Simon A. Lee

I'm a 2nd year Ph.D. student in the Computational Medicine Department at UCLA working with Prof. Jeffrey Chiang. Previously, I worked as a data science researcher at Celsius Therapeutics exploring immunology cell quantification through deconvolution methods for our Phase I clinical trials. Prior to this, I obtained a Bachelor's of Science in Applied Mathematics and also studied abroad at École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland. In my free time, I enjoy spending time with people, traveling, eating, and discovering the mysteries of the world.

Email  /  Scholar  /  Twitter  /  Github

News

• June 2024: Our paper on Text Serialization was accepted into ICML AI4Science Workshop
• June 2024: I was awarded the Warren Alpert Computational Biology and AI Network Fellowship
• May 2024: I began working with Optum AI Labs as part of a grant
• May 2024: I put out my second preprint on: Antibiotics Susceptibility Prediction
• January 2024: I put out my first preprint: MEME
• August 2023: I begin my Ph.D. and working with Prof. Jeffrey Chiang
• August 2023: I finish work at Celsius Therapeutics. Special Thanks to Mukund Varma and Nicolas Stransky
• February 2023: I get admitted to the Computational Medicine Ph.D. Program
• January 2023: I begin working as a data science researcher at Celsius Therapeutics
• September 2022: I move abroad and begin at EPFL in Lausanne Switzerland
• August 2022: I completed my work on ABP+EHR with Dr. Akos Rudas and Jeffrey Chiang
• June 2022: I obtained my Bachelors of Science in Applied Mathematics

Research

I am broadly interested in representation learning for electronics health records (EHR) and other AI4Science work. I am particularly interested in using modern AI techniques (e.g. BERT, LLMs, etc.) to learn low level representations from high level data. I am also broadly interested in other popular avenues like algorithmic development and improvement, multimodality and digital twins.

Showcased how Medical Foundation Models can be leveraged for predicting Antibiotic Susceptibility to combat Antibiotic Resistance.

Developed the Pseudonotes method that transforms Tabular EHR into text. Also build the Multiple Emebdding Model for EHR (MEME) which helps concatenate heterogenous ("Multimodal EHR") features for Emergency Department Prediction

Service