cv

Education, appointments, funding, honors, editorial service, and patents. The full Harvard-format CV is available as a PDF above.

Contact Information

Name Xiang Li
Professional Title Assistant Professor of Radiology, Harvard Medical School
Email xiangli.shaun@gmail.com
Location 399 Revolution Dr., 11th Floor, Somerville, Massachusetts MA 02145

Professional Summary

Assistant Professor in the Department of Radiology at Massachusetts General Hospital and Harvard Medical School, and Affiliate Faculty at the Harvard Kempner Institute. Research spans medical foundation models, generative AI, expert-machine alignment, medical informatics, and counterfactual causal inference.

Experience

  • 2023 - present

    Cambridge, MA

    Affiliate Faculty Member
    Harvard University — Kempner Institute for Natural and Artificial Intelligence
  • 2023 - present

    Boston, MA

    Assistant Professor, Department of Radiology
    Harvard Medical School
  • 2019 - present

    Boston, MA

    Research Staff, Department of Radiology
    Massachusetts General Hospital
  • 2019 - 2023

    Boston, MA

    Instructor, Department of Radiology
    Harvard Medical School
  • 2016 - 2019

    Boston, MA

    Research Fellow, Radiology
    Harvard Medical School and Massachusetts General Hospital
    • Mentors: Associate Prof. Quanzheng Li and Distinguished Prof. James H. Thrall.

Education

  • 2009 - 2016

    Athens, GA, USA

    PhD
    University of Georgia
    Computer Science
    • Advisor: Distinguished Professor Tianming Liu (AIMBE Fellow).
    • Outstanding Graduate Dissertation/Thesis, University of Georgia, 2016.
  • 2002 - 2006

    Shanghai, China

    B.E.
    Shanghai Jiao Tong University
    Automation

Current Research Funding

  • 2026–2030 · 360° Collaborative Hub for AI in TMD Research (360° Chat-TMD) — NIH U54, 1U54DE035412-01. Co-Investigator; direct cost $5,584,812 over 5 years. Establishing a Center of Excellence in TMD research, training and outreach.
  • 2026–2029 · UniBrain: An End-to-End Unified Toolkit for Comprehensive Neuroimaging Analysis — NIH R01, 1R01EB038264-01A1. Multiple PI; direct cost $2,525,515 over 4 years. Developing AI-empowered technologies for neuroimaging analysis.
  • 2026–2029 · Automated PET Report Generation with Visual Grounding — NIH R01, 1R01EB038426-01. Subcontract PI (MGH); direct cost $2,304,244 over 4 years. Visually grounded PET report generation using federated learning.

Honors and Awards

  • 2025
    Editor's Choice Article
    Medical Physics

    Fine-tuning open-source large language models to improve their performance on radiation oncology tasks.

  • 2025
    Featured Article
    IEEE Reviews in Biomedical Engineering

    Artificial General Intelligence for Medical Imaging Analysis.

  • 2024
    Distinguished Paper Award
    American Medical Informatics Association (AMIA)

    MKRAG: Medical Knowledge Retrieval Augmented Generation for Medical Question Answering.

  • 2024
    Research Scholar Program Award
    Google Inc.

    Tailoring Large Language Models for the Diagnosis and Management of Late-life Depression Patients.

  • 2022
    Best Paper Award
    IEEE Transactions on Radiation and Plasma Medical Sciences

    Deep Learning-Based Image Segmentation on Multimodal Medical Imaging.

  • 2021
    MGH Thrall Innovation Grants Award
    Massachusetts General Hospital

    Chest Radiographs-based Lung Cancer Screening by the DeepProjection Technique.

  • 2020
    Best Paper Award
    IEEE International Symposium on Biomedical Imaging (ISBI)

    ASCNet: Adaptive-Scale Convolutional Neural Networks for Multi-Scale Feature Learning.

  • 2018
    Most Cited Article
    Journal of the American College of Radiology

    Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls.

  • 2016
    Outstanding Graduate Dissertation/Thesis
    University of Georgia
  • 2015
    Cover and Feature Paper
    IEEE Transactions on Biomedical Engineering

    Holistic atlases of functional networks and interactions.

  • 2013
    Best Student Paper Award
    IEEE International Symposium on Biomedical Imaging (ISBI)

    Discovering Common Functional Connectomics Signatures.

  • 2011
    Best Student Paper Award
    IEEE International Symposium on Biomedical Imaging (ISBI)

    Brain State Change Detection via Fiber-centered Functional Connectivity Analysis.

Editorial Service

  • IEEE Transactions on Medical Imaging — Guest Associate Editor
  • Proceedings of the National Academy of Sciences (PNAS) — Ad-hoc Editor
  • Data Intelligence — Associate Editor
  • IEEE Transactions on Artificial Intelligence — Associate Editor
  • Meta-Radiology — Associate Editor
  • BMC Biomedical Engineering — Associate Editor
  • BMC Artificial Intelligence — Associate Editor
  • Connected Health and Telemedicine — Associate Editor
  • Frontiers in Oncology, Radiology, Neuroscience, and Cardiovascular Medicine — Associate Editor
  • Neurocomputing — Guest Editor, Special Issue on Recent Advancements in Foundation Models
  • IEEE Transactions on Neural Networks and Learning Systems — Guest Editor, Special Issue on Advancements in Foundation Models

Grant Review Service

  • Dutch Research Council (NWO), Open Technology Programme — Ad hoc Reviewer
  • Society of Nuclear Medicine and Molecular Imaging, Bradley-Alavi Student Fellowship — Study Section Member
  • NIH BRAIN Initiative, ZMH1 ERB-E (03) S — Ad hoc Reviewer
  • NIH, Academic-Industrial Partnerships PAR Panel, ZRG1 CTH-E (57) — Ad hoc Reviewer

Patents

  • System and Method for Tokenization of Anatomical Structures for Medical Imaging — anatomy-aware tokenization of 2D and 3D medical images as semantic disentangled token groups.
  • System and Method for Synthesizing 3D Volumetric Medical Images Using a Probabilistic-Based Machine Learning Model — pose-conditioned diffusion models synthesizing high-resolution 3D CT-like volumes from standard 2D radiographs.
  • Automated Detection and Management of Valvular Heart Disease Using Machine Learning (2024) — Aortic Stenosis Ensemble Risk Prediction (AS-ERP) model predicting length of stay and readmission from EMR data.