTS: 3 Course type: IS E-learning (45h) Exam type: t.b.d.

Proteomics describes large-scale analysis of proteins in a biological sample. The aim of these studies is to determine the protein parts that are present in such samples and to define their concentrations, molecular states, structures, functions, or connections. Today, there are different technologies being used and developed to study the different types of samples such as to find biomarker molecules that could help to diagnose diseases or even improve therapy of patients.

Course main content: The objective of the course is to present current trends for global protein analysis and to demonstrate its principles, challenges, and complexity. The course will therefore provide an overview of typical proteomics applications used today, such as for biomarker discovery and validation. The course is focused on different methods, technologies and strategies currently used within the field of proteomics in general and with an emphasis on biomarker discovery. The lectures will cover background and recent advances for both classical proteomics methods, such as 2D-gel electrophoresis and mass spectrometry,

and strategies based on high-throughput antibody generation, bioinformatics, and structural approaches.

Intended learning outcomes: The aim of the course is to provide the students with an introduction to current methodologies and trends in the field of proteomics. The students should also obtain an overview and awareness of typical proteomics applications. After completed course the student should be able to describe and discuss the possibilities and advantages, and the complexity and drawbacks of various proteomics technologies compare traditional methods with emerging technologies suggest suitable approaches for specified applications and motivate the choice speculate and argue about the future of proteomics technologies participate in scientific discussions regarding proteomics technologies critically evaluate scientific results.

Literature: Principles of Proteomics by R.M Twyman, Garland Science, ISBN: 9780815344728 (second edition)Handout and articles distributed at the lectures

APPLIED PROBABILITY AND STATISTICS Responsible lecturer: Anatoli Iouditski, Sana Louhichi ECTS: 6 Course type and working hours: 48h Exam type: Applied probe part (50%), statistics part (50%)

Description: The aim of this course is to provide basic knowledge of applied probability and an introduction to mathematical statistics. Contents of applied probability: Markov Chains: description, properties, applications. Contents of mathematical statistics:

• Estimation (parameter) • Sample comparison • Statistical tests

This course includes practical sessions. Mutualized with M1 SSD Applied probability and Statistics. See the associated content from UPMF.

ALGORITHMS AND SOFTWARE TOOLS Responsible lecturer: Laurence Pierre ECTS: 3 Course type and working hours: 36h Exam type: practical (50%), written exam (50%)

Description: The objective of this course is to present the computer sciences basics useful for applied mathematics.

Contents:

1. Compilation (const, inline, loops, Gnu Make ...) 2. C++: genericity (template), code reuse (STL), efficient programming 3. Objects and hierarchical memory, notions of cache and locality (e.g., BLAS) 4. Basics of algorithmics 5. Complexity 6. Error propagation, floating point computing

Prerequisite: UNIX/LINUX. At least some programming languages (C, python, java).

Learning outcome: Produce code using C++, algorithms, and compilation tools, taking into account complexity and errors.

SIGNAL AND IMAGE PROCESSING Responsible lecturer: Cecile Amblard ECTS: 6 Course type and working hours: 54h Exam type: practical (50%), written exam (50%)

Description: The aim of this course is to provide the basics mathematical tools and methods of image processing and applications. Content:

• Image definition • Fourier transform, FFT, applications. • Image digitalisation, sampling • Image processing: convolution, filtering. Applications • Image decomposition, multiresolution. Application to compression

This course includes practical sessions.

Prerequisite: Geometry and analysis from L3 mathematics/applied mathematics

Learning outcomes: Tools for image processing (see objectives above)

1st year, 2nd semester (spring/summer semester)

Innovation and Entrepreneurship 2 (9 ECTS) SHORT INTERNSHIP 6ECTS

Description: Science industrial and/or business project.

Master ś electives (21 ECTS) HIGH THROUGHPUT IN BIOLOGY Responsible lecturer: Mickael Charrier ECTS: 6 Course type and working hours: 70h Exam type: oral exam (50%), Research project (50%)

Course outline: The lectures present the basic methodology and some advanced techniques used for high throughput in vitrosmall molecule drug discovery. The principles and statistical methods used for assay optimization and validation will also be explained.

1. Molecular biology, Biochemistry and Protein expression 2. Proteomic analysis; Mass spectrometry 3. Lab-chips and Cell-chips 4. Structural biology: Crystallogenesis and Crystallization; RMN 5. Combinatory chemistry

Prerequisite: Background in biochemistry, molecular biology, and cellular biology. Knowledge in physiology, immunology and microbiology will be appreciated. Students with laboratory and/or practical skills will better understand technological benefits of the use of high throughput technologies in the lab work.

Targeted skills: After completion of this course, students should be able to better understand and compare different high throughput methods used, for instance, for the discovery and validation of new biomarkers. Be able to present this knowledge in oral and written form.

HOW TO BECOME A CANCER CELL Responsible lecturer: Arnaud Seigneurin ECTS: 6

Course type and working hours: 50h Exam type: t.b.d.

Description: The objective of this course is to acquire the fundamental knowledge necessary to understand key mechanisms of cancer development. This course includes a set of lectures on alterations of cellular and molecular mechanisms that are responsible for the cancer pathophysiology. These modified cellular functions include for example the cell division, apoptosis, gene expression, stem cells, angiogenesis, and degradation of the extracellular matrix. The fundamental notions will be illustrated via their implications in diagnosis and therapeutics. The publication analysis will allow emphasizing the medical interest.

Learning outcomes:

• Knowledge in fundamental cancer cell biology, cancer cell-host relationship, basis on corresponding targeted therapeutics. • Ability to analyse biological data from published scientific manuscripts.

CANCER DISEASE:EXPERIMENTAL AND THERAPEUTICALLY APPROACHES Responsible lecturer: Claire Rome ECTS: 6 Course type and working hours: 40h Exam type: t.b.d.

Description: To provide a comprehensive overview of cancer from basic research to clinical trials: cancer physiopathology; cancer metabolism; proteomics; circadian rhythm and cancer cell characteristics; development of new anti-cancer drugs; imaging in cancer disease.

BIOSTATISTICS,BIOINFORMATICS,MODELLING Responsible lecturer: Adeline Leclercq-Samson ECTS:6 Course type and working hours: 39h Exam type: Oral exam (60%, 30 min), Research project (20%, 30min)

Course outline: At the end of the course, the students should be able to analyse an "omic" dataset. More precisely, they should be able:

• to load, explore and summarize graphically a dataset. • to compute confidence interval estimates for proportions, means and variances. • to formulate hypotheses, compute tests statistics, interpret p-values and make practical decisions for the • standard parametric and non-parametric tests. • to adjust simple and multiple linear models, analyses of variance (anovas), logistic regression, Cox model. • to select genes that explain a response variable by applying multiple testing approaches.

• to analyse a data set of differential gene expression.

MOLECULAR AND CELLULAR IMAGING (MICROSCOPY) Responsible lecturer: Arnaud Seigneurin ECTS: 6 Course type and working hours: 50h Exam type: t.b.d.

Lectures: Optical Microscopy:

• Basics of light microscopy, Köhler illumination, Contrast generation for transmitted light (Dark field, Polarized light, Phase contrast, DIC...) • Fluorescence Microscopy, F-techniques, Optical Sectioning and Confocal Microscopy (Laser scanning confocal, multiphoton microscopy...) • Processing and analysis of biological images

Electron Microscopy:

• Ultrastructural studies of the architecture of cellular components, viruses, and macromolecular assemblies by electron microscopy (Transmission Electron Microscopy, Scanning, Cryo-EM...) • Sample preparation, Image analysis

Discovering of X-Ray and near-field microscopies

Lab sessions:

• Optical Microscopy: Köhler illumination, Contrasts for transmitted light, Fluorescence microscopy, Image Processing • Visits on research platforms allow the students to become familiar with modern microscopy techniques (laser scanning multiphoton, super resolution, F-techniques, imaging methods in electron microscopy ...).

Learning outcomes:

• Acquisition by the students of autonomy on wide field optical microscopes, • Thorough knowledge of the principles of electron microscopes • Discovery of X-rays and near field microscopies. • Practice of processing and analysis of biological images with open-source software

DATA ANALYSIS,LINEAR MODELS AND ANOVA Responsible lecturer: Jean-Baptiste Durand, Clementine Prieur ECTS: 6 Course type and working hours: 54h Exam type: t.b.d.

Description: The aim of this course is to present advanced statistics and linear modelling, variance analysis and provide practical implementation

Contents:

• Principal components analysis (PCA) • Classification (Linear Discr. Analysis) • Data mining (text mining) • Linear regression • Estimation and test of regression parameters • ANOVA • ANCOVA • Practical implementation

This is a two parts course, including practical sessions:

1. 3ECTS = Lecture 13h + Practical 5h + Lab 15h. Course mutualized with Ensimag 2A 4MMFDASM(head: Jean-Baptiste Durand) 2. 3ECTS = Lecture 14h + Lab 6h -MSIAM specific course (in-depth and practical session) (head: Clémentine Prieur)

A short description of the course content can be found here.

Prerequisites: Elementary notions in probability theory (probability distribution, joint probability density function for random vectors, conditional distribution, expectation, variance, covariance, Gaussian distribution) Elementary notions in mathematical statistics (estimator, confidence interval, statistical tests). As a bonus: simple linear regression. Notions in linear algebra (matrix reductions). As a bonus :elementary notions in Rstudio and the R software.

COMPUTING SCIENCE FOR BIG DATA AND HPC Responsible lecturer: Clementine Prieur ECTS: 6 Course type and working hours: 54h Exam type: t.b.d.

Description: The aim of this course is to introduce numerical and computing problematics of large dimension problems.

Contents:

• Introduction to database • Introduction to big data • Introduction to high performance computing (HPC) • Numerical solvers for HPC

Prerequisite: First semester of M1 MSIAM.

Learning outcome: Algorithmics of big data and HPC

NUMERICAL OPTIMISATION Responsible lecturer: Laurent Desbat ECTS: 6 Course type and working hours: 54h Exam type: t.b.d.

Description: This program combines case studies coming from real life problems or models and lectures providing the mathematical and numerical backgrounds.

Contents:

• Introduction, classification, examples. • Theoretical results: convexity and compacity, optimality conditions, KT theorem • Algorithmic for unconstrained optimisation (descent, line search, (quasi) Newton) • Algorithms for non-differentiable problems • Algorithms for constrained optimisation: penalization, SQP methods • Applications

Prerequisite: linear algebra, differential calculus

Learning outcomes:

• Recognize and classify optimisation problems. • Solve optimisation problems using adequate algorithms and methods. • Practical implementation

ADVANCED BIOSTATISTICS AND EXPLOITATION OF RESEARCH WORK Responsible lecturer: Jean-luc Bosson ECTS: 6 Course type and working hours: 44h Exam type: t.b.d.

Parametric tests (Correlation -Regression, ANOVA, censored data analysis, PCA ...) and nonparametric tests (small sample cases) are discussed, up to complex multivariate models. The R -Rstudio software tool is used for the application examples.

There are 11 face-to-face sessions of SEPI or TD with 1 collaborative assignment (2 or 3 max) to be produced before the final MCQ-based exam. These assignments will be structured as statistical analysis protocols, preliminary stage of a scientific article and associated with a synthesis of multimedia presentation of the realized works.

2nd year, 1st semester (autumn/winter semester)

Innovation and Entrepreneurship 3 (9 ECTS) THE HMDA Ś SCHOOL ON LEARNING FROM HEALTH DATA This summer school is offered and held by UGA for all HMDA students, which includes the preparation, execution, and documentation of a practical project, which is based on a real use case in the healthcare context.

Responsible lecturer: various ECTS: 5 Course type: workshop/laboratory course (150 hours in total, including self-study and preparation of tasks in teams) Exam type: /

Contents: With its yet unfulfilled promise to revolutionize the healthcare economy and save billions of euros in the process, Artificial Intelligence (AI) and health data management in general are exploding in popularity. Indeed, the growth of the global AI health market is expected to reach US$6.6 billion by 2021.

But can AI and data-driven technologies truly live up to expectations in the field of health?

Over 5 demanding days at this exciting bioHealth Computing school, graduate students (Master & PhD) and early career professionals in science, informatics and healthcare are immersed in a challenging mix of theoretical and practical sessions on AI technology and innovation and coached to develop business models of market-acceptable products and services using AI technologies.

Learning from Health Data is an accelerated learning programme proposed by a consortium of EIT- Health partner universities and co-organised by the Université Grenoble-Alpes and ESI-Archamps. The school is fully in line with EU goals to deliver innovation-led solutions enabling European citizens to live longer, healthier lives.

The school adheres to the 2030 Agenda for Sustainable Development of the UN, and in particular to the objectives of the UHC2030 programme whose mission is to create a movement for accelerating equitable and sustainable progress towards universal health coverage (UHC).

The application form includes a section where candidates should provide a 50 to 200-word outline of an innovative idea or project related to health and medical data analytics. This might be expressed in terms of:

• An unmet need in healthcare which could benefit from the development of data-driven products or services. • The (re)deployment of an existing technology to provide an innovative product or service for healthcare. • Currently unavailable but potentially marketable product or service involving data-driven technology for healthcare.

