Tin Yau Lee

Tin Yau  Lee
Eric and Wendy Schmidt Center funded Research and Innovation Scholars
Advisor: Caroline Uhler
Department: EECS
Years: 2021-2022
Research Project Title:

Profiling Cell Interactions in Tissue Images Using Machine Vision

abstract:Inventions require knowledge. Understanding how our cells act, react, and talk provides valuable puzzle pieces in discovering new drugs. The goal of my project is to capture cellular interactions at the tissue level to enable a better understanding of cells. With state-of-the-art computer vision techniques such as visual transformers, we will systematically extract patterns in cellular activities from high-resolution tissue images. We will develop a self-learning model, optimize it, and apply it to practical use cases to explain morphological patterns in tissues without manual labeling. With this scalable way of pattern interpretation, we hope to enable novel discoveries on cellular structures and ultimately leverage that knowledge for developing drugs to better save lives.

I am participating in SuperUROP because I am very interested in applying my machine learning knowledge (from 6.036) to research and gain real-life experiences working alongside scientists. I hope to be able to publish a paper by the end of the project and contribute to our understanding of applied biology.