Academic Profile

Thomas Buckley

PhD Student, Harvard Medical School

thomas_buckley [at] g [dot] harvard [dot] edu

[2025-04-04] Interviewed on JAMA+ AI Podcast about LLMs for Clinical Reasoning
[2025-03-31] My Research on Llama 3.1 for Complex Diagnosis Featured in The Harvard Crimson and Harvard Medicine News
[2024-11-06] Awarded inaugural Dunleavy Fellowship for Clinical AI at Harvard Medical School
[2024-05-06] Presented research on Accuracy of Vision-Language Models on Challenging Medical Cases at SAIL 2024
ABOUT

I am a PhD student in the inaugural AI in Medicine Program, in the Department of Biomedical Informatics at Harvard Medical School, advised by Prof. Arjun Manrai. My research focuses on developing and evaluating large language models for clinical reasoning and diagnosis, as well as AI for automated scientific discovery.

PUBLICATIONS
First Author
╭──[ LATEST ]──╮

Comparison of frontier open-source and proprietary large language models for complex diagnoses

TA Buckley, B Crowe, REE Abdulnour, A Rodman, AK Manrai

JAMA Health Forum 6 (3), e250040-e250040 • 2025

[DOI/Link]

Do Language Models Think Like Doctors?

LG McCoy, R Swamy, N Sagar, M Wang, J Cao, S Bacchi, N Fong, ...

medRxiv, 2025.02.11.25321822 • 2025

[DOI/Link]

Projected changes in statin and antihypertensive therapy eligibility with the AHA PREVENT cardiovascular risk equations

JA Diao, I Shi, VL Murthy, TA Buckley, CJ Patel, E Pierson, RW Yeh, ...

Jama 332 (12), 989-1000 • 2024

[DOI/Link]

Superhuman performance of a large language model on the reasoning tasks of a physician

PG Brodeur, TA Buckley, Z Kanjee, E Goh, EB Ling, P Jain, S Cabral, ...

arXiv preprint arXiv:2412.10849 • 2024

[DOI/Link]

Multimodal Foundation Models Exploit Text to Make Medical Image Predictions

T Buckley, JA Diao, P Rajpurkar, A Rodman, AK Manrai

arXiv e-prints, arXiv: 2311.05591 • 2023

[DOI/Link]

Artificial intelligence vs clinician performance in estimating probabilities of diagnoses before and after testing

A Rodman, TA Buckley, AK Manrai, DJ Morgan

JAMA Network Open 6 (12), e2347075-e2347075 • 2023

[DOI/Link]