DOCSLIB.ORG

Ds3-Datascience-Polytechniqu…

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Ds3-Datascience-Polytechniqu… Machine Learning for Personalised Healthcare: Advances and Challenges Danielle Belgrave Researcher Healthcare Machine Learning @DaniCMBelg My Research Patient 1 Patient 2 Patient 3 Patient 4 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (hours) Latent Variable Modelling Longitudinal Data Analysis Missing Data X Y Z Patient-Centric Approach Causality Multidisciplinary Top 8 Challenges: DS in Healthcare 1. Address some of the technical challenges facing the community of machine learning for doing impactful healthcare research 2. Presentation of current solutions 3. Steps for future research Challenge # 1: Estimating Treatment Effects Intervention Population is split into 2 Outcomes for both groups by random allocation groups are measured Patient Group Control = Cured = Still Diseased Supervised Learning Intervention Control p1 = proportion in the intervention group who are cured p2 proportion in the control group who are cured H0: p1 - p2 = 0 e vs H1: p1 - p2 ≠ 0 Mean: p1 - p2 푝 (1−푝 ) 푝 (1−푝 ) = Cured = Not cured Variance: 1 1 + 2 2 푛1 푛2 푝 − 푝 Assume well-labelled groups Test Statistic: Z = 1 2 1 1 푝(1−푝)( + ) Machine recognises a new example 푛1 푛2 Classification, regression Challenge #2: Heterogeneous Populations Patient Group Same diagnosis same prescription Understanding Heterogeneity Drug NOT toxic Drug toxic but and beneficial Patient Group NOT beneficial Same diagnosis same prescription Drug toxic but Drug NOT toxic and beneficial NOT beneficial Supervised Learning Unsupervised Learning is not Enough Not cured Cured Patient Group Assume well-labelled groups Infer patterns/ discover underlying Machine recognises a new example data structure from a dataset without reference to labelled outcomes Classification, regression Clustering, Latent variable modelling Endotype Discovery to Understand Heterogeneity To identify subgroups of complex disease risk Treatment outcome explained by distinctive underlying mechanism Foundation of Stratified Medicine Seeking better-targeted interventions Accounting for Heterogeneity Individualised Healthcare Outcomes • Probabilistic Graphical Models • Reinforcement Learning • Deep Learning Understanding Heterogeneity Through Probabilistic Graphical Modelling Identifying Heterogeneous Patient Groups Parsimonious description of the data inferred from what is observed Probabilistic Programming Probabilistic reasoning system The evidence contains The probabilistic model specific information expresses general about a situation Probabilistic knowledge about a model situation The inference algorithm uses the Evidence model to answer Inference queries given evidence Algorithm Queries The answers to queries are framed as probabilities of Answer different outcomes The queries express the things that will help The basic components of a probabilistic reasoning system you make a decision Adapted from Pfeffer, Avi. "Practical probabilistic programming." International Conference on Inductive Logic Programming. Springer Berlin Heidelberg, 2010. Eco-systems for Probabilistic Programming Infer.NET Edward Observed values (data, priors) Probabilistic Pyro program Stan Infer.NET Inference Engine Algorith Infer.NET C# Algorith C# m compiler compiler m execution Probability distributions Understanding Heterogeneity Motivating Probabilistic Graphical Examples Modelling Example 1: Heterogeneity in Asthma Asthma Phenotypes: Observable Manifestations of Disease Asthma Medication Exacerbations Allergy Poor Wheeze Asthma Lung Symptoms Function Don’t Grow out of Respond to Asthma Late Severity Respond to Asthma treatment in Childhood treatment Endotypes? Different Diseases With Different Causes Causal Mechanisms of Asthma and Allergy Manchester Longitudinal birth cohort ~2000 children Progression of allergy: Eczema -> Asthma -> Rhinitis