October 7, 2026
Amin Tajgardoon (SCI ‘17G, ‘25G) was selected as one of two runner-up finalists for the 2026 Edward H. Shortliffe Doctoral Dissertation Award by this year's AMIA Dissertation Awards Committee.
Tajgardoon received the award for his dissertation titled “Understanding and Quantifying Dataset Shift in Clinical Predictive Modeling.” His dissertation examines how dataset shift (changes in patient populations, clinical practices, or data collection) degrades clinical prediction models and evaluates unsupervised divergence metrics as early-warning signals. It finds that different shift types (case-mix, conceptual, and demographic) harm performance and fairness in distinct ways, identifies the most reliable metrics and thresholds for triggering model review, and proposes an ensemble alarm strategy for trustworthy, ongoing model monitoring.
Each year, the AMIA Edward H. Shortliffe Doctoral Dissertation Award honors the best doctoral dissertations that advance informatics science in one or more biomedical fields.
Please join us in congratulating Tajgardoon on this achievement!