The best ideas may serve as the basis for a group project in the Business Development & Innovation component of the school.

Learning outcomes and competencies: Students can

• apply industry standard techniques and team management. • develop, independently and within a team, their problem-solving and creative skills. • implement their ideas as prototypes by applying agile software development methods. • use the results of their projects for the creation of start-ups. • develop world-class solutions in the field of IT and health technologies, address societal challenges, contribute to the competitiveness of Europe, • focus on unmet needs in healthcare, AI candidate technologies, experienced-based co-design, business creation, health assessment and regulatory affairs. • participate a series of advanced courses and hands-on activities on IT proposed by experts from partner universities, hospitals, and industries. One breakout session on advanced application in health will examine several uses of machine learning, big data and internet of things presented by leaders in Health Research and Development, reviewing the latest techniques. • develop innovative ideas in multidisciplinary teams translating them into value creation through a business model and under the health regulation framework.

ENTREPRENEURSHIP AND INNOVATION Responsible lecturer: Daniel Bernard ECTS: 3 Course type and working hours: 39h Exam type: t.b.d.

Objective and learning outcomes: The aim of this course is to

• Understand the main concepts related to technology ventures management. • Understand and apply the main tools for business model generation. • Develop the ability to search, analyse and combine business and technology information to build a business plan.

Content:

1. Introduction to entrepreneurship: main concepts about technology new ventures. Main Founders’ characteristics. 2. Establishing the business idea: Technology search and analysis. 3. Customer analysis: Market segmentation; End user analysis; Total Addressable Market (TAM) analysis. 4. Value proposition: Full Life Cycle; Product specification; Quantifying the value proposition.; Competitive analysis. 5. Product acquisition: understanding customers’ decision process. 6. Business model generation: Business model canvas; Components’ analysis; Design the business model; Lifetime Value of a customer; Cost of a Customer Acquisition. 7. Product design: defining the Minimum Viable Business Product. 8. Scaling the business & Building up the company: Technology venture financing; Negotiating deals; Company registration.

Learning outcomes: Students can

• apply industry standard techniques and team management. • develop, independently and within a team, their problem-solving and creative skills. • implement their ideas as prototypes by applying agile software development methods. • use the results of their projects for the creation of start-ups. • develop world-class solutions in the field of IT and health technologies, address societal challenges, contribute to the competitiveness of Europe, • focus on unmet needs in healthcare, AI candidate technologies, experienced-based co-design, business creation, health assessment and regulatory affairs. • participate a series of advanced courses and hands-on activities on IT proposed by experts from partner universities, hospitals, and industries. One breakout session on advanced application in health will examine several uses of machine learning, big data and internet of things presented by leaders in Health Research and Development, reviewing the latest techniques. • develop innovative ideas in multidisciplinary teams translating them into value creation through a business model and under the health regulation framework.

ENTREPRENEURIAL PROCESS & TOOLS Responsible lecturer: Caroline Tarillon ECTS: 6 Course type and working hours: 90h Exam type: t.b.d.

Objective and learning outcomes: The aim of this course, is to

• Analyse the functions developed by a biomedical engineer within an organization. • Understand managerial concepts in a business environment.

Learning outcomes: During this course, students will

• Develop the ability to search, analyse and combine business information for decision making. • Understand management of the main functional areas of a company: marketing, operations, finance, human resources, and R&D.

Content:

1. Strategic management: Management Strategy; Business values and orientation; External analysis; Internal analysis; Corporate, Business and Functional strategies 2. Marketing: Strategic marketing; Operative marketing: Four P's 3. Operations Management: Definition & Evolution; Strategies; Supply chain; Quality management; Five P's (product, process, plan, programme and people) 4. Human Resources: Planning; Recruiting; Selection; Training; Performance appraisal; Compensation. 5. Innovation management (R&D): Sources; Innovation types ; Disruptive innovation ; Managing innovation. 6. Finance: General concepts on financial cycles; Main financial documents; Cost Accounting.

RESEARCH AND DEVELOPMENT PROJECTS Responsible lecturer: Jean Breton ECTS: 3 Course type and working hours: 70h Exam type: t.b.d.

Objective and learning outcomes: The aim of this course is to learn students how to apply the scientific method in the development of research and development projects, as well as in the dissemination of project results.

Learning outcomes: During this course, students will

• Work on bibliographical subject and make a critical discussion about the results in oral and public presentation. • Perform an individual and a teamwork, developing the ability to search information sources, analyse the legislation for the collaboration of public and private entities, and apply methods for management of research project.

BIOMEDICINES INNOVATIVE PROJECT Responsible lecturer: Jean Breton ECTS: 6 Course type and working hours: 30h Exam type: t.b.d.

Content: This curse takes a very practical, applied approach to the challenges of successful project management. The essentials cover the following: structuring projects to set realistic goals and identify milestones;

using effective tools for scheduling and be able to run single or parallel projects; identify project risks ; manage time, cost, and quality; implement control systems to keep on top of the project.

Target: The Biomedicines innovative project has a wide range of possible applications for any initiative whose completion is fixed within specific time limits

COMPUTATIONAL MEDICINE SUMMER SCHOOL [COPD and Chronic conditions as Case Study] Responsible lecturer: Philippe Sabatier ECTS: 6 Course type and working hours: 85h Exam type: t.b.d.

Objective and learning outcomes: CompMed is a Summer School, which expose participants to some of the latest biomedical advances. Students are invited to discover how engineering and computing solutions could be used to promote healthy living and active aging. The participants can leave a transformational experience, which will help them to develop transferable skills necessary for successful innovation. During the school, the students meet innovative scientist and high-tech start-up companies, and can discuss their ideas with highly qualified professors and young entrepreneurs. They are introduced to Creative Thinking, will have a unique opportunity to promote (pitch) their ideas in front of a medical/business panel, with the opportunity for further development of promising teams, and potential links to Health Accelerator.

Learning outcomes: By taking this course participants will

• Know the definition of biological/pathological process; chronic disease; electronic health records; semantic data, mathematical modelling; Investigate biological/pathological process. • Learn how design theoretical models; integrate multi-scale modelling; support interaction between deterministic model and probabilistic models; study perturbations of a biological process; explore the toolbox of biomathematics modelling. • Study data representation and integration; explore semantic technologies and translational medicine; assess data warehouse solutions in terms of their targeted medical use case : data sharing, data interoperability and knowledge discovery. • Discover and manage databases; Use specific and publicly available datasets (BioBridge and PAC-COPD); Create semantic mapping on inference engineering; Analyze associations of co- morbidities in PAC_COPD and in the Medicare database (diseasome) • Integrated oxygen transport models from atmosphere to cell with mitochondrial reactive oxygen species (ROS) generation and metabolic pathways. • Provide intuitive users interfaces for clinician and bio-researchers.

Content: CompMed learn clinicians, scientists, and engineers in computational medicine and healthcare optimization for chronic diseases (CD) focused on COPD as case study. CompMed aims to use a system approach for the study of the underlying mechanisms of diseases’ phenotypes associated with poor

prognosis. To address this tricky challenge, CompMed propose a dynamic approach, based on complex problem solving. COPD is caused by unhealthy lifestyles and increased life expectancy, complex gene environment interactions, intrinsic host responses, such as local and systemic inflammation, epigenetic changes and decoupling of basic regulatory mechanisms with impact on bioenergetics, metabolome, proteome, genome, microbiome, immune responses, and remodeling. Through the case study on COPD, participants are invited to work, step by step, on a better understanding of the physio-pathological mechanisms explaining co-morbidity, which seems to be crucial not only for a better diagnosis and treatment, but also to envision new formats of service delivery considering early patient stratification based on characterization of disease phenotypes. The learning process is based on the iterative loop of Systems Medicine approach:

1. Formulate and formalize a question. 2. Define, Integrate and Perturb systematically the system components. 3. Check model predictions; Collect appropriate data sets(targeted and global); Compare the 4. Observed/predicted results. 5. Refine the model so that its predictions fit better with the experimental observations; Iterate the 6. Process until an answer to the initial question is obtained.

The program culminates in a capstone design-project in which students work in interdisciplinary teams co-advised by faculty members and investigators from industries and hospitals. The participants are introduced to Creative Thinking, and throughout the last day, they present (pitch) their projects in front of a business/medical panel, who will join the session and provide their valuable feedback. CompMed provides participants with a broad but high-level scientific background and important skills such as conceptual approaches, teamwork, management of complex processes, entrepreneurship, and high intercultural awareness. The school combines an intensive program of lectures, hands-on sessions(experiments, simulation, and modeling) and group working. Courses are given by teachers from France, Spain, United States, Sweden, Germany, the Netherlands, Great-Britain, and Switzerland.

HMDA specialization (21 ECTS) MOLECULAR TOOLS FOR THE DIAGNOSIS AND TREATMENT OF GENETIC DISEASES Responsible lecturer: Jean Breton ECTS: 3 Course type and working hours: 29h Exam type: t.b.d.

Course module description This course will cover the principles of Molecular Diagnosis which is the process of identifying a disease by studying molecules, such as proteins, DNA, and RNA, in a tissue or fluid. Molecular diagnostics is a new discipline that captures genomic and proteomic expression patterns and uses the information to distinguish between two or more conditions at the molecular level. The conditions under investigation can be human genetic disease or infectious diseases. Molecular diagnostics is not confined to human diseases but can be also used in environmental monitoring, food processing...etc.

Many of the diagnostic techniques are developed and marketed in kit format by biotechnology companies. The main source of information is web sites of companies that develop and market the molecular diagnostic kits. New methods are continuously developed. The objective of this course is learning and understanding how molecular techniques that were studied in other classes can be developed and utilized in diagnosis and sold in diagnostic kits.

Intended learning outcomes:

• Knowledge and understanding of the basic principles used in molecular diagnosis. • Gain thinking and analysis skills to understand new diagnostic methods. • Ability to collect information to develop a new diagnostic kit. • Knowledge and skills gained in the course should be useful in practical life in developing or using diagnostic kits.

MOLECULAR MARKERS FOR MEDICAL IMAGING Responsible lecturer: Franz Bruckert ECTS: 3 Course type and working hours: 30h Exam type: Written report based on articles proposed by the lecturers written report.

Goals Series of seminars on the development and use of contrast agents and molecular markers in biomedical imaging and therapy.

Content Contrast agents and molecular markers for ultrasound, X-Ray, and MRI imaging Development and use of visible and infrared fluorophores. Application to cardiovascular diseases, cancer, and neurodegenerative diseases.

Prerequisites Molecular biology and physiology courses

NUMERICAL SIMULATIONS AND STATISTICAL DATA ANALYSIS Responsible lecturer: Judith Peters ECTS: 3 Course type and working hours: 30h Exam type: Written exam (2h)

Content: The course will deal with the uncertainties of the experimental data, including the notions of probability, the random variable, and distributions. Some hypothesis tests and confidence intervals will be presented. Simulation methods such as the Monte Carlo method will be introduced. The acquired techniques will be applied to selected examples in the field of life sciences and the environment.

MODELLING IN SYSTEMS BIOLOGY Responsible lecturer: Delphine Ropers ECTS: 3 Course type and working hours: 30h Exam type: only in English, continuous evaluation with homework assignment.

Goals The course introduces systems biology by focusing on the behaviors emerging from interactions between genes, proteins and RNAs, taking examples from microbes to mammals. The main goal of this course is to show students that abstract computational and mathematical methods can be effectively employed for in silico modeling and analysis of living organisms. Moreover, to enhance practical skills, students will apply some of the techniques and software tools to analyze genome-scale models and models of cell metabolism and gene expression.

Content The different steps of the model development will be presented: initial observations, hypotheses, model testing and validation. Different types of models will be described and illustrated, for instance: deterministic versus stochastic, static versus dynamic or versus non-parametric, lumped versus distributed. These notions will be illustrated by mathematical models in the biomedical field as for instance physiological models (Hodgkin-Huxley), compartment models or population models.

• Introduction to cellular networks and mathematical modelling (course: 2h) • Mathematical modelling of cell metabolism: flux balance analysis (course: 2 hours + 2 hours computer lab exercises) • Kinetic models of integrated networks and introduction to other modelling frameworks (4 hours + 2 hours computer lab exercises) • Identification and inference of metabolic network models (2 hours + 2 hours computer lab exercises)

Intended Learning Outcomes

• Understand basic knowledge about biological systems in order to model them. • Understand and be able to model different types of biological systems by using appropriate modelling tools. • Choose appropriate models and argue about these choices depending on the modelling application. • Make a critical analysis about the relevance and interest of mathematical models of biological systems in their capacity to predict new experimental results and inspire original experimental protocols. • Use software and computers to implement and simulate mathematical models of biological systems.

Learning outcomes

• Make a critical analysis of the scientific literature devoted to the development of original mathematical models of biological systems.

• Make a concise and critical presentation of a scientific article related to mathematical models of biological systems.

