Mohammad Asadi

PhD student, Electrical Engineering · Stanford University · Ashley Lab · STAI Lab

Mohammad Asadi, PhD student in Electrical Engineering at Stanford University

masadi [at] stanford.edu

I’m a PhD student in Electrical Engineering at Stanford, where I work in the Ashley Lab, advised by Euan Ashley and co-advised by Fei-Fei Li and Ehsan Adeli. I build multimodal models for cardiology that read ECG, echocardiograms, and cardiac MRI, and I study where frontier multimodal models fail. In MIRAGE, we found that they will confidently diagnose medical images they were never shown.

Before Stanford, I worked on human-motion generation at Samsung and on interpretable machine learning for education at EPFL, and I did my BSc in Electrical Engineering at Sharif University of Technology.

In The Press

MIRAGE was covered internationally. A few pieces:

See all coverage →

Not to be confused with other researchers named Mohammad Asadi, including the professor of chemical engineering at Illinois Institute of Technology. This site is about the Stanford Electrical Engineering PhD student working on AI for medicine.

News

Mar 23, 2026 Preprints for MIRAGE and MARCUS are online.
Feb 27, 2026 I served on the poster committee for Market Design in the Age of AI.
Nov 01, 2025 I joined the inaugural cohort of Amazon AI PhD Fellows and became a Stanford HAI Graduate Fellow.
Feb 17, 2024 Our project Healthiator won the “Smartest AI Agent” prize at TreeHacks.

Selected Publications

  1. MIRAGE: The Illusion of Visual Understanding
    MIRAGE: The Illusion of Visual Understanding
    Mohammad Asadi*, Jack W. O’Sullivan*, Fang Cao, Tahoura Nedaee, Kamyar Rajabalifardi, Fei-Fei Li†, Ehsan Adeli†, and Euan Ashley†
    arXiv preprint, 2026
    Abs arXiv Code Website
  2. MARCUS: An Agentic, Multimodal Vision-Language Model for Cardiac Diagnosis and Management
    MARCUS: An Agentic, Multimodal Vision-Language Model for Cardiac Diagnosis and Management
    Mohammad Asadi*, Jack W. O’Sullivan*, Lennart Elbe, Akshay Chaudhari, Tahoura Nedaee, Francois Haddad, Michael Salerno, Fei-Fei Li, Ehsan Adeli, Rima Arnaout, and Euan A. Ashley
    arXiv preprint, 2026
    Abs arXiv Code Website
  3. Deterministic Hallucination Detection in Medical VQA via Confidence-Evidence Bayesian Gain
    Deterministic Hallucination Detection in Medical VQA via Confidence-Evidence Bayesian Gain
    Mohammad Asadi, Tahoura Nedaee, Jack W. O’Sullivan, Euan Ashley, and Ehsan Adeli
    arXiv preprint, 2026
    arXiv
  4. EchoAtlas: A Conversational, Multi-View Vision-Language Foundation Model for Echocardiography Interpretation and Clinical Reasoning
    EchoAtlas: A Conversational, Multi-View Vision-Language Foundation Model for Echocardiography Interpretation and Clinical Reasoning
    Chieh-Ju Chao, Mohammad Asadi, and others
    medRxiv preprint, 2026
    HTML
  5. EchoGraph: Automated Quality Assessment of Echocardiography Reports
    EchoGraph: Automated Quality Assessment of Echocardiography Reports
    Chieh-Ju Chao, Jean-Benoit Delbrouck, Mohammad Asadi, and others
    npj Digital Medicine, 2025
    HTML
  6. Synthetic Hands Meet Legacy Data: A Synthetic Dataset for Structured, Controllable, and Multimodal Evaluation
    Synthetic Hands Meet Legacy Data: A Synthetic Dataset for Structured, Controllable, and Multimodal Evaluation
    Menghe Zhang, Haley M. So, Mohammad Asadi, and others
    In ICCV 2025 Workshops (DataCV), 2025
    HTML
  7. Ripple: Concept-Based Interpretation for Raw Time Series Models in Education
    Ripple: Concept-Based Interpretation for Raw Time Series Models in Education
    Mohammad Asadi, Vinitra Swamy, Jibril Frej, Julien Vignoud, Mirko Marras, and Tanja Käser
    In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023
    arXiv Code

Patents

  • System and Method for End-to-End Pipeline for Photo-Realistic 3D Motion Generation. U.S. Patent Application (Samsung).
  • MARCUS: An Agentic, Multimodal Vision-Language Model for Cardiac Diagnosis. U.S. patent application, in preparation.