Committees General Chair Zainal A. Hasibuan, University of /APTIKOM, Indonesia Program Co-Chairs

 Teddy Mantoro, Sampoerna University, APTIKOM, Indonesia  Fabrice Meriaudeau, Universiti Teknologi Petronas, Malaysi

Publication Co-Chairs

 Achmad Nizar Hidayanto, , Indonesia  Mochammad Wahyudi, STMIK Nusa Mandiri , Indonesi

Publicity Co-Chairs

 Harry Budi Santoso, University of Indonesia, Indonesia  Ina Agustina, , Indonesia

Technical Program Committee Chair Media Anugerah Ayu, Sampoerna University, Indonesia Organizing/Local Committee Co- Chairs

 Dwiza Riana, Universitas BSI, Indonesia  Khusnul Khotimah, Universitas Yapis Papua  Rangga Firdaus, , Indonesia

TPC members

 Abdullah Alkalbani University of Buraimi, Sultanate of Oman  Achmad Benny Mutiara, Universitas Guadarma, Indonesia  Adamu Ibrahim, International Islamic University Malaysia, Malaysia  Agus Buono, Bogor Agricultural University, Indonesia  Agus Hardjoko, Gajah Mada University, Indonesia  Ahmad Zeki, Bahrain University, Bahrain  Akram M. Zeki, International Islamic University Malaysia, Malaysia  Alamin Mansouri, Universite de Bourgogne, France  Anton Prabuwaono, King Abdul Azziz University, Saudi Arabia  Asep Juarna, Universitas Gunadarma, Indonesia  Ayu Purwarianti, Bandung Institute of Technology, Indonesia  Bharanidharan Shanmugam, University of Darwin, Australia  Christophoros Nikou, University of Ioannina, Greece  Dwiza Riana, Universitas BSI, Indonesia  Eko Kuswardono Budiardjo, University of Indonesia, Indonesia  Eri Prasetyo Wibowo, Gunadarma University, Indonesia  Evizal Abdul Kadir, Universitas Islam Riau, Indonesia.  Frederic Ezerman, Nanyang Technological Univiversity, Singapore  Fredy Purnomo, Binus University, Indonesia  H. Dawid, Universitaet Bielefeld, Germany  Heru Suhartanto, University of Indonesia, Indonesia  Iping Supriana Suwandi, Bandung Institute of Technology, Indonesia  Ismail Khalil, Johannes Kepler University, Linz, Austria  Kridanto Surendro, Bandung Institute of Technology, Indonesia  Lukito Edi Nugroho, Gajah Mada University, Indonesia  Michel Paindavoine, Burgundy University, France  Moedjiono, Budi Luhur University, Indonesia  Mohammad Essaaidi, Chair of IEEE Morocco Section, Morocco  Muhammad Zarlis, University of Sumatera Utara, Indonesia  Murni Mahmud , International Islamic University Malaysia, Malaysia  Naufal M. Saad, Universiti Teknologi Petronas, Malaysia  Normaziah Azis, International Islamic University Malaysia, Malaysia  Norshida Mohammad, Prince University, Saudi Arabia  Paulus Insap Santosa, Gajah Mada University, Indonesia  Prihandoko, Gunadarma University, Indonesia  Rila Mandala, Bandung Institute of Technology, Indonesia  Sabir Jacquir, Universite de Bourgogne, France  Salwani BTE Mohd Daud, Universiti Teknologi Malaysia, Malaysia  Shelvie Neyman, Institut Pertanian Bogor, Indonesia  Supriyanto, Universitas Gunadarma, Indonesia  Tole Sutikno, Ahmad Dahlan University, Indonesia  Tubagus Maulana Kusuma, Gunadarma University, Indonesia  Vincent Vajnovzski, Universite de Bourgogne, France  Waralak Siricharoen, University of the Thai Chamber of Commerce,Thailand  Wendi Usino, Budi Luhur University, Indonesia  Wisnu Jatmiko, University of Indonesia, Indonesia  Youssef Zaz, Abdelmalek Essaadi University, Morocco  Yusuf Yudi Prayudi, Universitas Islam Indonesia, Yogjakarta, Indonesia Message from the Program Chair