Symptoms Causally Linked Prevention strategy: Bristol Target children with eczema to reduce progression to asthma Longitudinal birth cohort and rhinitis ~10000 children Hidden Markov Model: “Allergic March” Eczema Age 1 Age 3 Age 5 Age 8 Age 11 Class Eczema State Eczema State Eczema State Eczema State Eczema State Wheeze Age 1 Age 3 Age 5 Age 8 Age 11 Class Wheeze State Wheeze State Wheeze State Wheeze State Wheeze State Rhinitis Age 1 Age 3 Age 5 Age 8 Age 11 Class Rhinitis State Rhinitis State Rhinitis State Rhinitis State Rhinitis State Children (n=9801) t e r A h e s s t h m n c a y Latent Class Disease a a n d u d M A l l e r g y S t Profile Longitudinal Latent Disease Profile Latent State Latent State Latent State Latent State Latent State Age 1 Age 3 Age 5 Age 8 Age 11 Class = 1,….,k Eczema Eczema Eczema Eczema Eczema Age 1 Age 3 Age 5 Age 8 Age 11 Wheeze Wheeze Wheeze Wheeze Wheeze Age 1 Age 3 Age 5 Age 8 Age 11 Rhinitis Rhinitis Rhinitis Rhinitis Rhinitis Age 1 Age 3 Age 5 Age 8 Age 11 Children (n=9801) Latent Class Disease Profile Disaggregating Symptom Heterogeneity The “Allergic March” reflects patterns at the population level, rather than the natural covariance of symptoms within individuals’ life courses Developmental profiles of are heterogeneous Danielle CM Belgrave, Raquel Granell, Angela Simpson, John Guiver, Christopher Bishop, Iain Buchan, A. John Henderson, and Adnan Custovic. Developmental Profiles of Eczema, Wheeze, and Rhinitis: Two Population-Based Birth Cohort Studies. PlosMedicine 2014. Example 2: Heterogeneity in Complex Chronic Diseases Scleroderma Aim: To predict a function of time that models the future trajectory of a single target clinical marker tracking a disease process of interest. Heterogeneity in Lung Function Schulam, Peter, and Suchi Saria. "Integrative analysis using coupled latent variable models for individualizing prognoses." The Journal of Machine Learning Research 17.1 (2016): 8244-8278. Probabilistic Models for Individualised Care Schulam, Peter, and Suchi Saria. "Integrative analysis using coupled latent variable models for individualizing prognoses." The Journal of Machine Learning Research 17.1 (2016): 8244-8278. Reinforcement Learning in Motivating Healthcare Examples Individualised Treatment Effects Task of Optimising Sequential Decision-Making Unobserved responses Observed decisions and response Unobserved Mechanical responses Ventilation? Sedate? Vasopressors Gottesman, O., Johansson, F., Komorowski, M., Faisal, A., Sontag, D., Doshi-Velez, F., & Celi, L. A. (2019). Guidelines for reinforcement learning in healthcare. Nature medicine, 25(1), 16-18. Reinforcement Learning in Healthcare What action maximises the reward Action (A) - State (S) - Reward (R) - Policy (π) - Value (V) Expected long-term return of the current state sunder policy π Vπ(s) Q-value or action-value (Q) Long-term return of the current state s, taking action a under policy π Qπ(s, a) v(s) = E[Rt+1 + λv(St+1)| St = s] 2 Qπ(s, a) = E[rt+1 + λrt+2 + λ rt+3 | s, a] = E[r + λ Qπ(s’, a’) | s, a] Q*π(s, a) = E[r + λmax Q*(s’, a’) | s, a] Successful Applications of RL in Healthcare 1. RL applied to optimizing antiretroviral therapy in HIV Parbhoo, S., Bogojeska, J., Zazzi, M., Roth, V. & Doshi-Velez, F. AMIA Summits on Translational Science Proceedings 2017, 239 (2017). 2. RL applied to tailoring antiepilepsy drugs for seizure control Guez, A., Vincent, R. D., Avoli, M. & Pineau, J. Treatment of epilepsy via batch- mode reinforcement learning. In Proceedings of the Twenty-Tird AAAI Conference on Artifcial Intelligence 1671–1678 (AAAI, 2008). 