Faisal Mahmood, PhD, Associate Professor, Harvard Medical School
Associate Professor & Brigham and Women's Hospital
Multimodal Foundation Models for Biomedical AI(opens in new tab)
(opens in new tab)
New research connects pathology, genomics, proteomics, and clinical data to create a comprehensive view of patient health.
From TriPath's three-dimensional pathology models to HEST's integration of histology and spatial transcriptomics, his work advances understanding across biological scales. The journey culminates with APOLLO, a multimodal temporal foundation model trained on approximately 25 billion clinical records from more than seven million patients, combining clinical events, text, and medical imaging to forecast disease progression, treatment response, and patient outcomes with unprecedented breadth and precision.