It is with great honour to welcome you all to The Second International Conference on Informatics and Computing (ICIC 2017) in Jayapura, Papua, Indonesia.The ICIC 2017 main topic is "Towards better competencies of ICT human resources and regional competitiveness in a global era" and the primary goal of the conference is to exchange, share and distribute the latest research and theories from our international community which provides an exchange of the state of the art and future developments in the two key areas of this process: Computer Science and Information Science. ICIC 2017 can be considered as a premier event for sharing knowledge and results in the area of a. Computer Engineering and Computer Systems, b. Computer Science/Informatics, Information Systems, c. Information Technology and Information Management and d. Software Engineering. It provides a platform to the researchers and practitioners from both academia as well as industry to meet and share the cutting-edge development in the field. In addition, this conferential meeting provides a great opportunity to exchange knowledge and experience for all the participants who join in from all over the world to discuss their new thinking and creations. ICIC 2017 serves as a forum for scientists, engineers and practitioners to meet and present their latest research results, ideas, and papers in the diverse areas of Computer Engineering, Computer Science, Information Technology and Information Systems. This conference is collaboratively organized by APTIKOM, hosted by APTIKOM Papua and Universitas YAPIS Papua, with co-hosted by STMIK Bumigoro, Universitas Pasundan, Universitas Bina Darma, Universitas Gunadarma, STMIK Bina Insani, Perguruan Tinggi Raharja, Bina Sarana Informatika, Universitas Komputer Indonesia, Universitas Dian Nuswantoro, Universitas Budi Luhur, STMIK Nusa Mandiri, STMIK Tasimalaya, Univeritas Tanjung Pura, STMIK , Universitas Katolik Parahyangan, Universitas BSI, LSP Informatika and STMIK Sepuluh November. This conference is sponsored by APTIKOM and technically co-sponsored by IEEE Indonesia Section This scientific conference covers 3 (three) keynote speakers and 5 (five) technical sessions. We accepted 95 papers out of 173 (Succes rate 55%) and has attracted researchers from 20 countries (paper submission came from 13 countries), i.e. Australia, Bahrain, Czech Republic, France, Germany, Greece, India, Indonesia, Ireland, Japan, Malaysia, Morocco, Nigeria, Oman, Poland, Saudi Arabia, Slovakia, South Korea, Sudan, and Thailand. The conference has 366 authors and 118 reviewers, who have carefully reviewed the papers. Each submitted paper was reviewed by at least 3 reviewers in the related fields, and all registered papers have the similary score of maximum 20% under the Turnitin plagiarism check. It is expected that this conference will become a main platform for providing and exchanging information, knowledge, skills, and experiences in the field of Computer Science and Information Science. ICIC conference will be held annually to make it an ideal platform for people to share views and experiences in informatics and computing. We are grateful to APTIKOM Papua and Universitas YAPIS Papua for hosting this ICIC This scientific conference covers 3 (three) keynote speakers and 5 (five) technical sessions. We accepted 95 papers out of 173 (Succes rate 55%) and has attracted researchers from 20 countries (paper submission came from 13 countries), i.e. Australia, Bahrain, Czech Republic, France, Germany, Greece, India, Indonesia, Ireland, Japan, Malaysia, Morocco, Nigeria, Oman, Poland, Saudi Arabia, Slovakia, South Korea, Sudan, and Thailand. The conference has 366 authors and 118 reviewers, who have carefully reviewed the papers. Each submitted paper was reviewed by at least 3 reviewers in the related fields, and all registered papers have the similary score of maximum 20% under the Turnitin plagiarism check.2017. We would like to take this opportunity to express our thanks to the members of the ICIC 2017 committees and our respected reviewers for providing proficient and valuable contribution in the preparation of this conference. We also would like to acknowledge all the participants and the co-hosts of ICIC 2017 for their supports to this conference. We do hope that the conference will be stimulating, informative, and enjoyable to everyone.

Thank you.

Prof. Dr. Teddy Mantoro, SMIEEE

Technical Session Schedule

Technical Session-1 Track 1

3 Security Comparison between Dynamic & Static WSN for 5G Abdullah Alkalbani and Teddy Mantoro Testing The Best Queue Method For Internet Network in Cibuntu Tourist

13 Village Halim Agung and Teddy Mantoro

25 Image Segmentation Using the Otsu Method in Dental X-Rays Anif Hanifa Setianingrum, Amanda Satya Rini and Nashrul Hakiem The Depth-First Search Column by Column Approach on the Game of Babylon

27 Tower Romi Fadillah Rahmat, Harry Harry, Opim Salim Sitompul, Mohammad Fadly

Syahputra and Erna Budhiarti Nababan

Track 2 Comparison of Nucleus and Inflammatory Cell Detection Methods on Pap

11 Smear Images Regina Lionnie and Mudrik Alaydrus Identification of Molar and Premolar Teeth in Dental Panoramic Radiograph

28 Image Romi Fadillah Rahmat, Silviani Silviani, Erna Budhiarti Nababan, Opim Salim

Sitompul, Rina Anugrahwaty and Silmi Lubis Developing Indonesian Corpus of Pornography using Simple NLP-Text Mining

33 (NTM) Approach to Support Government Anti-Pornography Program Miftah Andriansyah, Marshal Samos, Ali Akbar, Antonius Irianto

Sukowati, Muhammad Subali and Imam Purwanto Chess Piece Movement Detection and Tracking, A Vision System Framework

38 for Autonomous Chess Robot Dennis Aprilla Christie, Tubagus Maulana Kusuma and Purnawarman Musa

Track 3 Funding Eligibility Decision Support System Using Fuzzy Logic Tsukamoto

4 (Case: BMT XYZ) Bayu Sanggi Ardika, Anif Hanifa Setianingrum and Nashrul Hakiem An Enterprise Architecture Planning for Higher Education Using The Open

23 Group Architecture Framework (TOGAF): Case Study University of Lampung Gigih Forda Nama, Tristiyanto Tristiyanto and Didik Kurniawan Improving E-Government Services for the Poor through Transparency and

26 Trust Darius Antoni, Muhammad Akbar, Apriliani Apriliani and Muhammad Izman

Herdiansyah Concept-based Multimedia Information Retrieval System using Ontology

42 Search in Cultural Heritage Ridwan Andi Kambau and Zainal Arifin Hasibuan

Track 4 Generating Automatic Marker Based on Combined Directional Images from

54 Frequency Domain for Dental Panoramic Radiograph Segmentation Amelia Sahira Rahma, Eva Firdayanti Bisono, Agus Zainal Arifin, Dini Adni