ADVANCED ALGORITHMS FOR MACHINE LEARNING AND DATA MINING Responsible lecturer: Eric Gaussier ECTS: 3 Course type and working hours: 18h Exam type: t.b.d.

Description

• A prior algorithm (Frequent item sets) & Page Rank • Monte-carlo, MCMC methods: Metropolis-Hastings and Gibbs Sampling • Matrix Factorization (Stochastic Gradient Descent, SVD) • Generalized means and its variants (Bach, Online, large scale), Kernel clustering (Support Vector Clustering), Spectral clustering • Classification and Regression Trees, Support Vector regression • Alignment and matching algorithms (local/global, pairwise/multiple), dynamic programming, Hungarian algorithm,...

Prerequisite Fundamentals of probability/statistics, linear algebra, and computer science (data structures and algorithms)

Bibliography

1. C.D. Manning, P. Raghavan and H. Schütze. «Introduction to Information»Retrieval.Cambridge University Press, USA, 2008. 2. A. DasGupta. «Probability for Statistics and Machine Learning». Springer, 2011. 3. I. Goodfellow, Y. Bengio, A. Courville. «Deep Learning». MIT Press, 2016.C.M. Bishop. «Pattern Recognition and Machine Learning». Springer Verlag, 2006.

ADVANCED LEARNING MODELS Responsible lecturer: Julien Mairal ECTS: 3 Course type and working hours: 18h Exam type: written homework with theoretical exercise, participation in a data challenge.

Description Statistical learning is about the construction and study of systems that can automatically learn from data. With the emergence of massive datasets commonly encountered today, the need for powerful machine learning is of acute importance. Examples of successful applications include effective web search, anti-spam software, computer vision, robotics, practical speech recognition, and a deeper understanding of the human genome. This course gives an introduction to this exciting field, with a strong focus on kernels methods and neural network models as a versatile tool to represent data This course deals with:

Topic 1: Neural networks

• Basic multi-layer networks • Convolutional networks for image data • Recurrent networks for sequence data • Generative neural network models

Topic 2: Kernel methods

• Theory of RKHS and kernels • Supervised learning with kernels • Unsupervised learning with kernels • Kernels for structured data • Kernels for generative models

Prerequisites Fundamental notions in linear algebra and statistics. Basic programming skills to implement a machine method of choice encountered in the course from scratch → http://thoth.inrialpes.fr/people/mairal/teaching/2017-2018/MSIAM

BAYESIAN STATISTICS Responsible lecturer: Julyan Arbel ECTS: 3 Course type and working hours: 18h Exam type: t.b.d.

Objectives The course aims at providing an overview of Bayesian parametric and nonparametric statistics. Students will learn how to model statistical and machine learning problems from a Bayesian perspective and study theoretical properties of the models.

Syllabus This course is in two parts covering fundamentals of Bayesian parametric and nonparametric inference, respectively. It focuses on the key probabilistic concepts and stochastic modelling tools at the basis of the most recent advances in the field.

Part 1

• Foundations of Bayesian inference: exchangeability, de Finetti's representation theorem • Conjugacy in simple models (binomial, Poisson, Gaussian) • Some elements of posterior sampling, Markov chain Monte Carlo • Bayesian neural networks and their Gaussian process limit

Part 2

• Clustering and Dirichlet process, random partitions • Models beyond the Dirichlet process, random measures, Indian buffet process • Some elements of Bayesian asymptotic

Bibliography

• Hoff, P. D. (2009). A first course in Bayesian statistical methods. Springer Science & Business Media. • Neal, R. M. (2012). Bayesian learning for neural networks (Vol. 118). Springer Science & Business Media. • Hjort, N. L., Holmes, C., Müller, P., & Walker, S. G. (2010). Bayesian nonparametric. Cambridge series in statistical and probabilistic mathematics. Cambridge: Cambridge Univ. Press. • Orbanz, P. (2012). Lecture Notes on Bayesian Nonparametric • Kleijn, B., van der Vaart, A., & van Zanten, H. (2012). Lectures on Nonparametric Bayesian Statistics.

CATEGORY LEARNING AND OBJECT RECOGNITION Responsible lecturer: Jakob Verbeek ECTS: 3 Course type and working hours: 18h Exam type: t.b.d.

Description: In this course we present recent state-of-the-art methods for visual object category representation and recognition, and the techniques that underpin these methods. Methods will include so called "bag of features" approaches, Fisher vectors, and convolutional neural networks for tasks such as instance-level image retrieval, image classification, object localization, semantic segmentation, image caption generation and action recognition in videos. On the machine learning side we consider clustering methods (k-means, mixture of Gaussians), classification techniques (SVM, logistic discriminant), and kernels to obtain non-linear classifiers, as well as the principles underlying neural networks (multi-layer perceptron, back-propagation, convolutional networks, recurrent networks).

COMPUTATIONAL BIOLOGY Responsible lecturer: Olivier François, Michaël Blum ECTS: 3 Course type and working hours: 18h Exam type: oral presentation, written exam.

Description: This interdisciplinary MSc course is designed for applicants with a biomedical, computational, or mathematical background. It provides students with the necessary skills to produce effective research in

bioinformatics and computational biology. The objective of the course is to introduce mathematical biology questions, stochastic and deterministic approaches for modeling biological systems, and advanced tools for the analysis of biological data. The topics addressed in this course include a brief introduction to stochastic processes and differential equations and their application to biological problems. The first part of the course focuses on modelling in molecular biology and evolution, and on the analysis of molecular phylogenetic or population genetic data. The second part of the course focuses on models in cellular biology and biomechanics.

Prerequisites: No specific prerequisites.

MACHINE LEARNING FUNDAMENTALS Responsible lecturer: Massih-Reza Amini, Marianne Clausel ECTS: 3 Course type and working hours: 30h Exam type: Homeworks (30%), final exam (70%)

Description:

• Consistency of the Empirical Risk Minimization • Uniform Generalization Bounds and Structural Risk Minimization • Unconstrained Convex Optimization • Binary Classification algorithms (Perceptron, Adaboost, Logistic Regression, SVM) and their link with the ERM and the SRM principles • Multiclass classification • Application and experimentations

Prerequisites Statistics and probability (BSc)

Learning outcomes Understanding of fundamental notions in Machine Learning (inference, ERM and SRM principles, generalization bounds, classical learning models, unsupervised learning, semi-supervised learning.

Bibliography

• Massih-Reza Amini -Apprentissage Machine de la théorie à la pratique, Eyrolles, 2015. • Christopher Bishop -Neural Networks for Pattern Recognition, Oxford University Press, 1995. • Richard Duda, Peter Hart & David Strok -Pattern Classification, John Wiley & Sons, 1997. • John Shawe-Taylor & Nello Cristianini -Kernel Methods for Pattern Analysis, Cambridge University Press, 2004. • Colin McDiarmid -On the method of bounded differences, Surveys in Combinatorics, 141:148- 188, 1989. • Mehryar Mohri, Afshin Rostamzadeh & Ameet Talwalker -Foundations of Machine Learning, MIT Press, 2012. • Bernhard Schölkopf & Alexander J. Smola -Learning with Kernels, MIT Press, 2002.

• Vladimir Kolchinskii -Rademacher penalties and structural risk minimization, IEEETransactions on Information Theory, 47(5):1902–1914, 2001

FUNDAMENTALS OF PROBALISTIC DATA MINING Responsible lecturer: Jean-Baptiste Durand ECTS: 3 Course type and working hours: 30h Exam type: written exam (2h), 2 reports.

Description: This lecture introduces fundamental concepts and associated numerical methods in model-based clustering, classification, and models with latent structure. These approaches are particularly relevant to model random vectors, sequences, or graphs, to account for data heterogeneity, and to present general principles in statistical modelling. The following topics are addressed:

• Principles of probabilistic data mining and generative models; models with latent variables • Probabilistic graphical models • Mixture models and clustering • PCA and probabilistic PCA • Nonparametric density estimation • Generative models for series and graphs: hidden Markov models

Prerequisites: Fundamental principles in probability theory (conditioning) and statistics (maximum likelihood estimator and its usual asymptotic properties).Constrained optimization, Lagrange multipliers.

Learning outcomes: At the end of the course, the student will be able to perform model-based clustering, analysis, and segmentation of time-series with hidden Markov models, build a graphical model associated with a given distribution and represent numerical multivariate data with missing coordinates into planes.

Bibliography:

• Lauritzen, S.L. Graphical Models. Clarendon Press, Oxford, United Kingdom, 1996. • Koller, D. and Friedman, N. Probabilistic graphical models: principles and techniques. MIT press, 2009.Bishop, Christopher M. Pattern Recognition and Machine Learning. Springer Verlag, 2006.

MODEL SELECTION FOR LARGE-SCALE LEARNING Responsible lecturer: Emilie Devijver ECTS: 3 Course type and working hours: 60h Exam type: t.b.d.

Description When estimating parameters in a statistical model, sharp calibration is important to get optimal performances. In this course, we will focus on the selection of estimators with respect to the data. Particularly, we will consider calibration of parameters (e.g., regularization parameter for minimization of regularized empirical risk, like Lasso or Ridge estimators) and model selection (where each estimator minimizes the empirical risk on a specified model, as mixture models with several number of clusters). We will focus on the penalized empirical risk, where the penalty may be deterministic (as BIC or ICL) or estimated with data (as the slope heuristic).

Prerequisites: Basic knowledge in probability and statistics

Target skills: Learn.

• When model selection is needed. • What can be proved theoretically for existing methods. • How those results can help in practice to choose a criterion for some specific statistical problem • How the theory can serve to define new procedures of selection.

References:

• T. Hastie, R. Tibshiraniand J. Friedman, The Elements of Statistical Learning. Data Mining, Inference, and Prediction • P. Buhlmann and S. van de Geer, Statistics for High-Dimensional Data. Methods, Theory and Applications • P. Massart, Concentration Inequalities and Model Selection

INFORMATION VISUALIZATION Responsible lecturer: Renaud Blanch ECTS: 3 Course type and working hours: 18h Exam type: t.b.d.

Description Interactive Information Visualization (InfoVis) is the study of interactive graphical representations of abstract data (e.g., graphs linking people in social networks, series of stock-options values evolving over time). Graphical representations are a powerful way to leverage the human perceptual capabilities to allow the user to explore and make sense of abstract data, and also to expose findings and convey ideas. But to be efficient, a visualization has to be designed using knowledge about the human visual perception, the characteristics of the data, the kind of task that will be performed on those data. The aim of this course is to provide the keys, both theoretical and practical, to build usable and useful interactive visualizations.

Program summary:

• foundations: human visual perception, graphical variables, data types, the visualization pipeline. • linked data: tree and graph visualization • tabular data: time series and spatial data visualization • dealing with large data: aggregation, multiple views, interaction • validating visualization: visualization tasks, evaluation

DATA CHALLENGES Responsible lecturer: Jean-Baptiste Durand ECTS: 3 Course type and working hours: 60h Exam type: report (1/3), oral presentation (1/3), score (1/3)

Description Face up challenging real-world problems in machine learning, be involved in multidisciplinary teams of data scientists, computer scientists, mathematicians, and expert students in signal processing, and contribute to leading your team to the top rank! Different teams with M2 students issued from either MSIAM Data Science, MoSIG Data Science and SIGMAwork on a same challenge on either complex, structured, or big data, and maybe a combination of all three. Try and compare different approaches, take benefit from the computational power of clusters and from advice of your supervisors. The data challenges stretch on several months, include some tutored sessions, if needed mini-courses, and of course your regular involvement over that period of time.

Prerequisites Elementary notions in probability theory (multivariate distributions), machine learning (concepts of regression, classification, and clustering) and programming (usually python, although other languages may be chosen).

Learning outcomes At the end of the course, the student will be able to work in teams involving various skills (machine learning, statistical modelling, programming, data bases and others). They will acquire skills in data analysis and self-training in acquiring or reinforcing skills among the four listed above.

Bibliography

• Dopplick, R. Expanding minds to big data and data sciences. Inroads, 6(3) 88, 2015. • Yang, J.How we did it: Jie and Neeral on winning the first Kaggle-in-class competition at Stanford, 2010. • Organization of the data challenge in 2018-2019 • Detailed description of GBX9AM20; MSIAM Course list; Semester 3 (MSIAM tracks)

DATA MANAGEMENT IN LARGE-SCALE DISTRIBUTED SYSTEMS Responsible lecturer: Thomas Ropars ECTS: 3 Course type and working hours: 18h Exam type: written exam, practical work.

