Research outputs from the AI4ED project

Publications

Peer-reviewed papers, conference presentations, and preprints documenting our work on emergency department chart summarization, simulation, and workflow optimization.

AI4ED publications document the project across language-model evaluation, chart summarization, and agentic emergency department simulation. Together, they show how clinically grounded AI research can move from benchmarking and prototype development toward operational decision support.

AI4ED research output

EDSim: An Agentic Simulator for Emergency Department Operations

npj Digital Medicine Submitted

Authors: Jiajun Wu, Hutton Ledingham, Zirui Wang, Braden Teitge, Alexander Burn, Oussama Ouadihi, Mohamad Ghattas, Darin Vicaldo, Sergiu Cociuba, Megan Harmon, Tanvir Chowdhury, Zack Marshall, Tyler Williamson, Tracie Risling, Eddy Lang, Jiayu Zhou, Jessalyn Holodinsky, Steve Drew

Introduces an agentic emergency department simulator that combines LLM-driven clinician and patient behavior with data-driven operational modeling for workflow experiments.

Open publication

AI4ED research output

Fine-tuned large language models guided by physician feedback can improve patient chart summarization for emergency departments

ICEM 2025 – International Conference on Emergency Medicine Conference Presentation

Authors: Jiajun Wu, Hanzhe Wei, Braden Teitge, Jessalyn Holodinsky, Kyle Exner, Steve Drew

Demonstrates how clinician-in-the-loop fine-tuning boosts the quality and utility of ED chart summaries.

Open publication

AI4ED research output

Small Language Models for Emergency Departments Decision Support: A Benchmark Study

Accepted to 2025 IEEE International Conference on Autonomous and Trusted Computing (ATC 2025) Conference Paper

Authors: Zirui Wang, Jiajun Wu, Braden Teitge, Jessalyn Holodinsky, Steve Drew

Evaluates compact models for ED decision-support tasks, highlighting trade-offs between accuracy, latency, and deployability.

Open publication

AI4ED research output

Dual-stage and Lightweight Patient Chart Summarization for Emergency Physicians

Accepted at IEEE Annual Congress on Artificial Intelligence of Things (IEEE AIoT) 2025 Conference Paper

Authors: Jiajun Wu, Swaleh Zaidi, Braden Teitge, Henry Leung, Jiayu Zhou, Jessalyn Holodinsky, Steve Drew

Introduces a two-step summarization pipeline that balances comprehension with speed for frontline clinicians.

Open publication