Navastara and Rarasmaya Indraswari

62 Setiawan Hadi, Undang Ahmad Darsa, Erick Paulus and Mira Suryani

78 Nur Rokhman Solving Non-Linear Equations Containing Spline Interpolation Function by

Relaxing The Newton Method Implementation of Fuzzy C-Means Algorithm and TF-IDF on English Journal

85 Summary Mohamad Irfan, Jumadi Jumadi, Wildan Budiawan Zulfikar and Erik Erik Technical Session Schedule

Technical Session-2

Track 1

173 A Research Framework of Disaster Traffic Management to Smart City

Dedy Hartama, Herman Mawengkang, Muhammad Zarlis, Rahmat Widia Sembiring,

Dahlan Abdullah, Mhd Furqan and Robbi Rahim

168 Towards a Drug Information Interoperability Mechanism

Siti Saidah, Lintang Yuniar Banowosari and Aries Muslim

Diagnostic Decision Support System of Chronic Kidney Disease Using Support Vector

153 Machine

Mubarik Ahmad, Vitri Tundjungsari, Dini Widianti, Peny Amalia and Ummi Azizah

Rachmawati

150 Modified Adaptive Affinity Propagation with Similarity Distribution Based Preference

Rina Refianti, Achmad Benny Mutiara, Adang Suhendra and Asep Juarna

A Cluster Validity for Spatial Clustering Based on Davies Bouldin Index and Polygon

146 Dissimilarity Function

Ichwanul Muslim Karo Karo, Arief Fatchul Huda and Kiki Maulana Adhinugraha

Track 2 Security of Data Communications Between Embedded Arduino Systems with

156 Substitution encryption Julham, Ferry Fachrizal, Hikmah Adwin Adam, Yulia Fatmi and Arif Ridho Lubis

Estimation System of Occupant Behavior Against The Use Of Electricity Using Bayes

158 Method And Decision Tree Algorithm

Almas Shabrina, Feri Fahrianto and Nenny Anggraini

159 Identification of Active and Passive Sentence for Plagiarism Detection

Dina Anggraini, Achmad Benny Mutiara, Tubagus Maulana Kusuma and Lily

Wulandari

Improving the Accuracy of Complex Activities Recognition Using Accelerometer-

171 Embedded Mobile Phone Classifiers

Mohammed Mobark, Suriayati Chuprat and Teddy Mantoro

Velocity and Acceleration Analysis from Kinematics Linear Punch Using Optical

160 Motion Capture

Dharmayanti, Mohammad Iqbal, Adang Suhendra and Achmad Benny Mutiara

Track 3 EMPIRICAL STUDY ON CONSUMER ACCEPTANCE OF MOBILE APPLICATIONS IN

129 JAKARTA INDONESIA

Sri Nawangsari, Eri Prasetyo Wibowo and Raden Budiarto

130 Sri Nawangsari, Eri Prasetyo Wibowo and Raden Budiarto

Ira Puspitasari, Reza Pramudhika and Indra Kharisma Raharjana

COMPARATIVE ANALYSIS BETWEEN ONLINE E-LEARNING AND FACE TO FACE

151 LEARNING: AN EXPERIMENTAL STUDY

Anthony Anggrawan and Jihadil Qudsi S.

A recommendation system for culinary tourists in Jogjakarta based on collaborative

112 filtering

Bernard Suteja and Suryo Guritno

107 Learning Style Model Detection Based on Prior Knowledge in E-learning System

Muhammad Said Hasibuan

Track 4 System Architecture for a Distributed Digital Signage System in Developing Countries:

154 Leveraging Open Hardware, Open Source Software, and Open Standard

Rahmad Dawood, Fadhlul Fahmi and Mohd. Syaryadhi Web Services of Transformation Data Based on OpenEHR into Health Level Seven

142 (HL7) Standard

Aries Muslim, Achmad Benny Mutiara, Sulistyo Puspitodjati and Teddy Oswari

Problematic Internet Use (PIU): The Role of Emotional Factors on Social Media

109 Activities

Esther Andangsari

91 A Speech Intelligence Conversation Bot for Interactive Media Information

Jefri Yushendri, Alvian Rahman Hanif, Anneke Annassia Putri Siswadi, Purnawarman

Musa, Tubagus Maulana Kusuma and Eri Prasetyo Wibowo

84 EXPERT SYSTEM FOR CHOOSING PROPERTY BASED ON RULE BASED REASONING

Rosalina Rosalina, Nur Hadisukmana, Rikip Ginanjar, Budi Sulityo and R.B Wahyu

10 Land use growth simulation and optimization in the urban area Rahmadya Trias Handayanto, Nitin Kumar Tripathi, Sohee Minsun Kim and

Herlawati Implementation of Expert System for Selecting Appropriate Mobile

18 Application Architecture Using CLIPS Mr Prihandoko, Andrey Agassy Christian, Fuad Baskara Priyambada and Nur

Prilianti Kusuma Dewi Clustered Directed Diffusion Modelling for Tesso Nilo National Parks Case