Description Target skills : Data management and knowledge extraction have become the core activities of most organizations. The increasing speed at which systems and users generate data has led to many interesting challenges, both in the industry and in the research community. The data management infrastructure is growing fast, leading to the creation of large data centers and federations of data centers. These can no longer be handled exclusively with classic DBMS. It requires a variety of flexible data models (relational, NoSQL...), consistency semantics and algorithms issued by the database and distributed system communities. In addition, large-scale systems are more prone to failures, and should implement appropriate fault tolerance mechanisms. The dissemination of an increasing number of sensors and devices in our environment highly contribute to the “Big Data” and the development of ubiquitous information systems. Data is processed in continuous streams providing information related of user’s context, such as their movement patterns and their surroundings. This data can be used to improve the context awareness of mobile applications and directly target the needs of the users without requiring an explicit query. Combining large amounts of data from different sources offers many opportunities in the domains of data mining and knowledge discovery. Heterogeneous data, once reconciled, can be used to produce new information to adapt to the behavior of users and their context, thus generating a richer and more diverse experience. As more data becomes available, innovative data analysis algorithms are conceived to provide new services, focusing on two key aspects: accuracy and scalability. Program summary : In this course, we will study the fundamentals and research trends of distributed data management, including distributed query evaluation, consistency models and data integration. We will give an overview of large-scale data management systems, peer-to-peer approaches, MapReduce frameworks and NoSQL systems. Ubiquitous data management and crowdsourcing will also be discussed.

Prerequisites Fundamentals of DBMS, parallel programming (threads)

Learning outcomes At the end of the course, the students will know how to use Big Data software tools to efficiently store and process large amounts of data, including tools that can operate in real time.

Bibliography

• Dean, Jeffrey, and Sanjay Ghemawat. “MapReduce: simplified data processing on large clusters.” Communications of the ACM 51.1 (2008): 107-113. • Zaharia, Matei, et al. “Apache spark: a unified engine for big data processing.” Communications of the ACM 59.11 (2016): 56-65. • Murray, Derek G., et al. “Naiad: a timely dataflow system.” Proceedings of the Twenty-Fourth ACM Symposium on Operating Systems Principles. ACM, 2013. • Lakshman, Avinash, and Prashant Malik. “Cassandra: a decentralized structured storage system.” ACM SIGOPS Operating Systems Review 44.2 (2010): 35-40.

DATA SCIENCE SEMINAR Responsible lecturer: Jean-Baptiste Durand, Ronald Phlypo,Olivier Michel ECTS: 3 Course type and working hours: seminar Exam type: oral exam.

Description Our master programs now include a series of 6 or 7 seminars given by active researchers in the field of data processing methods and analysis. These seminars are intended to give students some insights on modern problems and solutions developed in a data science framework, with applications in a variety of fields. In order to make these seminars a most valuable experience for all students, a scientific paper dealing with the topic of the seminar will be selected by the speaker and dispatched to all students about 2 weeks before the seminar. Students are expected to read and study this paper, and to prepare questions, before attending the seminar. Presence at the seminars is compulsory for master students. At the end of the seminar series, some oral exam is organized. One of the topics presented during the seminars is randomly assigned to each student a few days in advance. The oral exam consists in a 25 min summarized presentation of the scientific issues that were addressed, and a 15 min session of discussion and questions. A second different topic is chosen by the student, and he/she must write a report on that topic, based on the seminar and associated articles.

SOFTWARE DEVELOPMENT TOOLS AND METHODS Responsible lecturer: Mourad Ismail ECTS: 3 Course type and working hours: seminar, 39h Exam type: practical session reports, oral presentation.

Description The aim of this course is to study various useful applications, libraries and methods for software engineering related to applied mathematics. For example:

• C++ project management (git and/or svn) • Development and profiling • Boost library • Linear algebra (Eigen) • Prototyping and interfacing using Python • Post processing and visualization tools (VTK,Paraview, GMSH)

This course deals with :

Topic 1: Software Engineering Topic 2: Programming

Prerequisite Linear algebra: fundamental notions (matrices, linear functions), Programming in C++ and python

Learning outcomes At the end of the course, students will be able to manage and couple different libraries, to correctly debug a code (find memory leaks for example).

2nd year, 2nd semester (spring/summer semester)

Master’s thesis in collaboration with industry partner or hospital 30 ECTS Objective and learning outcomes.

• The master thesis in Biomedical Engineering is typically a research project or study, or an extended analysis of a topic of scientific or technological nature. The goal is for students to perform research and apply the knowledge acquired during their studies while at the same time developing skills like initiative, autonomy skills, decision, and organization. • The main learning output of this master thesis is the students’ ability to work on a BME program, and to translate research into applications –in cooperation with a non-academic business partner. This innovative approach is based on the mobility of students, exchanging experiences in different disciplines, and establishing a common high-quality standard in education and training. • At the end of the master thesis students work should reflect the EIT OLOs 1-6.

Content The program is defined according to the supervisor orientation and to the type of theme and it is developed during one of the semesters of the last year of the MSc Course. Following the MSc procedures, the Education Committees has assessed the scientific quality and feasibility of the master thesis proposals. It is anticipated that the thesis will be relevant to the student's track and will address a question of importance in the student's field of expertise. Students are expected to design a research project, write a formal research protocol, perform the study described in it, and prepare a comprehensive scholarly scientific paper reporting the results. Optionally, part of the master project could be done at another institute or company outside Partner Universities, but in this case, it is always under supervision of a Partner University staff member. To achieve their research project, students are required to write a scientific paper under guidance of their research supervisors, and to give a presentation about the research performed. The scientific paper must be approved by the academic supervisor and be suitable for submission to an international, scientific journal. The thesis can take place at universities, research centers or companies. As mentioned above, the thesis requires an advisor from the Engineering side and a co-advisor from a non-academic partner.

SUMMER SCHOOL ARTIFICIAL INTELLIGENCE FOR HEALTH 6 ECTS, Philippe SABATIER [email protected]

Objective and learning outcomes AI4Health’s mission is to teach advanced topics related to Big Data computing and analytics for health and wellbeing, as well as to enhance innovation and entrepreneurial awareness amongst participants.

The participants can leave a transformational experience, which will help them to develop transferable skills necessary for successful innovation. During the school, the students meet innovative scientist and high-tech start-up companies, and can discuss their ideas with highly qualified professors and young entrepreneurs. They are introduced to Creative Thinking, will have a unique opportunity to promote (pitch) their ideas in front of a medical/business panel, with the opportunity for further development of promising teams, and potential links to Health Accelerator. IBD4Health is part of the bioHC program which educates since 2011 outstanding minds and cultivates future leaders who will explore fundamental principles underlying disease and design new biomedical technologies for health and wellbeing.

Outcomes By taking this course participants will:

• Know definition Big Data: (volume, velocity, variety, value), challenges. Study how and where Big Data challenges arise in several domains (including medicine, social media, insurance, and finance). Investigate web challenges and how to engineer around them. Discover user interfaces for Big Data and their functions. • Learn how designing theoretical models to investigate biological and pathological process of Chronic Diseases. Use the toolbox of biomathematics modelling. • Work on data representation and integration, in particular with semantic web technologies and data sharing, data interoperability and knowledge discovery. Explore the SQL & NoSQL systems, their capabilities, and pitfalls, and how the NewSQL movement addresses these issues in terms of scalability, performance, and robustness. • Explore data representation and integration, in particular with semantic web technologies and translational medicine. Assess existing solutions in terms of their target medical use case and of data sharing, data interoperability and knowledge discovery. • Analyse Big Data: statistical learning and evidence-based medicine. • Propose technology transfer in healthcare, and population wellbeing. Study the ethical issues and discover recent techniques that help building secure Big Data systems, resolving the challenges of sharing protected data and of integrating semantic-driven technologies into the clinical practice. • Solve the obesity study case by team working, Writing, and defending orally an executive summary.

Content IBD4Health focuses on the technology, unmet needs, business, value chain, privacy, and security aspects of future Health Clouds. Data collection on a large scale could provide materials for in-depth analysis of different causal and contributory factors, supporting the development of effective interventions and public health approaches to tackle obesity. IBD4Health comprises interactive morning sessions, including guest presentations, and computer practice on case study, and afternoon sessions focused on innovation and entrepreneurship, including workshops, group assignments, etc. The programme is organised in three sessions: 1st Session: Health, wellbeing, and data challenges; 2nd Session: Data oriented design, collection, computing and analytics; 3rd Session: Innovation and entrepreneurship. The School’s program starts with an Introductive round table (session 1) and ends with a Pitching session and a Design Thinking Workshop ‘session 3). Everyday sessions mix a comprehensive range of practical tools and real-life experiences. Time off between the sessions allowed participants to work on individual or team projects. Lectures & Practical Exercise on Data Design, Collection, Computing, Analytics, introduce the state-of- the-art on Big Data Technologies: from oriented design, collection, storage, computing, security and privacy and big data. Through the case study on obesity, multidisciplinary participants are invited to

discover, step by step, the main sources of data and efforts to develop new tools on large-scale data processing application, which involves very complex systems and sophisticated mechanisms. Innovation & Entrepreneurship Session is devoted to working on innovative ideas and translating them into value creation through business model approach. At the end of the Opening Day, 5 Group Projects (GP) are selected from the Personal Projects (PP) proposed on the student’s applications. Students are invited to explore the following aspects: unmet need, mobilized technology, developments to achieve, market analysis, etc. Participants are introduced to Creative Thinking as well as applied Design Thinking and Pitching. CERN BIC network proposes teaching based on the expertise of hosted SME and Spin-off. IBD4Health provides participants with a broad but high-level scientific background and important skills such as conceptual approaches, teamwork, management of complex processes, entrepreneurship, and high intercultural awareness. The school combines an intensive programme of lectures, hands-on sessions (experiments, simulation, and modelling) and group working. Courses are given by teachers from France, Spain, United States, Sweden, Germany, the Netherlands, Great-Britain, and Switzerland.

SUMMER SCHOOL ON SAFER NANOMATERIALS 6 ECTS, Philippe SABATIER [email protected]

Objective and learning outcomes SaferNano Design& Law (Safer Design for Nanomaterials) is a Summer School which focuses on advanced methods and innovative approaches to NTs’ safety-by-design, in order to reduce the need of, and/or foster substitution of Critical Raw Materials (CRMs) in the main EU industrial Value Chains. The three main objectives are:

• Educate students to become highly skilled European professionals with expertise in NT’s EHS. This expertise will enable them to develop new methods for life cycle assessment and safer designing of nanomaterials. • Enable participants to become leading practicing engineers, across all sectors of society including academia, industry, and public service, with transferable skills such as innovation, ethics, intellectual property, sustainability, and advanced research strategies. • Develop a deep entrepreneurship mindset with the help and expertise of associated businesses, incubators, and innovation services as well as a large panel of industries.

Outcomes SaferNanoDesign & Law provides participants with a broad but high-level scientific background in the field, of nano safety and important skills such as teamwork, management of complex processes, conceptual approaches, entrepreneurship, and high intercultural awareness. Students are trained on how to get and analyse omics data to perform gene ontology and pathway analysis. They also become familiar with predictive toxicology via the Adverse Outcome Pathways (AOP) and Effectopedia tool. By taking this course participants will have gained:

• Broad view of the nanotechnology market and the evolving regulatory framework, • Knowledge on theoretical and practical understanding of nanomaterial reactivity and transformation in the environment; and on the surface reactivity and on the ‘nano-specific’ properties useful for diverse applications.

• Knowledge on how to assess environmental impacts of nanomaterials using a life cycle assessment model, and to develop nanomaterials and nanoproducts using a safer by design approach. • Insight on the different types of assays available to assess the impact of nanomaterial exposure at different levels (environment, organism, cell, molecule etc. • Overview of nowadays and future nanotoxicology: the different types of assays available to assess the impact of nanomaterials exposure at the organism but also cellular and molecular levels. Finally, the students will have gained knowledge on how to assess the biological response to nanomaterial exposure. • Mastery of the general legislation concerning eco-design at EU and national levels, as well as working knowledge of value chain issues and marketing. It also includes the capacity to analyse in specific contexts how innovative strategies may lead to improved firm performance or to new business perspectives. Content Nanotechnology is now bringing new opportunities to reduce the need of, and/or foster sub-situation of Critical Raw Materials in the main EU Industrial Value Chains. SaferNano Design & Law address the tricky challenge of the Nanotechnology’s transition by promoting ‘safety-by-design’ that minimizes the risks associated to environment and population health. By working on case studies, participants learn the main computing tools and databases for addressing the life cycle of the products. Additionally, they are introduced to Creative Thinking & 1st Session Business Creation and invited to pitch their ideas in front of a business panel. The program is organized in four sessions: 1st session: Nanomaterials and their life- cycle analysis; 2nd Session: Nanomaterials transformation in the environment; Ecosystem and Human Exposure; 3rd Session: Human toxicity; 4th Session: Innovation, Technology transfer and Business development. The school combines an intensive program of lectures, hands-on sessions (experiments, simulation, and modelling) and group working. Courses are given by teachers from France, Spain, United States, Sweden, Germany, the Netherlands, Great-Britain, and Switzerland. Study track at the UL

OLOs – Checkpoints

1. Making value judgement and sustainability competencies 2. Entrepreneurship skills and competencies 3. Creativity skills and competencies 4. Innovation skills and competencies 5. Research skills and competencies 6. Intellectual transforming skills and competencies 7. Leadership skills and competencies

Modules 1 2 3 4 5 6 7

Network Science X X

Data Science X X

Data Analysis and Integration X X

Information Visualization X X

Business Process Management X X X X X

Advanced Algorithms X X

Natural Language X X

Information Processing and X X Retrieval

Machine Learning X X

Parallel and Distributed X X X Computing

Cloud Computing and X X X Virtualization

Anatomy and Histology X X

Bioinformatics X X

Health ICT X X X X X X

Multi-Omic Data Analysis X X X X

Medical Imaging X X X X

Neuroengineering X X X

Signals and Systems in X X Bioengineering

Information Systems Project X X X X Management

User Centered Design X X X X X X

Entrepreneurship, Innovation and X X X X X X Technology Transfer

Master Project in Information and X X X X X X Software Engineering

Project in Biomedical Engineering X X X X X X

Master Thesis X X X X X X

Common Core (20ECTS):

• (sem1) Network Science (7.5 ECTS). • (sem1) Data Science (7.5 ECTS). • (sem1) Data Analysis and Integration (7.5 ECTS). • (sem1) Information Visualization (7.5 ECTS)

Master‘s electives (20 ECTS):

• (sem 1) Business Process Management (7.5 ECTS), • (sem 1) Natural Language (7.5 ECTS), • (sem 2) Advanced Algorithms (7.5 ECTS), • (sem 2) Machine Learning (7.5 ECTS), • (sem 2) Cloud Computing and Virtualization (7.5 ECTS) • (sem 2) Parallel and Distributed Computing (7.5 ECTS),

HMDA specialization (20 ECTS):

• (sem1) Bioinformatics (7.5 ECTS). • (sem 2) Health ICT (7.5 ECTS). • (sem 2) Multi-Omic Data Analysis (3 ECTS) (for incoming students only). • Medical Imaging (6 ECTS). • Neuroengineering (6 ECTS). • Signals and Systems in Bioengineering (6 ECTS).

