Jamieson Lab

Shipping AI Solutions - Safely!

We develop and deploy state-of-the-art AI systems to solve real-world clinical, medical education, and research problems.
We embrace dynamic, interdisciplinary collaborations with creative and ambitious leaders who share our vision for how AI will transform health care and science.

[PATENTS PENDING! ✌️😁]

See our research

Get in touch!

News + recruiting

Come build the future with us!

https://ut-real-ai-project-maples.com/

MAY-2026: SAIL 2026 — Puerto Rico

Tra Ngo will present "Deploying a Multimodal AI Platform for Automated OSCE Grading in Medical Education" at SAIL 2026, the Symposium on Artificial Intelligence for Learning Health Systems (May 5–8, Río Grande, Puerto Rico).

APR-2026: Featured in UTSW Scientific Report

Our OSCE grading work is featured in Discovery@UTSW 2026, highlighting how AI replaced 91% of human grading while reducing turnaround from weeks to days. Read the full article

APR-2026: UT-REAL Health AI Grant Awarded

We received a $300,000 UT System grant to scale our AI grading platform across six UT medical schools over 18 months. We will validate AI-human concordance at each site and develop shared governance frameworks for AI in medical education. https://ut-real-ai-project-maples.com/

APR-2026: Dallas Innovates AI 75

Andrew Jamieson named to the Dallas Innovates AI 75 for 2026.

APR-2026: PCT International Patent Filed

Our second patent (UTSD 4539) entered the international phase: PCT/US2026/022629, covering automated audio-visual assessment of clinical encounters.

APR-2026: AIMW 2026 Workshop — Seattle

Presenting "From Research to Rollout: Lessons from Deploying AI-Based OSCE Assessment in Production" at AIMW26, April 20.

FEB-2026: AI for GME Program Review — ACGME Annual Educational Conference

Michael Holcomb presented "Leveraging AI: Automating Annual Action Plan Scoring as a Use Case" at the 2026 ACGME Annual Educational Conference ("Meaning in Medicine," San Diego, Feb 20–22). The work — a partnership with UTSW GME leadership (Drs. Scielzo, Green, and Scott) — uses a virtual panel of example-conditioned LLM personas to generate consensus evaluations of 200+ UTSW residency and fellowship program action plans, lifting pass/fail agreement from 46% (zero-shot) to 91% (virtual panel). ACGME Learn course · Video (6 min) · Open-source code.

FEB-2026: Published in JMIR AI

Kang S, Holcomb MJ, Hein D, Shakur AH, Dalton TO, Jamieson AR. "Physical Examination Identification in Medical Education Videos: Zero-Shot Multimodal AI With Temporal Sequence Optimization." JMIR AI. 2025. DOI

JAN-2026: Multimodal Physical Exam Grading Preprint

Kang S, Holcomb MJ, Shakur AH, Hein D, Ngo HT, Schuler H, Jarrett PC, Dalton TO, Jamieson AR. "Automated Assessment of OSCE Physical Exams using Multimodal AI." medRxiv 2026. DOI. Under review.

MAY-2025: Published in npj Digital Medicine

Hein D et al. "Iterative refinement and goal articulation to optimize LLMs for clinical information extraction." npj Digital Medicine 8, 301. 2025.

MAY-2025: Full OSCE Grading From Transcripts

Now grading entire OSCEs — communications and physical exams — from transcripts on a new set of pre-clerkship cases developed by Dr. Dalton.

APR-2025: Multimodal OSCE Preprint

Kang S et al. Automatic Physical Examination Segmentation within OSCE Videos. medRxiv 2025.

NOV-2024: Published in NEJM AI

Jamieson AR et al. Rubrics to Prompts: Assessing Medical Student Post-Encounter Notes with AI. NEJM AI. 2024.

OCT-2024: Production OSCE Grading Complete

Completed another year of AI-powered OSCE note grading at the SimCenter.

OCT-2024: Transcript Analysis Preprint

Shakur AH et al. Large Language Models for Medical OSCE Assessment: A Novel Approach to Transcript Analysis. arXiv. 2024.