PeopleofSuperintelligence

Researcher

Faisal Mahmood

Faisal Mahmood is a computational pathology researcher whose work develops generative and multimodal AI for pathology. With Nicholas Durr, he coauthored an endoscopy depth-estimation method in 2017. In 2020, he coauthored the TOAD preprint, a histology-based model for cancer-origin prediction. He is an Associate Professor of Pathology at Harvard Medical School and in the Division of Computational Pathology at Brigham and Women's Hospital and Massachusetts General Hospital; his group developed UNI, CONCH and PathChat for image-based and multimodal pathology. In January 2025, PathChat DX received FDA Breakthrough Device Designation; AstraZeneca acquired Modella AI, which he co-founded, in 2026; as of October 2026, he directs Mass General Brigham's AI Institute and is a scientific co-founder of Tremont AI.

Works
174
Citations
14,079
h-index
41
Badges

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

  1. 2017

    Endoscopy depth estimation

    With Nicholas J. Durr, he coauthors a 2017 paper that uses a deep convolutional neural network and conditional random field to estimate depth from conventional endoscopy images. The method reconstructs colon-surface topography from a single image.

  2. April 2019

    The Mahmood Lab opens

    In April 2019, he opens the Mahmood Lab and advertises postdoctoral fellowships.

  3. 2020

    TOAD for cancer origins

    In 2020, he coauthors a preprint introducing TOAD, a deep-learning algorithm that uses routine histology slides to provide a differential diagnosis for cancers of unknown primary.

  4. By 2024

    Associate Professor at Harvard

    By 2024, he is an Associate Professor of Pathology at Harvard Medical School and the Division of Computational Pathology at Brigham and Women's Hospital and Massachusetts General Hospital. He holds a PhD in Biomedical Imaging from OIST and previously trained as a postdoctoral fellow in Johns Hopkins University's Department of Biomedical Engineering. His research focuses on developing generative and multimodal AI for pathology applications.

  5. 2024

    UNI and CONCH models

    In 2024, Mass General Brigham introduces UNI and CONCH, two foundation models for computational pathology, with Mahmood as a corresponding author. UNI learns from pathology images; CONCH combines pathology images and language and is trained on more than 1.17 million histopathology image-text pairs.

  6. 2024

    PathChat for pathology

    In 2024, he coauthors PathChat, a vision-language assistant for human pathology. It combines a pathology encoder with a pretrained large language model and fine-tunes the system on more than 456,000 visual-language instructions.

  7. January 2025

    PathChat DX gets designation

    In January 2025, PathChat DX, an extension of PathChat developed in the Mahmood Lab, receives FDA Breakthrough Device Designation.

  8. As of October 2026

    Harvard and Tremont AI roles

    In 2026, AstraZeneca acquires Modella AI, which he co-founded. As of October 2026, he is Associate Professor of Pathology at Harvard Medical School and Brigham and Women's Hospital and the inaugural Director of the Mass General Brigham AI Institute. He is also a scientific co-founder of Tremont AI.

Career

  1. NowHarvard UniversityAssociate Professor of Pathology
  2. NowMass General BrighamAssociate Professor of Pathology
  3. NowMass General BrighamInaugural Director
  4. NowTremont AIScientific Co-founder
  5. 2021–2026Massachusetts Institute of Technology
  6. From 2019Harvard UniversityFounder of the Mahmood Lab
  7. 2019–2026Broad Institute
  8. Johns Hopkins UniversityPostdoctoral Fellow
  9. Modella AICo-founder

Education

  1. Okinawa Institute of Science and Technology Graduate UniversityPhD in Biomedical Imaging

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Faisal Mahmood yet. Send us a free-licensed photo.

  1. Deep Learning and Conditional Random Fields-based Depth Estimation and Topographical Reconstruction from Conventional Endoscopy arxiv.org. Checked 27 October 2026.
  2. Mahmood Lab – Computational Pathology – Computational and Quantitive Pathology at Harvard faisal.ai. Checked 27 October 2026.
  3. Deep Learning-based Computational Pathology Predicts Origins for Cancers of Unknown Primary arxiv.org. Checked 27 October 2026.
  4. Faisal Mahmood, PhD - 2024 World Medical Innovation Forum 2024.worldmedicalinnovation.org. Checked 27 October 2026.
  5. Mass General Brigham Researchers Develop AI Foundation Models to Advance Pathology news.massgeneralbrigham.org. Checked 27 October 2026.
  6. A multimodal generative AI copilot for human pathology nature.com. Checked 27 October 2026.
  7. Modella AI: Generative AI Co-Pilot PathChat Receives FDA Breakthrough Device Designation tissuepathology.com. Checked 27 October 2026.
  8. Team: Tremont AI tremont.ai. Checked 27 October 2026.