I & E 1 (10 ECTS):

• (sem 1)Information Systems Project Management ( 7.5 ECTS), • (sem 1) User Centered Design (7.5 ECTS), • Entrepreneurship, Innovation and Technology Transfer (6 ECTS)

I & E 2 (Medical Specialization, 10 ECTS):

• (sem1, sem2) Master Project in Information and Software Engineering (12 ECTS). • (sem 1) Project in Biomedical Engineering (6 ECTS) (for incoming students only)

I & E 3 (10 ECTS):

• Information Systems Project Management (7.5 ECTS), • User Centered Design (7.5 ECTS). • Entrepreneurship, Innovation and Technology Transfer (6 ECTS)

1st year, 1st semester (autumn/winter semester):

Innovation and Entrepreneurship 1 (10 ECTS) INFORMATION SYSTEMS PROJECT MANAGEMENT Responsible lecturer: Rosário Bernardo ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: written test (50%), project (40%), research (10%)

Contents: The syllabus of GPI is closely aligned with the course "IS 2010.4 IS Project Management" defined in the "ACM / AIS IS2010 Curriculum Guidelines" having the following main topics:

1. Project management related concepts in an organizational perspective 2. Project basics 3. Life Cycles models 4. Scope management 5. Time management 6. Project Organization and Communication management 7. Stakeholders management 8. Cost management 9. Quality management 10. Risk management 11. Procurement management 12. Project control 13. Project closure 14. Project Management competence elements 15. Projects alignment with the Organization and the Business.

Learning outcomes and competencies: The objectives of GPI are aligned with the same objectives as defined for the course “IS 2010.4 IS”

1. Project Management” of the curriculum “ACM/AIS IS 2010 Curriculum Guidelines”, namely: Understand the concepts of project and project management in the organizational context. 2. Understand the project management process groups. 3. Understand and properly relate the project management processes with the different project’s development lifecycles approaches. 4. Make use of project scope planning methods and techniques. 5. Make use of project scheduling methods and techniques. 6. Identify the project stakeholders, make use of project organization and responsibilities planning methods and techniques and develop the project communication planning. 7. Identify the main cost components and be capable to use cost planning methods and techniques to define the project budget. 8. Make use of quality planning, quality assurance and quality control in the project management context. 9. Make use of risk identification, assessment, treatment and control methods and techniques. 10. Understand the procurement management processes and the management of different project contract types. 11. Make use of information and tools to support project control, project close and suitable metrics. 12. Identify the main Project Manager technical, behavioral, and contextual competence elements. 13. Understand the concepts of project-based organization, change management, project value, programme management, portfolio management and governance of projects. 14. Make adequate use of MS-Project functionalities on practice exercises.

Literature:

• Rethinking Project Management. An organizational perspective: Erling S. Andersen 2008 Pearson Education, UK • Managing Information Technology Projects, Revised 6th Edition (International Edition):Kathy Schwalbe2011Cengage Learning • Project Management for Information Systems -Fifth Edition :Cadle, James & Yeates, Donald2008Pearson Education, UK

Keywords: Project Planning, Innovation, Entrepreneurship, Sustainability.

USER CENTERED DESIGN Responsible lecturer: Nuno Jardim Nunes ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: project (80%), laboratories (20%)

Contents: Lectures:

• Introduction to User Centered Design. • Users and Stakeholders. Inquiring Users and Experts. • Observing Users. • User Involvement and Participation. • User Needs and Requirements. • Usability Engineering. Data Analysis and Interpretation. • Building Prototypes. • Interface Types. • Affective Aspects. Accessibility. • Ethics in User Centered Design.

Laboratory: Workshops with the following themes:

• Who are the Users? • What the Users want? • Applying Cultural Probes • Workshops with Users • Initial Requirements • Validation of Requirements • Conceptual Model and First • Low Fidelity Prototypes. • Usability Testing. • Low Fidelity Prototypes. • Functional Prototype.

Learning outcomes and competencies:

• Understand the basic principles and the methodologies of interactive systems user centred design. • Understand users and their needs, how to really acquire them, and the need of user involvement in interactive systems design and implementation. Adapt the above knowledge to user centred design methodologies. • Design and implement an interactive system involving real users at various levels in light of the above.

Literature:

• Interaction Design: Beyond Human-Computer Interaction (3rd Edition):J. Preece, Y. Rogers, H. Sharp2011John Wiley & Sons • Software for use: a practical guide to the models and methods of usage-centered design: Larry L. Constantine Lucy A. D. Lockwood 1999Larry L. Constantine and Lucy A. D. Lockwood. 1999. Software for Use: A Practical Guide to the Models and Methods of Usage-Centered Design. ACM Press/Addison-Wesley Publ. Co., New York, NY, USA.

Keywords: User-Centered Design. Usability Engineering. Human-Computer interaction.

ENTREPRENEURSHIP,INNOVATION AND TECHNOLOGY TRANSFER Responsible lecturer: José Epifânio da Franca ECTC: 6 Course type and weekly hours: lecture (3h) + laboratory (1.5 h) Exam type: A team project, consisting of a business plan (65%) and the development of a product in terms of engineering, marketing, and manufacturing (35%).

Contents:

1. Innovation, entrepreneurship and competitivity. 2. Innovation factors and processes. 3. Technology transfer and intellectual property. 4. Connections between technology, products and services, and the market. 5. The product development process. 5.1. Creativity and product planning. 5.2. Customer needs and product specifications. 5.3. Concept generation, selection and testing. 5.4. Product architecture. 5.5. Industrial Design, engineering, and prototyping. 6. Risk management. Identification and evaluation of risk factors and contingency plans. 7. Design for manufacturing and assembly. Design for the environment. 8. Design for cost, Target price / Target cost. 9. Economic analysis and sources of financing. 10. Legal aspects associated to enterprise creation. 11. Business plan. 11.1. Marketing Plan. 11.2. Production Plan / operations 11.3. Management Plan and enterprise organization. 11.4. Financial Plan. 12. Design discussions and meetings / business plan discussions.

Learning outcomes and competencies: To develop the necessary skills of the business entrepreneur to generate and evaluate innovative ideas, to develop and materialize innovation in products and services, and to structure a business plan to incubate and explore technology-based innovation, with a specific knowledge of market mechanisms, finance, and management.

Literature:

• Proactive risk management: Controlling uncertainty in product development: Preston G. Smith, Guy M. Merritt2002Productivity Press • Creating Breakthrough Products: J. Cagan & C. Vogel2002Prentice Hall, ISBN 0-13-969694-6 • Product Design & Development, 3rd Ed:K. T. Ulrich, S. D. Eppinger2003McGraw-Hill, ISBN 0071232737

Keywords: Entrepreneurship, Innovation.

HMDA Common Core (20 ECTS) NETWORK SCIENCE Responsible lecturer: Francisco Correia dos Santos ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: project developed by group of 2-3 students (50%), exam (50%)

Contents: This course provides an introduction to the study of complex networks, including algorithms, models, and applications to both artificial and real networks, including social, biological, and technological networks, all sharing common features and properties. The course addresses the development of scalable algorithms and data structures so that we can efficiently study large complex networks, but also in the creation of theoretical models capable of describing empirically observed patterns. The number of applications is enormous, including web search engines, evolutionary dynamics, information diffusion on Internet, social networks and blogs, network resilience, network-driven phenomena in epidemiology and computer viruses, networks dynamics, with connections in the social sciences, physics, computational biology, and economics.

Learning outcomes and competencies: Introduction to complex systems and networks science: Theory and basic concepts. Properties and characterization of biological, social, and technological networks. Network models and random graphs. Efficient representation of large (sparse) networks. Succinct data-structures and coding strategies. Design and analysis of efficient and scalable algorithms for large network processing and analysis, including both sampling and randomization techniques. Databases and distributed platforms for the analysis of large networks. Link analysis and random walks. Community finding and graph partitioning. Ranking algorithms. Vertex relabeling. Dynamical processes on complex networks: The impact of network structure on economic, social, and biological systems. Introduction to stochastic processes, Monte-Carlo simulations, and large-scale multi-agent systems. Disease spreading and tolerance to attacks. Models of peer-influence and opinion formation. Game theory and population dynamics. Public goods problems, cooperation, and reputation dynamics. Decision making on (static and adaptive) interaction networks.

Literature:

• Algorithms on strings Trees, and Sequences: Dan Gusfield1997Cambridge press • Networks, Crowds, and Markets: Reasoning about a Highly Connected World: Easley, D. and Kleinberg, J.2010Cambridge University Press • Networks: An Introduction: M. E. J. Newmann2010Oxford University Press • Network Science:Barabási, A.-L.2016Cambridge University Press • Lectures on Complex Networks:Dorogovtsev, S.N.2010Oxford University Press • Graph Theory in the information age: F. Chung.2010Notices of AMS, 57(6):726-732 • Mining of Massive Datasets: J. Leskovec, A. Rajaraman, J. D. Ullman2011Cambridge Univ. Press

• Dynamical processes on complex networks: Barrat, A., M. Barthelemy, and A. Vespignani2008Cambridge University Press

Keywords: Network Science, Complex Networks

DATA SCIENCE Responsible lecturer: Cláudia Antunes ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: project (50%), exam (50%)

Contents:

1. Data Science. What is data science? The multidisciplinary nature. Data engineering vs. Data science. The role of a data scientist. 2. Knowledge Discovery Process. Formulating questions. Exploratory data analysis. Pre-processing overview. Evaluation overview -Occam’s razor. Information Visualization overview. Documenting the process: Notebooks. 3. Pre-processing. Data scaling and centering. Data reduction: PCA, SVD, DFT, wavelets, SAX. Data balancing: resampling and SMOTE. Data discretization: equal-width, equal-frequency, taxonomies. Labelling. 4. Pattern Mining. Association rules -apriori algorithm. Closed vs Maximal patterns. Evaluation metrics: support, confidence, lift and Jaccard. 5. Clustering. Algorithms: K-means, hierarchical. Evaluation: SSE (MSE), silhouette coefficient, Dunn, and DB indexes. 6. Classification and Regression. Supervised learning: overfitting, training strategies, cross-validation. Linear and logistic regression. Classification Algorithms: KNN, Naive Bayes, Decision trees: metrics and pruning. Ensembles: AdaBoost, Random forests. Evaluation: Metrics (Accuracy, sensibility and specificity, f-measure, ROC area, confusion matrix); ROC and Lift charts 7. Outliers detection. 8. Privacy-preserving data mining. 9. Large-scale data mining. Parallelization: map-reduce, online algorithms. Indexing: LSH, Multidimensional. 10. Case Studies / Advanced Topics (9h) Time series and sequential analysis. Social Networks analysis; Mining graphs. Recommender Systems, Computational Advertising. Text and opinion mining. Process Mining. Stream Processing and Mining. Computational biology.

Learning outcomes and competencies: Students should be able to:

• Understand the statistics and data processing concepts used in complex information processes. • Design systems for knowledge discovery processes automation, and communication of their outcomes using the appropriate algorithms and validation methods at each stage. • Understand the techniques for frequent patterns recognition and outlier detection in data sets.

• Identify sensitive data that might be subject to processing restrictions and data anonymization techniques that enable privacy-preserving data mining, • Address large-scale data processing challenges.

