Tag
This review paper proposes a unified framework for intervention-aware disease trajectory modeling in clinical AI, addressing static prediction failures by incorporating treatment confounder feedback and informative observation patterns.
DT-Transformer is a foundation model trained on 57.1 million structured EHR entries from 1.7 million patients across 11 hospitals in the Mass General Brigham health system, achieving strong discrimination for next-event prediction across 896 disease categories.