As a Summer Research Fellow at the RPI-IBM HEALS Research Center, I worked on a number of projects that aimed to improve the current state of medical knowledge representation and reasoning. My work focused on developing novel methods for extracting and analyzing medical knowledge from clinical practice guidelines, and incorporating that knowledge into knowledge graphs for use in medical diagnosis and treatment.
The HEALS project applies advanced cognitive computing capabilities to help people understand and improve their own health conditions. In particular, we are exploring areas including personalized and mobile medical care, improved healthcare analytics, and new data-based approaches to driving down the cost of medical care.
One of my key contributions was the development of the PaperRank framework, which utilized the PageRank algorithm to compute a community trust score for each publication in a citation network of academic articles. This score was derived from a non-parametric mixture model, and was designed to reduce bias in the final trust scores. The framework was highly scalable, allowing it to analyze the entire NCBI PubMed citation network of over 14 million articles.
In addition to my work on the PaperRank framework, I also developed the Guideline Explorer, a tool for efficiently visualizing and examining clinical practice guidelines. This tool allowed users to easily identify and analyze medical directives within the guidelines, and was applied to the American Diabetes Association's 2018 guidelines.
Alleviating the Burden of Diabetes with AI - THINK Blog
How many decisions do you make per day? From the minute you wake up in the morning until you fall asleep that same evening, you likely have to make endless choices: the route you drive to avoid traffic, what you should order for lunch, when to run errands, what type of exercise you want to [...]
Another project I worked on was the EHR Simulation engine, which utilized Monte Carlo simulations to suggest medical tests that would be statistically likely to identify previously unknown medical issues. This engine was designed to improve the efficiency and accuracy of medical diagnosis, by providing a probabilistic basis for test selection.
Overall, my time as a Summer Research Fellow at RPI was a valuable and enriching experience. I was able to contribute to cutting-edge research in the field of medical knowledge representation and reasoning, and I gained valuable experience working on large-scale data analysis and machine learning projects.
Sunset over the HEALS building