Literature:

• Data Mining and Analysis: Fundamental Concepts and Algorithms: Mohammed J. Zaki, Wagner Meira, Jr. 2014 Cambridge University Press

Keywords: Data Science, Data Mining, Data Analytics.

DATA ANALYSIS AND INTEGRATION Responsible lecturer: Helena Galhardas ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: exam (55%), 3 small projects(45%)

Contents:

1. Main challenges of data integration processes; data integration paradigms. Heterogeneous data sources: XML data management and processing. 2. Heterogeneous data sources: (sensor) data stream management and processing. Virtual data integration: wrappers and mediators; query expression manipulation. 3. Query answering using views; source descriptions. 4. Schema mapping languages: global-as-view and local-as-view; schema mapping and matching. 5. Wrappers: manual and automatic construction. 6. Data warehousing: multi-dimensional modeling and data warehouse conception. 7. OLAP (Online-Analytical Processing) and ETL (Extraction-Transformation-Loading). 8. Caching and partial materialization; reporting. Data Exchange: declarative warehousing. 9. Data cleaning: taxonomy of data quality problems; data quality dimensions. 10. Approximate duplicate detection: string and data matching algorithms. 11. Data fusion. Mashups: motivation, creation, and application. 12. Data Provenance and Applications

Learning outcomes and competencies: The course on Data Analysis and Integration aims at teaching the students the most important concept of data integration according to two different perspectives: virtual data integration, where the data sources can be accessed through a mediator-based architecture; and materialized data integration, where a materialized data repository (named data warehouse) is populated with data coming from the data sources. Additionally, the course will teach techniques that can be used to exploit information: OLAP (On-line Analytical Processing) and reporting in a warehoused architecture, and mash-up systems in a virtual architecture. The data integration processes aim at supplying, among other applications, a uniform view over a set of autonomous and heterogeneous data sources, making it easy the access to source data for analysis and visualization purposes. Their application domains are diverse, ranging from the Business Intelligence systems to scientific research systems (e.g., Bioinformatics).

Literature:

• Principles of Data Integration: Anhai Doan, Alon Halevy and Zachary Ives.2012 Morgan Kaufmann.

Keywords: Information Integration, Data Warehousing, OLAP, Data Quality.

INFORMATION VISUALIZATION Responsible lecturer: Daniel Gonçalves ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: exam (35%), labs (25%), project (40%)

Contents:

1. Introduction 2. Design Methodology 3. Datasets and variables 4. Human Factors in InfoVis 5. Visualization Types 6. Visualization Techniques 7. Dynamic visualizations and animations 8. Item and Attribute reduction 9. Legibility and fidelity of visualizations 10. Evaluation of InfoVis Solutions 11. Applications

Learning outcomes and competencies: The main goal is to provide students with knowledge in the area of Information Visualization, that allows them to design and develop high-impact visualizations of data and information, to effectively transmit qualitative and quantitative data. The area of Information Visualization will be introduced, after which we will teach a methodology for analyzing problem domains and conceiving effective visualizations. Afterwards, we will discuss the different kinds of variables (continuous, nominal, ratio, etc.), data (tabular, networks, text, etc.) and patterns to visualize. Next, we will describe the different relevant physiological and psychological factors (memory, visual processing, etc.) relevant for the creation of good visualizations. We will study the most common kinds of visualizations adequate for different information types (graphs, time series, etc.) and interaction techniques (focus+context, overview+detail, panning+zoom, brushing, etc.). Finally, we’ll address issues related with the evaluation of the effectiveness of InfoVis applications.

Literature:

• Visualization Design and Analysis: Abstractions, Principles, and Methods: Tamara Munzner 2014 AK Peters -(Draft version: http://www.cs.ubc.ca/~tmm/courses/533-11/book/)

• Interactive Data Visualization: Foundations, Techniques, and Applications, Second Edition: Matthew O. Ward, Georges Grinstein, Daniel Keim 2015 A K Peters/CRC Press ISBN 9781482257373

Keywords: Information Visualization.

1st year, 2nd semester (spring/summer semester):

Innovation and Entrepreneurship 2 (10 ECTS) –Medical specialization BIOINFORMATICS (ALTERNATIVE:COMPUTATIONAL BIOLOGY,6ECTS,FOR INCOMING) Responsible lecturer: Susana Vinga ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: 4 laboratory work (20%), 2 tests or 1 exam (70%), 1 seminar (10%)

Contents:

• Introduction, Molecular biology main concepts, Introduction to algorithms and complexity • Graphs and genetics • DNA sequence analysis • Pairwise alignment • Multiple Sequence alignment • Motif finding • NGS data, algorithms, and data structures • Probabilistic models • Gene expression data analysis • Data mining • Unsupervised Learning: Clustering and Biclustering • Molecular phylogenetics • Supervised Learning: Decision trees, Bayesian methods Integrative data analysis • Seminar

Learning outcomes and competencies: Bioinformatics aims at developing computational methods and algorithms to process biological data and uses mathematical and statistical modelling to generate testable hypotheses about biological entities and processes. The goal of this course is to introduce the basic techniques that support the most recent developments on this field. Additionally, it enables the development of the ability to critically assess research publications in this field. Practical assignments during the course aim at developing the student's ability to develop software for bioinformatics.

Literature:

• An Introduction to Bioinformatics Algorithms: N. C. Jones and P. Pevzner2005MIT Press

• Biological Sequence Analysis -Probabilistic models of proteins and nucleic acids :R. Durbin, S. Eddy, A. Krogh, G. Mitchison1998Cambridge • Data Mining: Practical Machine Learning Tools and Techniques :Ian H. Witten, Eibe Frank, Mark A. Hall2011http://www.cs.waikato.ac.nz/ml/weka/book.html • Bioinformatics and Biomarker Discovery: "Omic" Data Analysis for Personalized Medicine: Francisco Azuage2010Wiley Blackwell

Keywords: Bioinformatics Algorithms.

MEDICAL IMAGING Responsible lecturer: Patrícia Figueiredo ECTC: 6 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: Two tests or Final exam –70% + Lab work –30%

Contents:

1. Introduction 1.1. Historical perspective 1.2. General imaging principles 2. X ray imaging 2.1. X rays 2.2. Planar radiography 2.3. Computed Tomography (CT) 2.4. Image reconstruction 2.5. Specialized imaging techniques 3. Nuclear medicine imaging 3.1. Radionuclides 3.2. Scintigraphy 3.3. Single Photon Emission Computed Tomography (SPECT) 3.4. Positron Emission Tomography (PET) 3.5. Corrections and image reconstruction 4. Magnetic Resonance Imaging (MRI) 4.1. Nuclear Magnetic Resonance (NMR) 4.2. Image formation and reconstruction 4.3. Instrumentation 4.4. Contrast mechanisms 4.5. Imaging sequences 4.6. Rapid imaging 4.7. Specialized imaging techniques 5. Ultrasound imaging 5.1. Ultrasounds 5.2. Transducers 5.3. Imaging modes 5.4. Doppler ultrasonography.

Learning outcomes and competencies: The goal of this course is to provide both a theoretical and a practical background in biomedical imaging techniques, focusing on the main modalities and covering physical principles of image acquisition; basic instrumentation; image reconstruction and analysis methods; and applications in disease diagnosis and monitoring. By the end of the semester, the student should be familiar with 1) the physical principles and basic instrumentation used for the acquisition of the main biomedical imaging techniques; 2) the most important image reconstruction and analysis methods; and 3) the main applications in disease diagnosis and monitoring.

Literature:

• Introduction to Biomedical Imaging: Andrew Webb 2003 Wiley ISBN: 0-471-23766-2.

Keywords: X-Ray, Nuclear Imaging, MRI, Ultrasound.

Master ś Electives (20 ECTS) BUSINESS PROCESS MANAGEMENT Responsible lecturer: Pedro Sousa ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: 1 exam (60%), group project (40%)

Contents:

1. Introduction to Business Process Management 2. Process Identification 3. Process Modeling 4. Process Discovery 5. Process Conformance 6. Process Analysis 7. Process Redesign 8. Process Automation

Learning outcomes and competencies: This course provides an engineering perspective on the fundamental concepts, techniques and tools associated with the business process management life cycle. The topics addressed in this course focus on the identification, documentation, modelling, validation, and verification, and optimization of organizational business processes using process analysis, design, and automation techniques. The learning objectives are as follows:

1. Understand the role of business processes within and between organizations. 2. Understand the relationships and dependencies between processes, enterprise architecture and the application and technological infrastructure. 3. Analyse and design business processes using business process modelling languages.

4. Analyse business processes using manual, semi-automated and automated techniques, including architectural principles and process mining. 5. Redesign and optimize business processes while keeping the traceability to the transformation requirements. 6. Understand the role of business process management systems (BPMS). 7. Understand the role of BPM tools, especially modelling and analysis tools.

Literature:

• Fundamentals of Business Process Management: Marlon Dumas, Marcello La Rosa, Jan Mendling, Hajo A. Reijers. 2013 Springer.

Keywords: Business Process Management. Business Process Modeling and Analysis. Process Mining.

NATURAL LANGUAGE Responsible lecturer: Luísa Coheur ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: 4 tests or 1 exam (60%), 2 projects (each 10%), 12 exercises (20%)

Contents:

1. Course overview. (1h) 2. Introduction to Natural Language Processing. (3.5h) 2.1. Basic concepts 2.2. Ambiguity and linguistic variability 2.3. Associated knowledge 2.4. Methodology 2.5. Train/test corpus, Cross validation, Measures (precision, recall, etc.) 3. Regular expressions and automata (1.5h) 4. N-Grams (4.5 h) 4.1. N-grams as language models 4.2. Markov assumption and probabilities of an N-gram/sentence 4.3. Smoothing techniques 5. Morphology (9h) 5.1. Morphology and transducers 5.2. Part of speech tagging (POS) 5.3. Rule-based and stochastic HMMS and Viterbi algorithm 6. Syntax (9h) 6.1. Grammars 6.2. Context-free grammars 6.3. Dependency grammars ́ 6.4. Probabilistic grammars 6.5. Syntactic analysis 6.6. Unification-based Top-down and Bottom-up 6.7. Chat-parsers (Earley e CKY)

6.8. Probabilistic 7. Semantic (9h) 7.1. Meaning representation 7.2. Lexical semantics 7.3. Thematic roles 7.4. Semantic disambiguation 7.5. Semantic analysis 7.6. Compositional semantic analysis 7.7. Statistic-based semantic analysis 7.8. Classifiers and their application in semantic analysis 7.9. Applications (remaining classes) 7.9.1. Information extraction (named entity recognition, etc.) 7.10. Text classification 7.11. Question/answering systems. 7.12. Dialogue systems 7.13. Machine translation 7.14. Speech recognition

Learning outcomes and competencies:

• Learn the basic concepts, main formalisms, techniques and algorithms, knowledge bases and corpora, used in the Natural Language Processing area. • Understand the main tasks involved in the processing of a sentence, paragraph or text and understand the main challenges of each one of these tasks. • Learn the main applications and be able to identify the associated technology. • Understand which are the tasks that can be done considering the current state of the art.

Literature:

• Speech and Language Processing (3rd ed. draft):Dan Jurafsky & James H. Martin2017https://web.stanford.edu/~jurafsky/slp3/ • Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition, Second Edition: Daniel Jurafsky & James H. Martin2009Prentice-Hall

Keywords: Natural Language Processing, Text Mining.

NEUROENGINEERING Responsible lecturer: Patrícia Figueiredo ECTC: 6 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: Exam (70%): covering all the modules’ topics + Student presentation (30%): paper or essay regarding one of the courses topics.

Contents:

1. Current challenges for Neuroengineering 2. Neuroscience basics 2.1. Neural systems and behavior 2.2. Brain cells and circuitry 2.3. Neural communication, plasticity, and degeneration 2.4. Cognitive function and dysfunction 3. Computational neuroscience 3.1. Neural coding and neural networks 3.2. Computational cognitive neuroscience 4. Neuroimaging 4.1. EEG and MEG 4.2. Structural and functional MRI 4.3. PET 4.4. NIRS 5. Neural monitoring and diagnosis 5.1. Statistical inference and model-based classification methods 5.2. Emotion assessment and human identification 5.3. Longitudinal studies 6. Neural interfaces

6.1. Fundamentals of BCIs: neurophysiology, data acquisition and signal processing 6.2. Direct EEG Interfaces, VEP, P300 and ERD/ERS 6.3. Motor imagery and rehabilitation 6.4. Applications 7. Neural modulation 7.1. Neurofeedback using EEG and NIRS. 7.2. Neural stimulation: DBS, TDCS, TMS 7.3. Self-adaptive immersive neural stimulation 7.4. Clinical and performance enhancement applications 8. Nanotechnology for neural engineering 8.1. Nanoparticle engineering 8.2. Microsystems for neuroscience on a chip 9. Neural tissue engineering 9.1. Biomolecular-based strategies for neural regeneration 9.2. Cellular-based strategies for neural regeneration and disease modeling 9.3. Acellular biomaterial-based strategies for neural regeneration 9.4. Advanced tissue engineering strategies 10. Cognitive robotics 10.1. Sensorimotor coordination 10.2. Non-verbal communication 10.3. Tools for rehabilitation 11. Complex brain networks 11.1. Theory and basic concepts of complex networks 11.2. Properties, representation, processing, and analysis of large networks 11.3. Applications to brain networks.

