Tenure-track Assistant Professor · Biostatistics and Health Data Science
Indiana University School of Medicine
AI that turns real-world health data into fair, actionable evidence.
I'm a tenure-track Assistant Professor in the Department of Biostatistics and Health Data Science at Indiana University School of Medicine, an Investigator at the Regenstrief Institute's Clem McDonald Center for Biomedical Informatics, and an Adjunct Professor in Purdue University's Weldon School of Biomedical Engineering. I develop machine learning methods for electronic health records, medical images, and other real-world data: to model how diseases progress, to find patient subgroups, and to make sure the resulting models work fairly for everyone.
Before IU, I was a postdoctoral associate with Jiang Bian at the University of Florida. I received my Ph.D. in Computer Science from National Yang Ming Chiao Tung University in 2022, co-advised by Vincent S. Tseng and Gary G. Yen.
Research
Risk models that stay accurate and equitable across patient groups when they learn from real-world data.
Read more 02Graph and sequence models that find progression subtypes in EHRs and physiological signals and predict outcomes early.
Read more 03Patient-level computational models that forecast how health will change and test what-if treatment scenarios.
Read more 04Foundation models and interpretable learning for OCT and fundus images, including the eye as a window on dementia risk.
Read moreProjects
A multidisciplinary AI board for dementia risk assessment
An AI copilot that distills fragmented multimodal clinical data, including retinal imaging (OCT, OCTA, fundus), structured EHRs in the OMOP CDM, and clinical notes, into a single evidence-backed ADRD risk consensus through multi-agent deliberation.
Causal knowledge synthesis from the biomedical literature
Causal evidence about complex diseases is scattered across hundreds of thousands of papers. MedCausalAtlas uses large language models to extract causal relationships into evidence-traceable DAGs, so researchers can reason about intervention targets, find knowledge gaps, and generate testable hypotheses. Presented at ACM SIGKDD 2026.
News
Publications
A fair individualized polysocial risk score for identifying increased social risk in type 2 diabetes
Nature Communications, 15(1), 8653, 2024
Temporally Detailed Hypergraph Neural ODE for Disease Progression Modeling
The Fourteenth International Conference on Learning Representations (ICLR), 2026
Journal of Healthcare Informatics Research, 10(2), 317–339, 2026
Adaptation of a fair individualized polysocial risk score for hospitalization risk prediction
JAMIA Open, 9(4), ooag161, 2026
Snippet Policy Network for Multi-class Varied-length ECG Early Classification
IEEE Transactions on Knowledge and Data Engineering, 35(6), 6349–6361, 2022
Work with me
I'm looking for people who want to build AI methods for real-world health data, including EHRs, medical imaging, and multimodal data, with a focus on fairness, disease progression, and digital twins. Email me your CV and a short note on what you'd like to work on.