3. RL applied to interventions in ICU Komorowski, M., Celi, L. A., Badawi, O., Gordon, A. & Faisal, A. Nat. Med. 24, 1716– 1720 (2018). Confounding factors: Estimated effects of medication may be diminished if severely ill or high-risk patients are given a particular drug Effective sample size larger if learned policies are close to the clinician policies – RL better for refining existing practices rather than discovering new treatment approaches If the rewards we are trying to maximise are too simplistic, will the system behave as expected in real life Downfalls of RL in Healthcare Deep Learning Exceptionally effective at learning patterns of data Utilizes learning algorithms that derive meaning out of data by using a hierarchy of multiple layers that mimic the neural networks of our brain If you provide a system with tons of information, it begins to understand it and respond in useful ways Challenge #3: Heterogeneous Data Types Data “In-the-Wild” The Richer the Data Types, the more we need other methods Meta-challenges to Estimating Treatment Effect Heterogeneous Generalisability Causal Discovery Patient Populations Data “In-the-Wild” Missing Data Interpretability Challenge #4: Causality Causal Reasoning The questions that motivate most studies in the health, social and behavioral sciences are not associational but causal in nature. Before an association is assessed for the possibility that it is causal, other explanations such as chance, bias and confounding have to be excluded Require some knowledge of the data-generating process - cannot be computed from the data alone, nor from distributions governing data Aim: to infer dynamics of beliefs under changing conditions, for example, changes induced by treatments or external interventions. Pearl, Judea. "Causal inference in statistics: An overview." Statistics surveys 3 (2009): 96-146. Efficacy and mechanism evaluation: Causal Framework for Personalising Health U Mediator Predictive biomarker (moderator) Outcomes Random Allocation Prognostic biomarker (risk factor) Example: Personalisation of Cancer Treatment U Tumor Size Genetic Marker Outcome Treatment (Survival) Prognostic biomarker
Load more

Recommended publications
  • A Study on Hierarchical Clustering for Identifying Asthma Endotypes (J4R/ Volume 03 / Issue 05 / 004)
    Journal for Research | Volume 03| Issue 05 | July 2017 ISSN: 2395-7549 A Study on Hierarchical Clustering for Identifying Asthma Endotypes S. Poorani Dr. P. Balasubramanie Assistant Professor Professor Department of Computer Technology Department of Computer Technology Kongu Engineering College, Perundurai, India Kongu Engineering College, Perundurai, India N. Sasipriyaa Assistant Professor Department of Computer Science and Engineering Kongu Engineering College, Perundurai, India Abstract Large multitudes of medical data are attained due to advancement in recent technology. These large volumes comprise valuable information which helps for diagnosing diseases. Data mining techniques can be used to extract useful patterns from these mass data. It provides a user- oriented approach to the novel and hidden patterns in the data. One of the major challenges in medical domain is the extraction of comprehensible knowledge from medical diagnosis data. Healthcare system needs an automated tool that identifies and disseminates relevant healthcare information. Asthma is one of the dominant disease all over the world. Identifying subtypes of the disease will help in the treatment and prevention approaches. In this paper a clustering model is proposed based on observable symptoms of asthma. The hierarchical clustering technique is used to group the samples based on important characteristics of asthma. Keywords: Data Mining, Hierarchical Clustering, Asthma, Health Care _______________________________________________________________________________________________________ I. INTRODUCTION There are 300 million asthmatics worldwide with 1/10th of those living in India. A recent review about epidemiological studies showed that the asthma prevalence among children was 7.24%[1]. Recently the Global Initiative for Asthma (GINA) defines asthma as a heterogeneous disease usually characterized by chronic airway inflammation.