Learning outcomes and competencies: The main objective is to provide comprehensive background knowledge of the most important areas in the field of Neuroengineering, including the existing challenges and the main concepts and techniques that can be used to address them. The course comprises a series of modules addressing specific topics, each organized by an expert in the field. Students successfully completing the course are expected to:

1. have basic knowledge about the organization, structure, function, and pathological modifications of neural systems. 2. have general knowledge about the principles, methodologies and applications of the main engineering techniques used to study and interact with neural systems, with the objectives of brain monitoring, diagnosing, modulating, repairing, enhancing, or interfacing with machines; and 3. be prepared to critically evaluate different problems and techniques in Neuroengineering.

Literature:

• Neural Engineering: Bin He20132nd Ed., (ISBN-13: 978-1461452263)

Keywords: Computational Neuroscience, Neuroengineering, Brain Networks.

NEUROIMAGING Responsible lecturer: Patrícia Figueiredo ECTC: 6 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: A. Oral presentation of an assigned scientific paper during the semester (30%); B. Quizzes on taught material throughout the semester (30%); C. Written essay and oral presentation on a chosen Neuroimaging topic (40%).

Contents:

1. Introduction a. Historical perspective b. Overview of the human brain (Functional Specialization, Main Divisions, Brain Systems) 2. Neurophysiology basics a. Biophysics of neuronal function b. Micro-electrophysiology c. Neuronal models: from single neurons to neuronal masses d. Brain metabolism and hemodynamics 3. Electrophysiology neuroimaging a. Electro-Encephalography (EEG) b. Magnetoencephalography (MEG) c. Spontaneous brain rhythms and evoked potentials, synchronization, and desynchronization d. Transcranial Magnetic Stimulation (TMS) 4. Structural neuroimaging and spectroscopy a. Overview of magnetic resonance imaging (MRI)

b. Structural MRI c. Diffusion Tensor Imaging (DTI): structural connectivity d. Magnetic resonance spectroscopy (MRS) 5. Hemodynamic neuroimaging a. Functional MRI (fMRI): Blood Oxygenation Level Dependent (BOLD) contrast b. Stimulus/task-related and resting-state activity; functional connectivity. c. Perfusion imaging: Arterial Spin Labeling (ASL) d. Radiotracer techniques (PET and SPECT) e. Diffusion optical imaging (DOI) f. Multimodal techniques: EEG-fMRI, PET-MRI

Learning outcomes and competencies: The course takes a multidisciplinary approach, in order to provide training in both the principles of Neuroimaging techniques, as well as in their application to the understanding of brain function. Students successfully completing the course are expected to:

1) have general background knowledge of the basic principles, methodologies, and applications of the most important Neuroimaging techniques; and 2) to be prepared to critically evaluate the applicability of, and the results provided by, Neuroimaging techniques for different problems in basic and clinical Neuroscience. Literature:

• Brain Mapping: The Methods: Arthur W. Toga (Editor), John C. Mazziotta (Editor)2002Academic Press; 2nd edition. ISBN-10: 0126925402.

Keywords: neurophysiology, electrophysiology, neuroimaging

SIGNALS AND SYSTEMS IN BIOENGINEERING Responsible lecturer: João Miguel Sanches ECTC: 6 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: Tests/Exams (70%) and Labs (30%)

Contents:

1. Introduction to signal and systems a. Analog and discrete signals b. Systems c. Typical biomedical sensors 2. Signal spaces a. Metric Spaces b. Norm function and inner product c. Finite dimension representation and manipulation of continuous spaces d. Interpolation 3. Transforms a. Z Transform

b. Fourier Transform of discrete signals. c. Fourier transform properties. d. Discrete and fast Fourier transforms, DFT and FFT e. Short Time Fourier Transform (STFT) f. Long sequence filtering. Overlap add and save. 4. Sampling a. Continuous and discrete sampling b. Aliasing c. Sampling of band-pass signals d. Canonical ADC and DAC topologies. Anti-aliasing and reconstruction filters 5. Systems a. Linear and Time Invariant (LTI) b. Convolution theorem and LTI eigen signals c. FIR and IIR systems/filters d. Magnitude and Phase response. Bode plots. e. Canonical topologies of discrete filters 6. Random signals a. Discrete time random signals b. Average and correlation sequences c. Response of LTI systems to random signals 7. Feedback and control a. Linear feedback systems. b. Feedback effects. Stability issues c. Root Locus analysis d. Bode diagram and Nyquist stability criterion. e. Gain and phase margins.

Learning outcomes and competencies: This course is intended to complement the basic mathematical theory of signals and systems and provide the basic concepts of feedback and control theory in the scope of Bioengineering. The course focuses on the practical aspects related to implementation and use of the fundamental concepts of signals and systems theory, but advanced topics will be also addressed such as spaces of signals, time- frequency analysis, canonical structures of digital filters and feedback and control theory. In this latter topic canonical structures of feedback will be addressed as well as some techniques for stability analysis and compensation.

Literature:

• Discrete-time signal processing: Alan V. Oppenheim and Ronald W. Schafer-Prentice-Hall • Understanding Digital Signal Processing (3rd Edition): Richard G. Lyons-Pearson

Keywords: Signals, Transforms, Sampling, Systems, Feedback, and control.

ADVANCED ALGORITHMS Responsible lecturer: Alexandre Francisco ECTC: 7.5

Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: exam (60%), assignments (40%)

Contents:

• Advanced data structures. B-trees. Binomial heaps, Fibonacci heaps, and relaxed heaps. • Approximation algorithms for NP-hard problems. • Probabilistic techniques, random algorithms, and game theory. • Algorithms with random choices. • Online and real-time algorithms. • Parallel algorithms and algorithms using external memory. • Approximation algorithms for polynomial problems, e.g., linear algorithms for MSTs. • Fast algorithms for minimum cuts. • Graph partitioning. • Approximate counting. • String algorithms and pattern matching. • Suffix trees and suffix arrays. • Tree algorithms, LCA. • Amortized Analysis. Learning outcomes and competencies: Data structures and algorithms are the basic building blocks of any computer system and they become even more relevant when such systems have to process huge volumes of data and/or have to meet real time processing requirements. The aim of this course is to provide advanced training in techniques for the development and implementation of efficient algorithms and applications, with particular focus on advanced data structures and algorithms for indexing and compression, and on randomization, sampling, and approximation schemes, taking into account real time processing requirements and distributed computing environments. This course will follow a problem-based learning approach where techniques and methods will be intuitively and constructively explored.

Literature:

• Algorithms on Strings, Trees, and Sequences: Dan Gusfield 1997 Cambridge University Press • Randomized Algorithms: Rajeev Motwani and Prabhakar Raghavan 2000 Cambridge University Press

Keywords: Algorithms.

MACHINE LEARNING Responsible lecturer: Manuel Cabido Lopes ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: 5 mini-projects, lab projects, individual examination.

Contents:

1. Introduction to Machine Learning

2. Background a. Probability and information theory b. Linear algebra c. Optimization 3. Introduction to supervised learning -Linear Methods a. Linear regression b. Logistic regression and perceptron 4. Fundamentals of learning theory a. The bias-variance tradeoff b. Overfitting and underfitting c. Regularization d. Model selection e. Statistical learning theory 5. Supervised learning -Non-parametric methods a. k-nearest neighbors b. Locally weighted regression 6. Supervised learning -Decision Trees and ensemble methods a. Decision trees

b. Regression trees c. Ensemble methods 7. Supervised learning -Bayesian methods a. Naive Bayes b. Bayesian linear regression c. Bayes nets 8. Supervised learning -Kernel methods a. Max-margin classifiers b. Kernel regression 9. Supervised learning -Artificial neural networks a. Multilayer perceptron b. Backpropagation c. Convolutional networks d. Recurrent networks e. Regularization 10. Unsupervised learning a. k-means b. Mixture models and Expectation-Maximization c. PCA and ICA d. Autoencoders 11. Applications a. Text classification b. Image classification

Learning outcomes and competencies: This course aims to provide a complete and up-to-date introduction to key concepts in machine learning. After completing the course, students should be able to:

• Understand the main challenges involved in machine learning. • Understand and correctly apply the steps needed to train and validate a model that is able both to explain a set of data and make predictions about unseen data. • Understand and correctly apply the more common machine learning algorithms, recognizing their corresponding domain of application.

Literature:

• Machine Learning and Pattern Recognition: C. Bishop2006Springer

Keywords: Machine Learning.

2nd year, 1st semester (autumn/winter semester)

Innovation and Entrepreneurship 3 (10 ECTS) THE HMDA Ś SCHOOL ON LEARNING FROM HEALTH DATA This summer school is offered and held by UGA for all HMDA students, which includes the preparation, execution, and documentation of a practical project, which is based on a real use case in the healthcare context. For more information, please visit page 25.

ANATOMY AND HISTOLOGY Responsible lecturer: Carlos Plancha ECTC: 6.0 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: quizzes (20%), seminar (20%), exam (60%)

Contents:

1. Major cell types and tissues Epithelium. Connective. Muscle. Nerve. 2. Stem cells and cell therapy Concept of stem cell, embryonic stem cells, in vitro models. Induced pluripotent cells. Clinical applications. 3. Musculoskeletal System Bone: ossification / Remodeling, Repair in traumatology. Joins and Movement. Prostheses. Skeletal muscle: contraction; kinetic evaluation. 4. Nervous system Brain / Spinal Cord / Spinal Nerves / Cranial Nerves / Autonomic nervous system. Neurosurgical procedures. 5. Blood and Immune System Cells and plasma; differential blood cell count. Lymphoid organs: Bone marrow / Thymus / Lymph nodes /Spleen / Tonsils; Flow cytometry and transplantation. 6. Digestive System Esophagus and Gastrointestinal tract, glands attached to the gut; endoscopy / biopsies. 7. Respiratory system Airways / Lung / Ventilation. Assisted ventilation. 8. Endocrine System Hypothalamus / Pineal / Thyroid / Parathyroid / Adrenal / Pancreas 9. Cardio-Vascular System Heart and circulatory system. Major arteries and veins. Valvular and vascular prostheses.

10. Urinary System Kidney and urinary tract. Dialysis. 11. Reproductive System (Male and Female) Testis and spermatogenesis. Ovary and follicle development. Infertility and Medically Assisted Reproductive Technology 12. Eye and Vision; Ear and Hearing

Learning outcomes and competencies: At the end of the curricular unit the student must know:

1. the basic structure of cells, tissues, and organs, being able to correlate them with their respective functions in the body. 2. the language used in these scientific areas of Medicine, that will greatly facilitate the future interactions with the different health professionals.

Literature:

• Introduction to the Human Body: The Essentials of Anatomy & Physiology, 9th Edition: Gerard J. Tortora, Bryan Derrickson 2011 John Wiley & Sons • Color Atlas of Cytology, Histology and microscopic Anatomy, 4th Edition: Wolfgang Kuehnel 2003 Thieme

Keywords: Human Anatomy and Physiology.

INFORMATION PROCESSING AND RETRIEVAL Responsible lecturer: Bruno Martins ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: final exam (55%), 3 mini-projects(45%)

Contents:

1. Introduction to Information Retrieval and Information Extraction a. IR system architecture b. Document pre-processing 2. Non-structured data models a. Boolean model b. Vector-space model c. Dimensionality reduction d. Probabilistic models 3. Information Extraction from text a. Classification and clustering of documents b. The naive Bayes classifier c. Information Extraction with hidden Markov models 4. Evaluation of IR Systems a. Evaluation metrics b. Reference collections c. Cross-validation and other issues

5. Semi-structured data models a. Semi-structured data models b. The Extensible Markup Language (XML) c. Markup languages based on XML (e.g., TEI, METS, MODS) d. Other languages (e.g., SGML, HTML e RDF) 6. Web Data Extraction a. Wrapper generation b. The XQuery language c. IR in XML documents 7. Link analysis a. Web models b. Basic concepts on graphs and link analysis c. Using links to rank documents. d. Web crawling

8. Indexing and querying non-structured information a. Regular expressions b. Inverted Indexes c. Query processing 9. Similarity search a. Document shingling and the Jaccard similarity measure. b. Similarity-preserving summaries of sets and min-hash c. Locality-sensitive hashing d. Applications in multimedia retrieval 10. Recommendation systems a. Context, personalization, and information filtering b. Content-based recommendations c. Collaborative filtering 11. Distributed processing for IR and IE a. Data partitioning b. Federated search and meta-search engines c. Map-Reduce processing 12. IE and IR applications a. Enterprise search and expert search b. Digital libraries c. Opinion mining d. Other applications

Learning outcomes and competencies: This course aims to provide the students with a complete and updated introduction to the key concepts and technologies used for data processing in the areas of Information Retrieval (IR),Information filtering (IF) and Information Extraction (IE). Students of this course will learn the basic theoretical concepts and acquire the practical skills needed to:

1. Design modern solution for processing, managing, and querying large volumes of information. 2. Classify and group automatically sets of resources (e.g., large sets of textual documents).

