About me

I am a first year PhD student at National University of Singapore, Department of Anatomy, Yoo Loo Lin School of Medicine, attached to Dr. Jayantha Gunarathne's lab at Institute of Molecular and Cell Biology, A-Star.

My research area is focused on developing proteomic-centric computational pipeline to discover diagnostic and prognostic markers in breast cancer. I will also focus on exploring the ablity to use machine learning and artificial intelligence in the molecular biology area.

I completed my undergraduate degree computer science and engineering at University of Moratuwa, Sri Lanka, and worked as a software engineer at Enactor. I did my final year project on development of Neuroscience Decision Support System for ADHD Identification , supervised by Prof. Dulani Meedeniya.

Further, I would like to explore the development of deep learning models for brain disease/disorder identification and the study of current modalities like MRI,fMRI and EEG in application of advanced computational techniques.

My final year project led into discovery of novel method of using two modalities of eye movement and fMRI data in ADHD identification resulting high precision and recall values in deep learning model derivation. This research was recognized as the best research project of the year 2020 in Computer Science and Engineering at University of Moratuwa Awards Ceremony.

I like to write, photography and explore nature during my free time.

Neuroscience Decision Support System for ADHD Identification.

Children with ADHD are in danger of developing other mental health difficulties, particularly behavior and learning disorders such as oppositional defiant disorder, conduct disorder, and learning and language disorders. As researchers found 5-11% of children affected ADHD in United States representing 6.4 million children in countrywide (Alaa, 2015).

Therefore, early detection and classification of ADHD is important to prevent relevant children from future symptoms and difficulties in executive functions such as planning, organizing, initiating activities, and monitoring. Therefore, there is a major requirement to generate an objective biological tool which is capable of classifying ADHD and non-ADHD using a multi-modality approach including functional MRI (fMRI) and eye movement data.

Project website | Demo

---

Emotion Diary for Depression Prediction

This system is developed to give insights to the user upon his/her emotional stability and predict future emotional data using the user’s personal data collection. Further the system is going to be developed in order to provide a platform to keep track of the mood of the user and produce a report for a given period of time. Also, the system is able to identify the relationship between the face emotion and the text emotion.

In addition, the system is going to be designed in as a chrome extension to collect data and as a web application that it will be able to give notifications if any server mental health problem is identified. Unlike most similar applications this project automates the process of taking real moods of the user accurately. As a long term objective, the data collection of this project can be beneficial to use as a comprehensive data set for the research purposes.

---

Reporting feature for Siddhi

Report generation for Siddhi was a new project as it was a highly requested feature by the customers. As per the initial discussion it was required to generate reports from perceiving the current state in the snapshot view and report in PDF format. The widgets and the dashboard pages should be able to download as PDF files. The importance of perceiving the current snapshot view is that, since it shows streaming data which has a huge variation over the time; it should show the data in the report generation moment.

As the second part of the product, it was decided to generate scheduled reports via Jasper reporting library. It was decided to handle the scheduling part via Siddhi triggers and what was left to do is that the report generation sink extension.

Siddhi Extension | Documentation

---

DengAI: Predicting Disease Spread

Dengue is a viral infection which is identified as mosquito-borne tropical disease. The severe symptoms typically reveal in three to fourteen days after infection. This fatal disease cause to loose the lives of thousands of infected people all around the world DengAI challenge of this competition is to predict the number of total_cases for (city, year, weekofyear)in the test dataset. The given scenario consists of two major cities as San Juan and Iquito with the spanning of 5 and 3 years.

Applied novel data preprocessing pipeline and developed an ensembled classification model with Random Forest, LASSO, SVM (linear kernel), SVM (RBF kernel), Gradient Boost. Our team was able to achieve a mean abosulte error of 18.95, and ranked 152 among 6920 competitors.

---

Proof of Concept Implementation of Causal Consistency in Redis

Distributed databases with geographically replicated data nodes are widely used to support low latency and fault tolerance database services to the clients. In replicated distributed databases, the data are replicated in several locations to enhance their availability.