    [Show full text]
  • Towards Using AI to Augment Human Support in Digital Mental Healthcare
    Towards Using AI to Augment Human Support in Digital Mental Healthcare Prerna Chikersal Abstract School of Computer Science To address the need for more access to, and increase Carnegie Mellon University, US the effectiveness of, mental health treatment, internet- [email protected] delivered psychotherapy programs such as iCBT are shown to achieve clinical outcomes comparable to face- Gavin Doherty to-face therapy. While offering ubiquitous access to School of Computer Science & Statistics healthcare, a key concern of digital therapy programs is Trinity College Dublin, IRL to sustain users’ engagement with the treatment to [email protected] attain desired benefits. Research has demonstrated that including a trained ‘human supporter’ to the digital Anja Thieme mental health ecosystem can provide useful guidance Healthcare Intelligence and motivation to its users, and lead to more effective Microsoft Research Cambridge, UK outcomes than unsupported interventions. Within this [email protected] context, we describe early research that makes use of machine learning (ML) approaches to better understand how the behaviors of these human supporters may benefit the mental health outcomes of clients; and how such effects could be maximized. We discuss new opportunities for augmenting human support through personalization, and the related ethical challenges. Author Keywords Mental health; digital behavioral intervention; ethics; responsible AI; machine learning; NLP; data mining. CSS Concepts •Human-centered computing → Human computer Submission
    [Show full text]
  • Atopic Dermatitis and Respiratory Allergy: What Is the Link
    Curr Derm Rep (2015) 4:221–227 DOI 10.1007/s13671-015-0121-6 ATOPIC DERMATITIS (C FLOHR, SECTION EDITOR) Atopic Dermatitis and Respiratory Allergy: What is the Link Danielle C. M. Belgrave1,2 & Angela Simpson1 & Iain E. Buchan2 & Adnan Custovic1 Published online: 28 September 2015 # Springer Science+Business Media New York 2015 Abstract Understanding the aetiology and progression of Introduction atopic dermatitis and respiratory allergy may elucidate early preventative and management strategies aimed towards reduc- Understanding the aetiology and progression of atopic derma- ing the global burden of asthma and allergic disease. In this titis and respiratory allergy may elucidate early preventative article, we review the current opinion concerning the link and management strategies aimed towards reducing the global between atopic dermatitis and the subsequent progression of burden of asthma and allergic disease [1, 2]. There is increas- respiratory allergies during childhood and into early adoles- ing interest in determining the causes of atopic dermatitis and cence. Advances in machine learning and statistical method- respiratory allergies, especially in the light of the increasing ology have facilitated the discovery of more refined defini- prevalence and economic burden associated with these condi- tions of phenotypes for identifying biomarkers. Understand- tions. Globally, up to 20 % of children suffer from atopic ing the role of atopic dermatitis in the development of respi- dermatitis [3]. According to the 2014 Global Asthma Report, ratory allergy may ultimately allow us to determine more ef- it is estimated that 334 million people have asthma with 14 % fective treatment strategies, thus reducing the patient and eco- of the world’s children and 8.6 % of young adults exhibiting nomic burden associated with these conditions.