30 Study Indra Yasri

94 Comparison Methods for Selecting Best Images to Solar Panel Monitoring Sara Lafkih and Youssef Zaz Classification on Indonesian Social Media Celebrity (Selebgram) using Support

31 Vector Machine Method Miftah Andriansyah and Ali Akbar

Track 4

90 An ICT Adoption for Education: A Proposed Framework Sofiana Nurjanah, Zainal A. Hasibuan and Harry Santoso The Expert-Judgement Validation and Finalization of Proposed Interaction

86 Design Process Maturity Instrument Isnaeni Nurrohmah, Dana Indra Sensuse and Harry Budi Santoso The Effect of e-CRM towards Service Quality and Net Benefits Using

103 Structure Equation Modeling Harrizki Arie Pradana, Bob Subhan Riza, Muchammad Naseer, Djoko Soetarno

and Teddy Mantoro Researching Computing Teachers’ Attitudes Towards Changes in the

1 Curriculum Content – an Innovative Approach or Resistance? Jiří Dostál, Anna Brosch, Xiaojun Wang and William Steingartner Performance Evaluation of Accounting Information System for Restaurants

122 SMEs In Jabodetabek Lana Sularto Fixed Asset Management System Development FOR PT. INVOSA SYSTEMS

128 “AVOSA SYSTEMS” Rosalina Rosalina, Rikip Ginanjar, Rb Wahyu, Nur Hadisukmana and Afifah

Muftinisa Technical Session Schedule

Technical Session-4

Track 1 A Survey Regarding the Readiness of Campus in Indonesia on the Adoption of Green

63 Computing

Sofwan Hanief, Luh Gede Surya Kartika and Ni Luh Putri Srinadi

Media Interactive Learning and Biology Subjects Implementation with Augmented

47 Reality Application

Sardjoeni Moedjiono, Nurcahyadi and Aries Kusdaryono

Solving Shortest Path Problems On Installing Data Network Connection Using Apriori

35 Algorithm

Ali Akbar, Nurul Adhayanti, Ike Kusumawijaya and Hendri Putra

Determining Components of National Cyber Security Framework Using Grounded

68 Theory

Farisya Setiadi, Panca O Hadi Putra, Yudho Giri Sucahyo and Zainal A. Hasibuan

Track 2 Decision Support System In Giving Recommendation For Flat Screen Television

65 Purchase Using Analytical Hierarchy Process (AHP) Method

Marla Sheilamita Shalin Pieter, Fegie Y Wattimena and Iriani Inggrit Lamia

Sentiment Analysis of Students’ Perception on the Use of Smartphones: A Cross

46 Sectional Study

Aminu Onimisi Abdulsalami, Barroon Isma'Eel Ahmad, Muhammad Aminu Umar, Amina Hassan Abubakar, Fatsuma Jauro, Aliyu Muhammad Kufena and Emmanuel Ameh Ekoja

A Short Film Making With 3D CGI And Live Action Footage Usage Using Compositing

21 Technique And Key Frames Method

Ina Agustina, Fauziah Fauziah and Maulina Dwi Utami Increasing Messenger Service Capability and Security on Mobile Devices Using Hybrid

14 Compression-HR2 Algorithm

Teddy Mantoro and Yosep Lazuardi

Track 3 Levensthein Distance as a Post-Process to Improve the Performance of OCR in

101 Written Road Signs

Satria Priambada and Dwi Widyantoro

File Encryption and Hiding Application Based on Advanced Encryption Standard (AES)

113 and Append Insertion Steganography Method

Gotfried C. Prasetyadi, Achmad Benny Mutiara and Rina Refianti

A Novel Scheme for Handwritten Binarization Method on Sundanese Palm Leaf

117 Document Images

Erick Paulus, Mira Suryani, Setiawan Hadi and Intan Nurma Yulita

Identifying The Relevant Page Numbers that Referred by The Back-of-Book Index

139 Using Syntactic Similarity and Semantic Similarity

Sherly Christina and Enny Dwi Oktaviyani

Track 4 Analysis of Comparison between Sequential and Parallel Computation Using OpenMP

114 for Molecular Dynamic Simulation

Adang Suhendra, Helen Wijaya and Achmad Benny Mutiara

Kolmogorov Watermarking Technique for Secure the data of Wireless Sensor

135 Networks

Bambang Harjito

Information Technology Strategic Plan Development Methodology Governing from

167 the Perspectives of Enterprise Architecture

Richardus Eko Indrajit

Application Of GIS In The Spatial Analysis to Assessing the Infrastructure Dynamics of

95 Slum Upgrading In Papua, Indonesia

Iis Roin Widiati

Technical Session Schedule Technical Session-5

Track 1 Hierarchical Decision Approach Based on Neural Network and Genetic Algorithm

12 Method for Single Image Classification of Pap Smear

Yudi Ramdhani and Dwiza Riana

100 Development of Extensible Open Information Extraction

Elvan Owen and Dwi Widyantoro

Comparison Analysis between Implementation of Principal Components Analysis and

137 Haar Wavelet as Feature Extractors in Palmprint Recognition System

R. Rizal Isnanto, Risma Septiana, Ajub Ajulian Zahra, Ilina Khoirotun Khisan Iskandar

and Galih Wicaksono

Database: Taxonomy of Plants Nomenclature for Borneo Biodiversity Information

79 System

Edy Budiman and Sitti Nur Alam

81 User Perceptions of Mobile Internet Services Performance in Borneo

Edy Budiman and Sitti Nur Alam

98 Designing a Wheeled Robot Model for Flammable Gas Leakage Tracking

Heru Supriyono and Ahmad Nur Hadi

Track 2 Analysis of Intrinsic Factors of Mobile Banking Application Users' Continuance

