Chirath Hettiarachchi

Research Fellow and HDR Convener Data Science

Picture of Chirath Hettiarachchi

Location
Hanna Neumann Blg (145), 3.16

Email
Chirath.Hettiarachchi@anu.edu.au

Clusters

Publications
ORCiD
dblp
Google Scholar

ANU Research Profile
chirath-hettiarachchi

Social
Twitter
Github
LinkedIn

Biography

Chirath Hettiarachchi is a Research Fellow at the ANU School of Computing working on closed-loop healthcare applications. He completed his PhD at ANU, in the Big Data program of the “Our Health In Our Hands (OHIOH)” project, a strategic initiative of the university. His research focused on using ML to develop a control system for the artificial pancreas. He received his BSc (Hons) in Electronics & Telecommunication Engineering & MSc (Research) degrees from the University of Moratuwa, Sri Lanka in 2018 and 2020. He completed his professional qualification in Management Accounting at the Chartered Institute of Management Accountants in 2014 (CIMA-UK). He worked as a Machine Learning Engineer in the FinTech industry from 2018 - 2019, developing algorithms for identifying outliers in corporate financial transactions, forecasting, name screening and risk prediction applications. During the final year of his undergraduate studies, he pursued his entrepreneurial aspirations by co-founding a healthcare startup which achieved multiple awards and publications.

Research

I’m interested in developing Machine Learning and Reinforcement Learning (RL) algorithms for closed-loop clinical treatment strategies. You can try out our work on RL-based Artificial Pancreas Systems (APS) for automating glucose regulation treatment in Type 1 Diabetes by visiting CAPSML and learn more about this work using the following open-sourced resources:

Interests

  • Reinforcement Learning
  • Machine Learning
  • Control Systems
  • Signal Processing
  • Biomedical Engineering
  • Neuroinformatics

If you are interested to work with us on exciting research projects please reach out and find the latest available student research projects here.

bars search times