    [Show full text]
  • Pediatric Pulmonology, on Pediatric Pulmonology, International Congress Th 15 ISSN 8755-6863 PEDIATRIC
    PEDIATRIC PULMONOLOGY PEDIATRIC VOLUME 51 • SUPPLEMENT 43 • JUNE 2016 PEDIATRIC PULMONOLOGY Volume 51 • Supplement 43 • June 2016 Proceedings S1 Foreword S2 Postgraduate Course on LFT S6 Keynote Speaker S8 I. Plenary Sessions S21 II. Topic Sessions S57 III. Young Investigator Oral Communications S60 IV. Posters PEDIATRIC PULMONOLOGY S90 Index Volume 51 Volume • Supplement 43 15th International Congress on Pediatric Pulmonology, Naples, Italy, June 23–26, 2016 • June 2016 June Pages S1–S92 Pages ISSN 8755-6863 ONLINEhttp://mc.manuscriptcentral.com/ppul SUBMISSION AND PEER REVIEW PEDIATRIC PULMONOLOGY Editor-in-Chief: THOMAS MURPHY, Watertown, MA USA Deputy Editor: TERRY NOAH, Chapel Hill, NC USA Associate Editors: ANNE CHANG, Brisbane, Queensland, Australia STEPHANIE DAVIS, Indianapolis, IN USA ALEXANDER MOELLER, Zurich, Switzerland KUNLING SHEN, Beijing, China PAUL STEWART, Chapel Hill, NC USA STEVEN TURNER, Aberdeen, Scotland, United Kingdom Topic Editors: JUDITH VOYNOW, Richmond, VA USA © 2016 Wiley Periodicals, Inc. All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior permission in writing from the copyright holder. Authorization to copy items for internal and personal use is granted by the HEATHER ZAR, Cape Town, South Africa copyright holder for libraries and other users registered with their local Reproduction Rights Organization (RRO), e.g. Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.com), provided the appropriate fee is paid directly to the RRO. This consent Managing Editor: CARLENE RUMMERY, Winnipeg, MB Canada does not extend to other kinds of copying such as copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale.
    [Show full text]
  • Neurips 2020 Workshop Book
    NeurIPS 2020 Workshop book Schedule Highlights Generated Fri Dec 18, 2020 Workshop organizers make last-minute changes to their schedule. Download this document again to get the lastest changes, or use the NeurIPS mobile application. Dec. 10, 2020 None Topological Data Analysis and Beyond Rieck, Chazal, Krishnaswamy, Kwitt, Natesan Ramamurthy, Umeda, Wolf Dec. 11, 2020 None Privacy Preserving Machine Learning - PriML and PPML Joint Edition Balle, Bell, Bellet, Chaudhuri, Gascon, Honkela, Koskela, Meehan, Ohrimenko, Park, Raykova, Smart, Wang, Weller None Tackling Climate Change with ML Dao, Sherwin, Donti, Kuntz, Kaack, Yusuf, Rolnick, Nakalembe, Monteleoni, Bengio None Meta-Learning Wang, Vanschoren, Grant, Schwarz, Visin, Clune, Calandra None OPT2020: Optimization for Machine Learning Paquette, Schmidt, Stich, Gu, Takac None Advances and Opportunities: Machine Learning for Education Garg, Heffernan, Meyers None Differential Geometry meets Deep Learning (DiffGeo4DL) Bose, Mathieu, Le Lan, Chami, Sala, De Sa, Nickel, Ré, Hamilton None Workshop on Dataset Curation and Security Baracaldo Angel, Bisk, Blum, Curry, Dickerson, Goldblum, Goldstein, Li, Schwarzschild None Machine Learning for Health (ML4H): Advancing Healthcare for All Hyland, Schmaltz, Onu, Nosakhare, Alsentzer, Chen, McDermott, Roy, Akera, Kiyasseh, Falck, Adams, Bica, Bear Don't Walk IV, Sarkar, Pfohl, Beam, Beaulieu-Jones, Belgrave, Naumann None Learning Meaningful Representations of Life (LMRL.org) Wood, Marks, Jones, Dieng, Aspuru-Guzik, Kundaje, Engelhardt, Liu, Boyden, Lindorff-Larsen,
    [Show full text]
  • Ascertaining Patterns of Asthma Symptoms in Childhood Using Machine Learning Methods
    Ascertaining Patterns of Asthma Symptoms in Childhood using Machine Learning Methods A thesis submitted to The University of Manchester for the degree of Doctor of Philosophy In the Faculty of Biology, Medicine and Health. 2019 Matea Deliu School of Health Sciences Division of Informatics, Imaging and Data Science Blank page 2 Table of Contents List of Tables ................................................................................................................... 6 List of Figures .................................................................................................................. 8 Abstract .......................................................................................................................... 9 Declaration .....................................................................................................................11 Copyright Statement.......................................................................................................11 Acknowledgements ........................................................................................................12 About the Author ...........................................................................................................13 1.1 Candidate Degrees ........................................................................................................... 13 1.2 Research Interests ............................................................................................................ 13 1.3 Publications ....................................................................................................................