125 Intention

Muhammad Baharudin Jusuf, Achmad Nizar Hidayanto, Nilamsari Putri Utami and

Muhammad Rifki Shihab

97 Digital Government Services in Papua

Mursalim Tonggiroh

Why Does People Use E-Payment Systems in C2C E-Marketplace? A Trust Transfer

56 Perspective

Rito Septi Tombe, Rika K. Ekawati, Nur Fitriah Ayuning Budi, Achmad Nizar Hidayanto

and Pornthep Anussornnitisarn

Analysis of Internal and External Factors Influencing User’s Knowledge Sharing

61 Behavior on TMC Polda Metro Jaya’s Twitter Using Theory of Planned Behavior

Annisa Monicha Sari, Nizar Hidayanto, Prahastiwi Utari, Solikin and Betty Purwandari Factors influencing citizen’s intention to participate in e-participation: integrating

66 Technology Readiness on Social Cognitive Theory

Khoirunnida, Achmad Nizar Hadiyanto, Betty Purwandari and Meidi Kosandi

Track 3

108 Hamiltonicity on Enhanced Extended Fibonacci Cube (E2FC) Interconnection Network

Mufid Nilmada, Ernastuti Ernastuti and Djati Kerami

96 Information Credibility Factors on Information Sharing Activites in Social Media

Afira Putri Ghaisani, Qorib Munajat and Putu Wuri Handayani

Factors Influencing Continuance Intention of Travel Agency Information System Use:

83 A Case Study of PowerSuite

Tubagus Kusuma, Puspa Sandhyaduhita and Muhammad Rifki Shihab

The Effects of Pictures, Review Credibility and Personalization on User's Satisfaction

87 of Using Restaurant Recommender Apps

Nur Fazri Ilham, Putu Wuri Handayani and Fatimah Azzahro

Policy and Procedure Design for Video Conference Service using Soft-System

104 Methodology: A Case Study of PT Pertamina (Persero)

Tami Utiwi Handayani, Satrio Baskoro Yudhoatmojo, Puspa Indahati Sandhyaduhita

and Adhi Yuniarto Laurentius Yohanes

Track 4 User Acceptance Factors Affecting the Usage of Mobile Learning In Enriching Outside

57 Classroom Learning at High School Level

Dini Seprilia, Putu Wuri Handayani and Ave Adriana Pinem

60 The Analysis of Tourism Information to Enhance Information Quality in E-Tourism

Hartasia Susan Panadea, Putu Wuri Handayani and Ave Adriana Pinem

Antecedents and Patterns of E-business Adoption among Small and Medium

72 Enterprises

Panca O. Hadi Putra, Harry B. Santoso and Zainal A. Hasibuan

Detection Water Level in Smart-Home's Bathtub In Saving Water Using Fuzzy

166 Ultrasonic Approach

Teddy Mantoro, Wirawan Istiono and Indrayutta Karima

Social Media Strategies for Public Diplomacy: a Case Study in the Ministry of Foreign

67 Affairs of the Republic of Indonesia Hary Hadiansyah, Betty Purwandari, Riri Satria and Satrio Baskoro Yudhoatmojo

Risk-Assessment Based Academic Information System Security Policy Using OCTAVE

110 Allegro and ISO 27002

Muhammad Taher Jufri, Mokhamad Hendayun, Toto Suharto

Comparison of Nucleus and Inflammatory Cell Detection Methods on Pap Smear Images

Dwiza Riana Achmad Nizar Hidayanto STMIK Nusa Mandiri Jakarta Universitas Indonesia Jakarta, Indonesia Depok, Indonesia [email protected] [email protected]

Mochamad Wahyudi STMIK Nusa Mandiri Jakarta Jakarta, Indonesia [email protected]

