ASCI / Emerging-Generation Awards, 2024

Theodore Pak, MD, PhD is an infectious diseases physician-scientist at Massachusetts General Hospital and Harvard Medical School. Having trained in bioinformatics and software engineering, his current research interest is the use of machine learning and artificial intelligence techniques on routinely collected clinical data to improve prevention and management of high-impact infectious diseases, including sepsis and COVID-19. He has published widely on his research, with first and co-first authorships in JAMA Internal Medicine, Clinical Infectious Diseases, Genome Medicine, and Molecular Systems Biology.

He grew up in New York, completed undergraduate studies in Biochemical Sciences and Computer Science at Harvard College, and worked professionally as a software engineer. He then matriculated into the MD/PhD program at the Icahn School of Medicine at Mount Sinai. For his PhD dissertation, he built bioinformatics tools enabling surveillance of several hospital outbreak pathogens in New York City, including Staphylococcus aureus and influenza A, using whole genome sequencing. He also performed immune profiling analyses of dengue and chikungunya infections and analyzed the costs of hospital-onset Clostridioides difficile using machine learning models built from diverse electronic medical record data. His PhD work was recognized with a National Research Service Award from NIH/NIAID (F30AI122673).

After medical school, he completed Internal Medicine residency at Massachusetts General Hospital and remains there as an Infectious Diseases fellow. His experiences as a front-line physician throughout the COVID-19 pandemic sparked his interest in analyses of large-scale EHR data that can provide evidence informing improved outcomes for critically ill patients. He now works in the Population Medicine research group of Drs. Michael Klompas and Chanu Rhee, with ongoing investigations of associations between time-to-antibiotics and sepsis outcomes in large multi-hospital cohorts, COVID-19 infection control strategies, and the use of large language models and other artificial intelligence methods to power massive retrospective clinical studies.