    [Show full text]
  • Elsevier Editorial System(Tm) for the Lancet Respiratory Medicine Manuscript Draft
    Elsevier Editorial System(tm) for The Lancet Respiratory Medicine Manuscript Draft Manuscript Number: THELANCETRM-D-17-00681R2 Title: Lung function trajectories from pre-school age to adulthood and their associations with early life factors: a retrospective analysis of three population-based birth cohort studies. Article Type: Article (Original Research) Keywords: Respiratory function tests; Forced expiratory volume; Child development; Lung diseases, obstructive; Genetics Corresponding Author: Professor John Henderson, Corresponding Author's Institution: University of Bristol First Author: Danielle Belgrave Order of Authors: Danielle Belgrave; Raquel Granell; Steve w Turner; John A Curtin; Iain E Buchan; Peter N LeSouëf; Angela Simpson; John Henderson; Adnan Custovic Manuscript Region of Origin: UNITED KINGDOM Abstract: Background: Maximal lung function in early adulthood is an important determinant of mortality and COPD. We investigated whether there are distinct trajectories of lung function during childhood, and whether these extend to adulthood and infancy. Methods: To ascertain trajectories of forced expiratory volume in 1 second (FEV1), we studied two population-based birth cohorts with repeat spirometry from childhood into early adulthood (1046 and 1390 participants). To ascertain whether trajectories extend to early life, we used a third cohort with repeat lung function measures in infancy (V′maxFRC) and childhood (FEV1; n=196). We identified trajectories using latent profile modelling. We created an allele score to investigate genetic associations of trajectories, and constructed a multivariable model to identify their early-life predictors. Results: We identified four childhood FEV1 trajectories: Persistently High; Normal; Below Average; and Persistently Low. Persistently Low trajectory (~5% of participants) was associated with persistent wheezing and asthma throughout follow-up.
    [Show full text]
  • Challenges in Using Hierarchical Clustering to Identify Asthma Subtypes: Choosing the Variables and Variable Transformation Mate
    Challenges in using hierarchical clustering to identify asthma subtypes: choosing the variables and variable transformation Matea Deliu, University of Manchester, Health e-Research Centre, Manchester Tolga Yavuz, University of Manchester, Manchester Matthew Sperrin and Umit Sahiner, Imperial College London, London Danielle Belgrave and Cansin Sackesen Adnan Custovic Omer Kalayci, Haceteppe University, Ankara Introduction The use of unsupervised clustering has identified different subtypes of asthma. Choosing the variables to input into the clustering algorithm is one of the important considerations. The majority of previous studies selected variables based on expert advice, whilst others used dimension reduction techniques such as principal component analysis (PCA). We aimed to compare the results of unsupervised clustering when using raw variables, or variables transformed using dimensionality reduction techniques. Methods We recruited 613 asthmatics aged 6–23 years (Ankara, Turkey). We conducted extensive phenotyping, with 49 variables including demographic data, sensitization, lung function, medication, peripheral eosinophilia, and markers of asthma severity. We performed hierarchical clustering (HC) using: (1) all variables and (2) PCA-transformed variables. Results PCA revealed 5 components describing atopy and variations in asthma severity, which were then used to infer cluster assignment. The optimal HC solution in both PCA-transformed and raw untransformed data identified 5- clusters which were not identical. Both identified mild asthma with good lung function, severe atopic asthma and late- onset atopic mild asthma. Clustering without PCA identified early-onset severe atopic asthma and late-onset atopic asthma with high BMI, whilst early onset non-atopic mild asthma in females was identified in HC with PCA. However, cluster stability was poor.