3. Design search and filtering mechanisms for large collections. 4. Design systems to extract information from text and/or the Web. 5. Evaluate empirically such systems.

Literature:

• Modern Information Retrieval, the concepts and technology behind search -2nd edition: Ricardo Baeza-Yates and Berthier Ribeiro-Neto2011Addison-Wesley Professional • Web Data Mining: Exploring Hyperlinks, Contents and Usage Data -2nd edition :Bing Liu2011Springer.

Keywords: Search, Information Extraction, Text mining, Opinion Mining, Recommendation Systems.

MASTER PROJECT IN INFORMATION AND SOFTWARE ENGINEERING Responsible lecturer: Luis Veiga ECTC: 12 Course type and weekly hours: individually supervised self-study Exam type: oral exam (100%)

Contents: /

Learning outcomes and competencies: Students develop individually a plan fora master dissertation project.

Literature: /

Keywords: Project Planning, Innovation, Entrepreneurship, Sustainability.

PROJECT IN BIOMEDICAL ENGINEERING (FOR INCOMING STUDENTS ONLY) Responsible lecturer: Mónica Oliveira ECTC: 6 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: Oral presentation of team projects addressing a challenge. The final document for evaluation will be the presentation slides, complemented by a summary of the solution found by the team.

Contents: In the beginning of the semester 4 problems/challenges, one in each of the areas that are the majors/profiles of the degree course, will be proposed. These problems will be described in detail in a document that will be made available to the students. The students will be organized into teams. Each team will study in-depth their chosen problem, gather information either in the literature or through interviews, will review the state-of-the-art, and will develop a proposal to solve the problem which will include both technical aspects and also market analysis, cost assessment, and social impact.

The work will be supported by the teaching staff and by external mentors with practical experience in the area of the proposed problems.

Learning outcomes and competencies: To develop the capabilities to integrate the knowledge and competencies acquired during prior coursework in the analysis and proposal to solve a practical problem in Biomedical Engineering.

Literature:/

Keywords: Project Planning, Innovation, Entrepreneurship, Sustainability.

HMDA specialization (20 ECTS) HEALTH ICT(OR BIOMEDICAL INFORMATICS FOR INCOMING Responsible lecturer: Mário Gaspar da Silva ECTC: 7.5(or 6 for incoming students) Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: 5 Biweekly assignments (50%) + written exam (90min.) 50%

Contents:

1. Information Technology in the life sciences 2. Clinical information systems 3. Acquisition processing and use of biomedical data. The Electronic Health Record. 4. Health Informatics data interchange standards. Thesauri and Ontologies. 5. Natural language processing and biomedical text mining. 6. Clinical Decision-support Systems. 7. Tele-monitoring 8. Tele-Health 9. Bioinformatics and Biomedical Research Infrastructures. 10. Information Search 11. Personalised medicine 12. Ethical Legal and Social Issues in IT in Health. 13. Public Health Informatics 14. IT for Healthy Living and Active Ageing. Consumer Health Informatics. 15. IT in user training and education of health professional

Learning outcomes and competencies: The general objective of the course is to provide the fundamental principles and concepts related to the use of information technology in health care. The students will acquire essential competencies and knowledge on the use of information technology in biomedical research and its crucial role in the provision of health care services.

Literature:

• Biomedical Informatics: Computer Applications in Health Care and Biomedicine: Edward H Shortliffe and James J. Cimino2014ISBN: 978-0-38728986-1 • Medical Informatics: Knowledge Management and Data Mining in Biomedicine:Hsinchun Chen, Sherrilynne S . Fuller, Carol Friedman, William Hersh (eds.)2005Springer. ISBN: 978-0387-2438 1-8

Keywords: Biomedical Informatics, Health ICT.

HANDS-ON EPIGENETICS:MULTI-OMIC DATA ANALYSIS (MOOC) Responsible lecturer: Prof. Rui Henriques ECTC: 3 Course type and weekly hours: MOOC (for incoming students) Exam type: written exam (30 min.)

Contents:

• 1: Introduction to epigenetics □ The genetic code □ Code and personalized medicine □ Genome-wide association studies □ Limitations of genome-centric studies □ The role of epigenetics □ The central dogma □ Multi-omic data collection □ The need for multi-omic data analysis □ Epigenetics and personalized medicine □ Epigenetics in our daily life □ Case studies on identical twins □ Case studies on ancestral influence • 2: Essential background on (biomedical) data analysis □ Sample omic datasets □ Data exploration □ Data preprocessing □ Clustering □ Biclustering and pattern mining □ Classification □ Regression • 3: Integrative multi-omics for personalized medicine □ Integrating multiple sources of omic data □ Essentials on heterogeneous data analysis □ The role of exposomics in personalized medicine □ Combining multi-omic and medical data for personalized medicine □ Unsupervised analysis of multi-omic data □ Enrichment analysis as the way of increasing current knowledge on epigenetics. □ Comprehensive study of epigenetics from integrative patterns of disease □ Supervised analysis of multi-omic data

□ Discovery of multi-source epigenetic markers for personalized medicine

Learning outcomes and competencies:

• structured view on epigenetics and its role in personalised medicine • be familiar with current findings, opportunities, and challenges in personalize medicine (along its prevention, diagnostic and treatment components) • understand the relevance of genomic, proteomic, metabolomic, clinomic and exposomic data in epigenetics. • master essentials of supervised and unsupervised data analysis • be able to analyze multiple sources of omic data and master principles on how to learn from heterogeneous multi-omic data.

Keywords: epigenetics, multi-omic data analysis, personalised medicine.

EPIGENETICS AND PERSONALIZED MEDICINE (MOOC, to be developed in the future) Responsible lecturer: Prof. Rui Henriques ECTC: 3 Course type and weekly hours: MOOC (for incoming students) Exam type: written exam (30 min.)

Contents:

• Omic markers for personalized medicine □ Description and prediction tasks from multi-omic data □ Specifying the problem at hands: data and learning requirements □ Discovery of putative regulatory modules from omic data □ Statistical significance versus biological significance □ Enrichment analysis as the way of increasing current knowledge on epigenetics. □ Expanding and assessing current knowledge repositories □ Learning omic markers for personalized diagnostics □ Learning omic markers for personalized prevention and treatment decisions • Epigenetics and personalized medicine: a new era □ Current findings on oncology: immunotherapy and beyond □ Current findings on neurodegenerative diseases □ Other evidence for disease prevention, diagnosis, and treatment □ Drug design from multi-omic findings2. State-of-the-art epigenetic therapies □ Structured overview of modules A–D2. Concluding remarks

Learning outcomes and competencies:

• formulate personalised medicine problems as clustering, pattern mining, Biclustering, classification, regression, and tasks over one or more sources of omic data. • master state-of-the-art findings of epigenetics and the corresponding cutting-edge therapeutics

• be able to learn integrative markers from multi-omic time series data and critically validate them (statistical significance, robustness to noise and overfitting risks) • formulate personalised medicine problems using advanced learning tasks. • understand current ways of expanding current knowledge on epigenetics from multi-omic data analysis.

Literature: /

Keywords: epigenetics, multi-omic data analysis, personalised medicine.

CLOUD COMPUTING AND VIRTUALIZATION Responsible lecturer: Luís Antunes Veiga ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: exam (40%), lab project (45%), paper presentation & feedback (15%)

Contents:

• Introduction to Virtualization and Cloud Computing, Infrastructure-as-a-Service, Platform-as-a- Service, Software-as-a-Service. • System-level virtualization: system VM architecture, CPU virtualization, OS core, memory, • I/O; hardware support for virtualization, case studies (VMWare, QEMU/KVM, Xen). • Cloud computing systems (Amazon EC2,OpenStack, XenCloud, OpenNebula); VM scheduling, migration, and replication; monitoring and scalability (CloudWatch, Autoscaling). • Process-level virtualization: Java VM specification and reference implementation, security model, code management and binary translation, just-in-time compilation, and optimization, • Garbage collection, case studies (Jikes RVM). Cloud computing platforms (Azure, Google App Engine); distributed virtual machines. • Monitoring and scalability (Azure Fabric Controller). • Data and Storage services: block storage, file storage, key-value stores (Dynamo, S3, Datastore), tabular storage (BigTable, Percolator). • Cloud computing scalability: Map-reduce, dataflows (Pig, Dryad, OOzie), streams (S4), applications, monitoring, elasticity, and optimization. • Cloud computing cross-cutting concerns: virtualization energy efficiency, dynamic provisioning, energy centered cloud design.

Learning outcomes and competencies:

• Attain an integrated perspective of cloud computing and virtualization, with combined approaches for the design of modern large scale and distributed computing systems, and with their underlying mechanisms and algorithms. • Understand a vertical approach to the various virtualization and cloud computing technologies, enhancing applications and services with improved flexibility, resource and economic efficiency, scalability, and adaptability. • To be able to develop reliable and scalable systems and applications, on cloud computing over current virtualization platforms and applications models.

• To be able to assess and evaluate solutions, given the alternatives and tradeoffs involved in the employment and management of virtualization infrastructure for cloud computing.

Literature:

• Virtual Machines: Versatile Platforms for Systems and Processes: James Smith and Ravi Nair2005Morgan Kaufmann • The Cloud at Your Service: Jothy Rosenberg and Arthur Mateos2010Manning Publications • Programming Amazon Web Services: James Murty2008O'Reilly Media • Hadoop: The Definitive Guide:Tom White2012O'Reilly Media • Cloud Computing -Theory and Practice: Dan C. Marinescu20139780124046276

Keywords: Cloud Computing.

PARALLEL AND DISTRIBUTED COMPUTING Responsible lecturer: José Monteiro ECTC: 7.5 Course type and weekly hours: lecture (3 h) + laboratory (1.5 h) Exam type: exam (40%), project (60%)

Contents:

• Parallel computing models: multiprocessors and multicomputer. Memory organization. • Communication complexity. Interconnection networks. Flynn’s taxonomy. • Programming message passing systems: MPI. • Programming shared memory systems: OpenMP, threads, race conditions, deadlock detection. Analysis and synthesis of parallel algorithms: problem partitioning; data organization; synchronization; balancing and scheduling. • Performance analysis for parallel algorithms. • Foundations of distributed computing and their applications to parallel algorithms. • Limits of parallel computing. • Analysis of parallel algorithms: sorting algorithms; numerical algorithms, matrix multiplication, solving systems of linear equations; algorithms on graphs; search and optimization algorithms.

Learning outcomes and competencies:

• Understanding the models, techniques, and programming methods for parallel algorithms. • Analyzing and designing parallel algorithms. Understanding the foundations of distributed computing.

Literature:

• Parallel Programming: Michael Quinn2003McGrawHill • Parallel Programming: Techniques and Applications Using Networked Workstations and Parallel Computers: Barry Wilkinson and Michael Allen2005Prentice Hall

Keywords: parallel programming, distributed computing.

2nd year, 2nd semester (spring/summer semester)

Master’s thesis in collaboration with industry partner or hospital (30 ECTS) Objective and learning outcomes:

• The master thesis in Biomedical Engineering is typically a research project or study, or an extended analysis of a topic of scientific or technological nature. The goal is for students to perform research and apply the knowledge acquired during their studies while at the same time developing skills like initiative, autonomy skills, decision, and organization. • The main learning output of this master thesis is the students’ ability to work on a BME program, and to translate research into applications –in cooperation with a non-academic business partner. This innovative approach is based on the mobility of students, exchanging experiences in different disciplines, and establishing a common high-quality standard in education and training. • At the end of the master thesis students work should reflect the EIT OLOs 1-6.

Content: The program is defined according to the supervisor orientation and to the type of theme and it is developed during one of the semesters of the last year of the MSc Course. Following the MSc procedures, the Education Committees has assessed the scientific quality and feasibility of the master thesis proposals. It is anticipated that the thesis will be relevant to the student's track and will address a question of importance in the student's field of expertise. Students are expected to design a research project, write a formal research protocol, perform the study described in it, and prepare a comprehensive scholarly scientific paper reporting the results. Optionally, part of the master project could be done at another institute or company outside Partner Universities, but in this case, it is always under supervision of a Partner University staff member. To achieve their research project, students are required to write a scientific paper under guidance of their research supervisors, and to give a presentation about the research performed. The scientific paper must be approved by the academic supervisor and be suitable for submission to an international, scientific journal. The thesis can take place at universities, research centers or companies. As mentioned above, the thesis requires an advisor from the Engineering side and a co-advisor from a non-academic partner.

www.eithealth.eu

Authors

Dr. Felix Schmutterer FAU

Prof. Pascal Mossuz UGA

Prof. Mário Gaspar da Silva UPM

Prof. Sergio Paraiso UPM

Prof. David Perez del Rey UPM

Daniel Garza Reyna FAU