New Haven, Connecticut, United States
Personal website with full CV: https://kevinwjin.com I am a third-year PhD candidate in Computational Biology and Biomedical Informatics at Yale University. My research concentrates on training a large language model for evidence-grounded psychiatric treatment recommendation. I am grateful and fortunate to be advised by Prof. Hua Xu, my thesis committee Profs. Hua Xu, Qingyu Chen, Sarah Yip, and Carolyn Rodriguez, and supported by the NSF Graduate Research Fellowship. I completed my bachelor's degree in molecular and cellular biology at Johns Hopkins University. As an undergraduate, I was a research assistant under Prof. Soojung Claire Hur in the Department of Mechanical Engineering, and under Prof. Steven S. An in the Department of Environmental Engineering at the Johns Hopkins Bloomberg School of Public Health; I worked on microfluidic approaches to investigate the olfaction of prostate cancer. Following my undergraduate work, I was a research intern under Prof. Guanghua Xiao at the Quantitative Biomedical Research Center at The University of Texas Southwestern Medical Center, and a research assistant under Prof. Qiwei Li in the Department of Mathematical Sciences at The University of Texas at Dallas; I worked on applying natural language processing methods to electronic health records to investigate lupus care, and a Bayesian shape clustering algorithm.
Conduct 1-on-1 consultations with students and researchers to support complex data, computational, and statistical inquiries. Collaboratively create peer-reviewed, published research guides and workshops. Supervisor: Ted Ellsworth
Conduct 1-on-1 consultations with graduate students requesting assistance with their writing projects. Supervisors: Ryan Wepler and Lauren Gonzalez
Affiliated with the Functional Neuroimaging and Bioinformatics Lab led by Prof. Justin Baker. Supervisors: Justin Baker and Einat Liebenthal
Construct a foundation model with applications in psychiatry by leveraging electronic health record and wearable device data.
Constructed a knowledge graph of clinical trial study reports using Graphlit, for use in GraphRAG. Improved prompt engineering for LLM-assisted structured content authoring of clinical trial study reports. Supervisors: Matt Renda and Shishir Jain