Bitterman Lab

ABOUT US

Learn more about us.

The Bitterman Lab is a research group within the AI in Medicine Program at Mass General Brigham and the Department of Radiation Oncology at Brigham and Women’s Hospital/Dana-Farber Cancer Institute, Harvard Medical School. We advance the science and clinical translation of foundation models for healthcare. Our work spans algorithm development, evaluation, and real-world deployment, with an emphasis on safety, reliability, and clinical utility. As a multidisciplinary team of physician-scientists and computer scientists, we ground methodological innovations in high-impact medical applications.

Our areas of active research include:

• Foundation model evaluation, alignment, and oversight: Developing principled evaluation frameworks, alignment objectives, stress-testing methodologies, and oversight architectures to characterize and control foundation model behavior in healthcare settings.
• Oncology information extraction from EHRs: Building NLP pipelines and representation learning methods to derive structured signals from heterogeneous cancer clinical texts for downstream modeling and causal inference.
• AI for patient-facing education and clinical decision-support in oncology: Building AI-powered systems that produce accurate, personalized educational materials and integrate actionable insights into clinician workflows.
• Translational AI and clinical evaluation: Executing clinical studies of AI systems, analyzing human–AI interaction and failure modes, and developing deployment frameworks that satisfy regulatory, ethical, and workflow constraints.

Learn more about our Research, Publications, and Team.