    [Show full text]
  • Features of Asthma Which Provide Meaningful Insights for Understanding the Disease Heterogeneity
    Received: 20 May 2017 | Revised: 31 July 2017 | Accepted: 15 August 2017 DOI: 10.1111/cea.13014 ORIGINAL ARTICLE Epidemiology of Allergic Disease Features of asthma which provide meaningful insights for understanding the disease heterogeneity M. Deliu1 | T. S. Yavuz2,3 | M. Sperrin1 | D. Belgrave6 | U. M. Sahiner5 | C. Sackesen4,5 | O. Kalayci5 | A. Custovic6 1Division of Informatics, Imaging and Data Sciences, Faculty of Biology, Medicine and Summary Health, University of Manchester, Background: Data-driven methods such as hierarchical clustering (HC) and principal Manchester, UK component analysis (PCA) have been used to identify asthma subtypes, with incon- 2Department of Pediatric Allergy, Gulhane School of Medicine, Ankara, Turkey sistent results. 3Department of Paediatric Allergy, Objective: To develop a framework for the discovery of stable and clinically mean- Children’s Hospital, University of Bonn, Bonn, Germany ingful asthma subtypes. 4School of Medicine, Pediatric Allergy Unit, Methods: We performed HC in a rich data set from 613 asthmatic children, using Koc University, Istanbul, Turkey 45 clinical variables (Model 1), and after PCA dimensionality reduction (Model 2). 5Pediatric Allergy and Asthma Unit, Hacettepe University School of Medicine, Clinical experts then identified a set of asthma features/domains which informed Ankara, Turkey clusters in the two analyses. In Model 3, we reclustered the data using these 6 Department of Medicine, Section of features to ascertain whether this improved the discovery process. Paediatrics, Imperial College London, London, UK Results: Cluster stability was poor in Models 1 and 2. Clinical experts highlighted four asthma features/domains which differentiated the clusters in two models: age Correspondence Adnan Custovic, Department of Medicine, of onset, allergic sensitization, severity, and recent exacerbations.
    [Show full text]
  • Advances in Data Science 2018: Final Speakers & Discussion Themes
    Advances in Data Science 2018: Final Speakers & Discussion Themes Author, Gwern Hywel A Data Science Foundation Blog March 2018 --------------------------------------------------- www.datascience.foundation Data Science Foundation Data Science Foundation, Atlantic Business Centre, Atlantic Street, Altrincham, WA14 5NQ Tel: 0161 926 3670 Email:[email protected] Web: www.datascience.foundation Registered in England and Wales 4th June 2015, Registered Number 9624670 Copyright 2016 - 2017 Data Science Foundation Advances in Data Science Conference 2018 Speakers & Discussion Themes Date: Monday 21st May - Tuesday 22nd May 2018 Venue: Museum of Science & Industry, Liverpool Road, Manchester, M3 4FP The Data Science Institute invites you to join us at Advances in Data Science 2018, a two- day conference exploring data science's potential to support societal well-being. Jointly organised by The University of Manchester's Data Science Institute and the Cathie Marsh Institute for Social Research, Advances in Data Science 2018 is set to build upon the success of the 2017 conference in May 2017. With our keynote speakers confirmed, the international line-up comprises of world-leading academic and industry experts in data science and social sciences. Join us for this unique opportunity to learn about the latest global research in societal well-being, share knowledge and innovation, and expand your own network by connecting with world-class researchers, analysts and industry leaders. Key Applications Focusing on Gaussian processes, Deep
    [Show full text]
  • Developmental Profiles of Eczema, Wheeze, and Rhinitis: Two Population-Based Birth Cohort Studies