Abstract— Detection and identification of cells in a Pap smear the abnormality of the cell. Therefore, we need a method to test is very important for determining cell abnormalities. extract the inflammatory cells and identify the nucleus cells. Detection of cells becomes an important stage in early detection of cervical cancer. The presence of inflammatory cells in Pap smear In the Pap smear examination, there are two objects to be images often results in the identification of nuclei in the early concerned, that are the nucleus and cytoplasm. Stadium or level detection of cervical cancer in Pap smear image becomes difficult of cancer can be seen from the proximity between the wide area to do. The success of inflaming the inflammatory cells and of nucleus and cytoplasm. Many previous studies have talked detecting the nucleus will facilitate the process of identifying the about identification of nucleus and cytoplasm such as the nucleus. This will greatly help the development of information identification using the nucleus feature [2], combining the shape, acquisition system technology and the classification of single Pap texture, and intensity features on the nucleus [3]. Many methods smear cell image for early detection of cervical cancer. The paper of nucleic and cytoplasmic detection were proposed by previous aims to compare three methods of nuclear detection and researchers. To perform the detection on the edge of the nucleus inflammatory cell in a single Pap smear image. The results of the and cytoplasm, several segmentation methods were proposed comparison of nucleus and inflammatory cell in the test of data such as automatic segmentation of the nucleus and cytoplasm consisting of 84 images with inflammatory cells, showed that the [4] [5] [6] using an efficient gap-search [7] and using intensity K-Means and Bayesian classification method has not been able to variation between super pixels [8] and contour Detector [9]. accurately detect the nuclei and inflammatory cell, rather than the There is another segmentation method to detect the nucleus like nuclei and inflammatory cells detection base on the combination using morphological reconstruction and clustering [10], of gray level thresholding method. including fuzzy clustering for nucleus segmentation [11], color Keywords— Nucleus Detection, Inflammatory Cells, K-means, canal modification [12], and Naïve Bayes [13]. Even research Bayesian, Gray Level Thresholding, Pap Smear Images. that discussed not only with the identification of nuclei itself is associated with the reduction of inflammatory cells [14], [15], I. INTRODUCTION [16], and [17]. Many studies of Pap smear always contain information about The study discussed about inflammation cells associated the disease that affects all women of all ages. Women are very with the nucleus and cytoplasm in the Pap smear image is still susceptible to cervical cancer. But this disease tends to affect rare. Whereas the presence of inflammatory cells is quite women who are sexually active. Around 75% of cancer deaths disturbing when the identification of cell abnormalities. So it occur in low- and middle-income countries, where the number needs to be investigated which method is more precise and of cancer cases is rising most rapidly [1]. Until now the successful to detect nuclei, inflammatory cells and cytoplasm. incidence of cervical cancer can be lowered by early detection So this study will compare three methods in detecting nuclei, of cervical cancer through Pap smear tests. inflammatory cells and cytoplasm. The Pap smear test has various outputs depending on which The proposed algorithm is based on the combination of gray type of test is selected, liquid base or conventional. In level thresholding and the definition of a distance rule, which developing countries like Indonesia, the Pap smear test is entails in the identification of inflammatory cells. The proposed conventionally coveted to be excellent because of its affordable method is suitable for the extraction of the inflammatory cell in cost, but the problem that appears is the number of inflammatory the normal superficial class [15]. The other method proposed by cells whose existence is very similar to nucleus where that [17] is based on pixel classi¿cation techniques for the detection makes it quite difficult to detect the nucleus in order to identify of the nuclei where the K-means and a Bayesian classi¿er were used for the detection of the nuclei and other pixels class. The method compared in this study is not a new method, but so far these methods have not been tested for inflammatory cells. The success of the segmentation method of inflammatory cell lies in the success of identifying a single cell object in the initial process or pre-processing. The object recognition techniques on these methods are expected to separate the cytoplasm from the background perfectly to enter at the stage of cell segmentation. To obtain the process of introducing nucleus object and superior inflammatory cells, this study identifies nucleus cells and inflammatory cells base on pixel classification using K means and Bayesian classification compared with the (a) (b) method of nuclei cell detection base on the combination of gray Fig 2. Pap smear images without inflammatory cells (a), and with inflammatory level thresholding. The selection of both methods in this study, cells (b). based on K-Means and Bayesian classification as well as thresholding techniques, provides the effective image In this work we have considered several cells on Pap smear segmentation. images with inflammatory cells. The input images used in this This paper is organized into sections as follow. Section 2 method are conventional Pap smear images in single cells. The describes the methodology in detail. Section 3 presents the identification method of nucleus and inflammatory cells in this results of the methods applied to our image data base and its study used image input of an original image that had an performance, and section 4 is the conclusions of the proposed inflammatory cell while the output was detection candidate method. nucleus and cytoplasm image. Figure 3 is a proposed research design. The research II. METHOD design consists of several stages consisting of nuclei and Object recognition was required to get the specific object of inflammatory cells detection base on pixel classification using Pap smear cell image. Fig 1 a shows the objects that become the K-Means and Bayesian classification and base on the focus in this method, they are the nucleus cell, cytoplasm, and combination of gray level thresholding. inflammatory cell. Pap smear cell images with

Nucleus (plural: nuclei) in the general meaning is a nucleus inflammatory cells or a central part surrounded by other parts in a group or set. In cellular biology, the nucleus has a special meaning, namely nucleus of a cell containing chromosomes (DNA). Inflammatory The Pixel classification using The combination of gray cells or inflammation are small round or black; in the context of K means and Bayesian level thresholding. this study is that the inflammatory cell disrupts the field of view of preparation reading. If there are too many inflammatory cells, it can complicate the diagnosis of pathologists because the cells The Image Result covered by inflammation and the field of view becomes dirty. This complicates cell assessment.

