My research work is focused on leveraging structure in medical imaging/sensors to facilitate personalized treatment and a better understanding of the disease. In my work, I have used ideas from machine learning, computer vision, and statistics, to extract signatures in imaging and sensor data collected from patients. I envision that the algorithms I have developed can be used in a clinical setting to stratify patients and determine the optimal treatment regimen for each individual patient. I specifically aim to contribute to the nascent but emerging field of computational pediatrics, or using data science to understand pediatric disorders better.
Major goals of my research:
1. Data-driven diagnosis and personalized treatment: With the unprecedented increase in computational power and storage capability in the last few years, it has become possible to build data-driven models using machine learning to find clinically meaningful signatures in medical data. For example, one of my research goals is finding signatures in glucose monitor data that can predict the likelihood of hyperglycemia in young diabetes patients.
2. Understanding disease biology through data science: Over and above the most common machine learning task, prediction, my research seeks to uncover the biological phenomenon driving an ML model’s performance. For example, I am working with Baylor College of Medicine’s psychiatry department to discover signatures in brain imaging that help us understand the biological drivers of depressive disorders.