PeopleofSuperintelligence

Researcher

Weishen Pan

Weishen Pan is a machine-learning researcher at Weill Cornell Medicine who applies AI to clinical prediction and trial design. He earned a B.Eng. in 2015 and a Ph.D. in 2022 at Tsinghua University. He coauthored work on federated risk prediction, predictive patient subgroups, and EmulatRx, a multi-agent system for clinical trial design. In March 2026, he coauthored a multimodal machine-learning model for advanced heart failure. As of October 2026, he is an assistant professor of research in population health sciences and director of innovations at Weill Cornell's Institute of AI for Digital Health.

Works
26
Citations
244
h-index
8

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

  1. 2015–2022

    Tsinghua degrees

    He earns a B.Eng. at Tsinghua University in 2015 and completes a Ph.D. there in 2022.

  2. 2024

    Federated risk prediction

    In 2024, he coauthors an adaptive federated-learning framework that predicts sepsis and acute kidney injury from intensive-care records across multiple institutions.

  3. 2025

    Machine learning for subphenotypes

    In 2025, he coauthors Graph-Encoded Mixture Survival (GEMS), a machine-learning framework for identifying predictive patient subgroups from clinical data. Applied to advanced non-small cell lung cancer patients receiving first-line immune checkpoint inhibitor therapy, it identifies three groups linked to distinct patient characteristics and overall survival.

  4. 2025

    Federated target trial emulation

    He coauthors FL-TTE, a federated-learning framework that estimates treatment effects from electronic health records held at different institutions without sharing records. The 2025 study evaluates it on Alzheimer's disease and sepsis data.

  5. March 2026

    AI model for heart failure

    In March 2026, he coauthors a multimodal machine-learning model that combines cardiac ultrasound, blood-flow waveforms and electronic health records to identify advanced heart failure.

  6. July 2026

    Five-agent trial design

    In July 2026, he coauthors a study on EmulatRx, a five-agent AI system that uses real-world health records to design and evaluate clinical trials. Researchers can monitor its work and correct its steps.

  7. As of October 2026

    Assistant professor and director

    As of October 2026, he is an assistant professor of research in population health sciences at Weill Cornell Medicine and director of innovations at its Institute of AI for Digital Health.

Career

  1. NowCornell UniversityAssistant Professor of Research in Population Health Sciences
  2. NowCornell UniversityDirector of Innovations, Institute of AI for Digital Health
  3. 2019–2021State Key Laboratory of Intelligent Technology and Systems
  4. 2019–2021Beijing National Research Center for Information Science and Technology

Education

  1. 2022Tsinghua UniversityPhD
  2. 2015Tsinghua UniversityBEng

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 Weishen Pan yet. Send us a free-licensed photo.

  1. Pan, Weishen vivo.weill.cornell.edu. Checked 27 October 2026.
  2. An adaptive federated learning framework for clinical risk prediction with electronic health records from multiple hospitals pmc.ncbi.nlm.nih.gov. Checked 27 October 2026.
  3. Identification of predictive subphenotypes for clinical outcomes using real world data and machine learning nature.com. Checked 27 October 2026.
  4. Federated target trial emulation using distributed observational data for treatment effect estimation nature.com. Checked 27 October 2026.
  5. AI Tool Shows Promise in Diagnosing Advanced Heart Failure news.weill.cornell.edu. Checked 27 October 2026.
  6. AI Research Team Could Streamline Clinical Trial Design phs.weill.cornell.edu. Checked 27 October 2026.
  7. Team phs.weill.cornell.edu. Checked 27 October 2026.