(a) (b) (c) Application of Nucleus Detection

Fig 3. The Proposed Research Design

A. Nuclei and inflammatory cells detection base on pixel classification using K means and Bayesian classification. There were three stages performed in the process of identifying nuclei and inflammatory cells in the image. The steps of the algorithm consist of 3 stages as follows: 1. Image Simplification

Simplifying the image by removing the background: Fig 1. The research object focus inflammatory cell (a), nucleus cell (b) and a) The image was converted into a black-and-white cytoplasm (c). image

b) Detected cells and nuclei were formerly colored only The data used for this research can be accessed through the in black backgrounds. website page: http://sipk.dwiza.web.id/citra. Figure 2 shows 2. Application of K-means Classification cell images without inflammation and cell images with 3. Application of Bayesian Classification inflammation. In this method, we performed the background cleaning and TABLE I. THE DISTRIBUTION OF DATASET classification of each pixel as pixel nucleus or cytoplasmic Data pixel. Fig 4 is a training image for the application of this Type of Data Cytological Isolated Overlapping method. Image cells cells

Dataset 200 230 147 Data training 140 163 121

Data test 60 85 39

In our experiments, we used 200 cytological images of conventional Pap smears, from which we automatically obtained 230 isolated cells and 147 images of overlapping cells. For the Fig 4. Training image for Bayesian classification construction of the training set we have selected 140 out of the 200 aforementioned images. Consequently, we came out with B. Nuclei and inflammatory cells detection base on the 163 images of isolated cells and 121 images of overlapping cells. combination of gray level thresholding. Results from both Bayesian and K-means classifications The nucleus and inflammatory cell detection were achieved by tested to 60 images obtained the condition that these two applying the following algorythm. methods have not been able to detect nuclei, inflammatory cells, 1. Convert the image from RGB to grayscale. and cytoplasm accurately when compared with combination of 2. Improve the image with adjustment and Unsharp filter. gray level thresholding method. 3. Segment the image by applying global threshold of 0.65 to obtain a crystal black and white image of cytoplasmic As shown by the image processing results in Fig 5 and 6, in all tested images, K-Means and Bayesian classification show the candidate. results which are not as accurate as expected where in some 4. Calculate the cytoplasm feature, namely centroid, area, conditions the cytoplasm and the nucleus were not perfectly and bounding box. detected. 5. Crop cytoplasm automatically with bounding box> 200x200 pixels. 6. If the automatically cropped nucleus candidate > 200x200 pixels, then S is the detected cytoplasm image or sub- image of the initial image. 7. Sub image segmentation was done for k = 1,2,3, ..., n; where n is the sub image. 8. Convert the sub image from RGB to grayscale. 9. Increase the sub image with adjustment and Unsharp. 10. Apply global threshold of 0.25 to get a black and white image of nucleus candidate. 11. Calculate the nucleus candidate features which were centroid, area, and bounding box. 12. Crop the candidate nucleus automatically with bounding box> 13x13 pixels. 13. If the cropped nucleus candidate > 13x13 pixels, then Cn is the nucleus candidate.

III. RESULT AND DISCUSSION Both methods were evaluated on a conventional Pap smear image data set each containing 222 original images in our data set. All of these images were acquired through a Logitech camera (Logitech HD web cam C525) adapted to an optical microscope (Olympus CH20). A 40x magnification was used, Fig 5. Image collection of the running results of K-Means method and then the results were stored in JPEG format. In Table.1, the distribution of the data was given.

Original K-means Bayesian Combination Images of Gray Level Thresholding

Not detected Not detected

Cytoplasm and several candidate nucleus was detected

Nucleus detected Not detected

Cytoplasm and candidate nucleus Fig 6. Image collection of the running results of Bayesian method. was detected

At the same time, the results of combination of gray level thresholding method show that cytoplasm and nucleus candidate From the image data of 60 cytological images obtained 85 were successfully detected. The final result of the process was images of single or isolated cells and 39 images of overlapping successfully classified into two images that are cytoplasm and cells, the detection results using K-means and Bayesian indicate nucleus candidate. The comparison of the running results of that both classification methods have not been able to produce some images can be seen in Table II. detection of nucleus candidate. Dif¿cult to obtain ef¿cient training since learning rate must be de¿ned for different data TABLE II. THE RESULT OF COMPARING IMAGES situation. Hard to distinguish between nucleus and cytoplasm. Original K-means Bayesian Combination The difficulty of detecting nucleus candidate was probably Images of Gray Level due to too many image variations in the data set while the Thresholding combination of gray level thresholding method of processing 60 cytological images (obtained 85 images of single or isolated cells and 39 images of overlapping cells) can be detected as many as 448 nucleus candidates, so the combination of gray level thresholding method with a series of image processing processes can be used for detection of nucleus candidates