    Developmental Profiles of Eczema, Wheeze, and Rhinitis: Two Population-Based Birth Cohort Studies Danielle C. M. Belgrave1,2.*, Raquel Granell3., Angela Simpson1., John Guiver4., Christopher Bishop4., Iain Buchan2., A. John Henderson3., Adnan Custovic1. 1 Centre for Respiratory Medicine and Allergy, Institute of Inflammation and Repair, University of Manchester and University Hospital of South Manchester, Manchester, United Kingdom, 2 Centre for Health Informatics, Institute of Population Health, University of Manchester, Manchester, United Kingdom, 3 School of Social and Community Medicine, University of Bristol, Bristol, United Kingdom, 4 Microsoft Research Cambridge, Cambridge, United Kingdom Abstract Background: The term ‘‘atopic march’’ has been used to imply a natural progression of a cascade of symptoms from eczema to asthma and rhinitis through childhood. We hypothesize that this expression does not adequately describe the natural history of eczema, wheeze, and rhinitis during childhood. We propose that this paradigm arose from cross-sectional analyses of longitudinal studies, and may reflect a population pattern that may not predominate at the individual level. Methods and Findings: Data from 9,801 children in two population-based birth cohorts were used to determine individual profiles of eczema, wheeze, and rhinitis and whether the manifestations of these symptoms followed an atopic march pattern. Children were assessed at ages 1, 3, 5, 8, and 11 y. We used Bayesian machine learning methods to identify distinct latent classes based on individual profiles of eczema, wheeze, and rhinitis. This approach allowed us to identify groups of children with similar patterns of eczema, wheeze, and rhinitis over time. Using a latent disease profile model, the data were best described by eight latent classes: no disease (51.3%), atopic march (3.1%), persistent eczema and wheeze (2.7%), persistent eczema with later-onset rhinitis (4.7%), persistent wheeze with later-onset rhinitis (5.7%), transient wheeze (7.7%), eczema only (15.3%), and rhinitis only (9.6%).
    [Show full text]
  • Uncertainty As a Form of Transparency: Measuring, Communicating, and Using Uncertainty
    Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertainty Umang Bhatt1,2, Javier Antorán2, Yunfeng Zhang3, Q. Vera Liao3, Prasanna Sattigeri3, Riccardo Fogliato1,4, Gabrielle Gauthier Melançon5, Ranganath Krishnan6, Jason Stanley5, Omesh Tickoo6, Lama Nachman6, Rumi Chunara7, Madhulika Srikumar1, Adrian Weller2,8, Alice Xiang1,9 1Partnership on AI, 2University of Cambridge, 3IBM Research, 4Carnegie Mellon University, 5Element AI, 6Intel Labs, 7New York University, 8The Alan Turing Institute, 9Sony AI ABSTRACT 1 INTRODUCTION Algorithmic transparency entails exposing system properties to Transparency in machine learning (ML) encompasses a wide va- various stakeholders for purposes that include understanding, im- riety of efforts to provide stakeholders, such as model developers proving, and contesting predictions. Until now, most research into and end users, with relevant information about how a ML model algorithmic transparency has predominantly focused on explain- works [17, 149, 194]. One form of transparency is procedural trans- ability. Explainability attempts to provide reasons for a machine parency, which provides information about model development learning model’s behavior to stakeholders. However, understanding (e.g., code release, model cards, dataset details) [8, 62, 136, 159]. An- a model’s specific behavior alone might not be enough for stake- other form is algorithmic transparency, which exposes information holders to gauge whether the model is wrong or lacks sufficient about a model’s behavior to various stakeholders [103, 164, 180]. knowledge to solve the task at hand. In this paper, we argue for con- The ML community has mostly considered explainability, which at- sidering a complementary form of transparency by estimating and tempts to provide reasoning for a model’s behavior to stakeholders, communicating the uncertainty associated with model predictions.
    [Show full text]