Hence, the issue in updating the database replicas in the same operational order has been addressed in this study based on the concept of totally ordered multicast with Lamport timestamping. Redis distributed database which supports the replication of nodes has been chosen to implement Lamport’s totally ordered multicast-based causal order of events in each replica.

White paper

---

Bio-metric Authentication System

A bio-metric authentication system which uses voice and keystroke dynamics to identify the user.

Singapore International Graduate Award

PhD training scholarship to pursue graduate studies at A*STAR Research Institute and National University of Singapore

2021 | September

---

CINTEC Award

Best Computer Science and Engineering project who has obtained highest marks at the CS 4202 - Research and Development Project. (By Council for Information Technology.)

Academic Year 2018/2019 2020| December

---

Best Paper Award

Biomedical Engineering Category: A Rule-Based System for ADHD Identification using Eye Movement Data

Moratuwa Engineering Research Conference 2019 | July

---

Best Project Award

Institute of Software Engineering Award Ceremony 2016 | July

---

Publications

[1] De Silva, S., Dayarathna, S., Ariyarathne, G., Meedeniya, D., Jayarathna, S., Michalek, A. M., & Jayawardena, G. (2019, July). A Rule-Based System for ADHD Identification using Eye Movement Data. In 2019 Moratuwa Engineering Research Conference (MERCon) (pp. 538-543). IEEE. doi.org/10.1109/MERCon.2019.8818865

[2] De Silva, S., Dayarathna, S., Ariyarathne, G., Meedeniya, D., & Jayarathna, S. (2019). A Survey of Attention Deficit Hyperactivity Disorder Identification Using Psychophysiological Data. International Journal of Online and Biomedical Engineering (iJOE), 15(13), 61-76. doi.org/10.3991/ijoe.v15i13.10744

[3] Ariyarathne, G., De Silva, S. , Dayarathna, S., Meedeniya, D., & Jayarathne, S. (2020, February). ADHD identification using convolutional neural network with seed-based approach for fMRI data. In Proceedings of the 2020 9th International Conference on Software and Computer Applications (pp. 31-35). doi.org/10.1145/3384544.3384552

[4] De Silva, S., Dayarathna, S. U., Ariyarathne, G., Meedeniya, D., & Jayarathna, S. (2021). fMRI feature extraction model for ADHD classification using convolutional neural network. International Journal of E-Health and Medical Communications (IJEHMC), 12(1), 81-105. doi.org/10.4018/IJEHMC.2021010106

[5] De Silva, S.,Dayarathna, S., Ariyarathne, G., Meedeniya, D., Jayarathna, S., & Michalek, A. M. (2021). Computational decision support system for ADHD identification. International Journal of Automation and Computing, 18(2), 233-255. doi.org/10.1007/s11633-020-1252-1

[6] S. de Silva, S. Dayarathna, & D. Meedeniya, Alzheimer’s Disease Diagnosis using Functional and Structural Neuroimaging Modalities, Wadhera, T., & Kakkar, D. (Eds), in Enabling Technology for Neurodevelopmental Disorders from Diagnosis to Rehabilitation, Ch. 11, pp. ..... Taylor & Francis CRS Press, 2022. https://www.routledge.com/Enabling-Tec-hnology-for-Neurodevelopmental-Disorders-From-Diagnosis/Wadhera-Kakkar/p/book/9780367761189 [To be published]

Visiting Lecturer, University of Moratuwa

B.Sc. Engineering Module CS2963 Presentation Skills

2019 – Jun 2019

Software Engineer, Enactor

Performance and Scalability Team

Feb 2020 – Dec 2021

---

Software Engineering Intern, WSO2

Siddhi Analytics Team

July 2018 – December 2018

Content Creator, Project Nenathambara

This project aims to bridge the gap between the technology and ICT education in Sri Lanka. I have been involved in creating content for Data Processing and Anlytics Python course.

Oct 2021 – Present