Candidate nucleus was Not detected consisting of the nucleus itself and inflammatory cells. It must detected be noted that the training and the test set of the images were independent. These images were acquired through a Logitech Cytoplasm and camera (Logitech HD C525) adapted to an optical microscope candidate nucleus (Olympus CH20). We have used a 40x magnification lens and was detected the acquired images were stored in JPEG format. Based on this result, the research was continued by making an application from combination of gray level thresholding method to facilitate the researcher in analyzing the image on Pap smear image. Fig 7 is the result of the application output when Cytoplasm and candidate Not detected nucleus was detected processing the image. With this application then the execution time of the image can be calculated. The execution of 60 images using the combination of gray level thresholding application Cytoplasm and took a processing time of 0.1645 ± 0.1159 seconds, and that was candidate nucleus was quite promising. detected [3] M. E. Plissiti, C. Nikou, and A. Charchanti, “Combining shape, texture and intensity features for cell nuclei extraction in Pap smear images,” Pattern Recognition Letters, vol. 32, no. 6, pp. 838–853, 2011. [4] Z. Lu, G. Carneiro, and A. P. Bradley, “Automated nucleus and cytoplasm segmentation of overlapping cervical cells,” in Proceedings of Medical Image Computing and Computer-Assisted Intervention MICCAI 2013, Lecture Notes in Computer Science, 2013, vol. 8149, pp. 452– 460 [5] Z. Lu, G. Carneiro, and A. P. Bradley, “Automated nucleus and cytoplasm segmentation of overlapping cervical cells,” in Proceedings of Medical Image Computing and Computer-Assisted InterventionMICCAI 2013, Lecture Notes in Computer Science, 2013, vol. 8149, pp. 452– 460. [6] Hasanuddin. Dwiza Riana, Dyah ekashanti Octarina Dewi, dwi H. Widyantoro.”Detection of Cytoplast Area of Pap smear Image Using Image Segmentation”.2012. [7] Lili Zhao, Kuan Li, Mao Wang, Jianping Yin, En Zhu, Chengkun Wu, Siqi Wang, Chengzhang Zhu, "Automatic cytoplasm and nuclei segmentation for color cervical smear image using an efficient gap-search MRF", Computers in Biology and Medicine, vol. 71, pp. 46, 2016, ISSN 00104825. [8] M. E. Plissiti, M. Vrigkas, and C. Nikou, “Segmentation of cell clusters in Pap smear images using intensity variation between superpixels,” pp. 184–187, 2015.

[9] M. H. Tsai, Y. K. Chan, Z. Z. Lin, S. F. Yang-Mao, and P. Fig 7. The Application of Nucleus Detection. C. Huang, “Nucleus and cytoplasm contour detector of cervical smear image,” Pattern Recognition Letters, vol. IV. CONCLUSION 29, pp. 1441–1453, 2008. This research has compare a methodology for detecting the [10] M. E. Plissiti, C. Nikou, and A. Charchanti, “Automated detection of cell nuclei in Pap smear images using nuclei and the inflammatory cell in Pap smear images. The best morphological reconstruction and clustering,” IEEE method is particularly useful to correctly interpret the cervical Transactions on Information Technology in Biomedicine, cells microscopic images and accurately diagnose the derivation vol. 15, no. 2, pp. 233–241, 2011. conclusions. The method has been tested in terms of the Pap [11] Saha, Ratna, Mariusz Bajger, and Gobert Lee. "Circular smear images application, and it was also verified by shape constrained fuzzy clustering (CiscFC) for nucleus pathologist. We can come out with the fact that the combination segmentation in Pap smear images." Computers in Biology and Medicine 85, pp 13-23, 2017. of gray level thresholding method is suitable for the detection of [12] D. Riana, D. Ekashanti, O. Dewi, D. H. Widyantoro, and nuclei and inflammatory cell in Pap smear images. This study is T. L. R. Mengko, “Segmentation and Area Measurement in a preliminary study in an attempt to find the detection methods Abnormal Pap Smear Images Using Color Canals of nuclei and inflammatory cells. As future work, we intend to Modification with Canny Edge Detection,” in International extend our method in order to be able to accurately recognize the Conference on Women’s Health in Science & Engineering, detection nuclei and inflammatory of the overlapping cells. 2012, pp. 1–4. Furthermore, we consider extracting inflammatory cells that [13] D. Riana, , and Fitriyani, “Integration of Bagging and Greedy Forward Selection on Image Pap Smear Classification using Naïve Bayes”. The overlap in both normal and abnormal Pap smear cell images. 5th International Conference on Information Technology for Cyber and IT Service Management (CITSM), 2017. ACKNOWLEDGMENT [14] I. Muhimmah, R. Kurniawan and Indrayanti, "Analysis of Dwiza Riana would like to thank The Ministry of Research, features to distinguish epithelial cells and inflammatory Technology and Higher Education (RISTEKDIKTI), Indonesia cells in Pap smear images," 2013 6th International Conference on Biomedical Engineering and Informatics, through Penelitian Pasca Doktor (2017) for supporting this Hangzhou, 2013, pp. 519-523. research, Laboratorium Khusus Patologi Veteran,Bandung, [15] D. Riana, M. E. Plissiti, C. Nikou, D. H. Widyantoro, and Indonesia for the Pap smear images database. T. L. R. Mengko, “Inflammatory cell extraction and nuclei detection in Pap smear images,” Int. J. e-Health Med. REFERENCES Commun., vol. 6, no. 2, pp. 27–43, 2015. [16] D. Riana, D.H. Widyantoro, T. L Mengko, "Extraction and classification texture of inflammatory cells and nuclei in [1] World's health ministers renew commitment to cancer normal Pap smear images", Instrumentation prevention and control. 2017. Communications Information Technology and Biomedical http://www.who.int/cancer/media/news/cancer- Engineering (ICICI-BME) 4th International Conference prevention-resolution/en/, accessed May 2017 on, 2015, pp. 65-69. [2] M. E. Plissiti and C. Nikou, “Cervical cell classi¿cation [17] Lezoray, O., dan Cardot, H. “Cooperation of Color Pixel based exclusively on nucleus features,” in Proceedings of Classification Schemes and Color Watershed: A Study for the 9th International Conference on Image Analysis and Microscopic Images”, IEEE Trans. Image Process, vol 11, Recognition, Lecture Notes in Computer Science, 2012, no 7, pp 783–789, 2002. vol. 7325, pp. 483–490.