Click to View Presentation

Presentation Time: 08:30 AM - 11:30 AM

Abstract Keywords: Artificial Intelligence, Large Language Models (LLMs), Data Mining
Working Group: Knowledge Discovery and Data Mining Working Group
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Large Language Models (LLMs) have revolutionized healthcare applications, from clinical decision support to biomedical research. While commercial models like GPT-4 and Claude have demonstrated strong performance, their closed-source nature limits transparency, reproducibility, and domain-specific customization. In contrast, open-source LLMs such as LLaMA 3, DeepSeek, and Mistral offer greater flexibility, control over data privacy, and opportunities for healthcare-specific fine-tuning. However, practical adoption presents challenges related to performance, multimodal integration, computational demands, and regulatory compliance.

This half-day workshop, organized by the AMIA KDDM, NLP, and KRS Working Groups, explores the opportunities and challenges of using open-source LLMs in healthcare. The program includes a keynote on benchmarking open- vs. closed-source models, a panel discussion on implementation challenges, brief research talks on multimodal LLMs, and a hands-on tutorial using LangChain and LangGraph to build LLM-powered tools.

Participants will gain insights into the trade-offs between model types, best practices for HIPAA-compliant deployment, and the development of interpretable, multimodal AI systems. This workshop offers both strategic perspectives and practical tools for researchers, clinicians, and informaticians advancing open-source AI in healthcare.

Speakers:
Kaleb Smith, PhD
NVIDIA

Zhe He, PhD, FAMIA
Florida State University

Zhiyong Lu, PhD
National Library of Medicine, NIH

Yonghui Wu, PhD
University of Florida

Mattia Prosperi, PhD, FAMIA
University of Florida

Balu Bhasuran, Ph.D
Florida State University

Hong Yu, PhD
University of Massachusetts Lowell

Authors:
Zhe He, PhD, FAMIA - Florida State University; Nansu Zong, Ph.D. - Mayo Clinic; Rui Zhang, PhD, FAMIA, FACMI - University of Minnesota, Twin Cities; Hilda Klasky, FAMIA - Oak Ridge National Laboratory (ORNL) - UT Battelle; Amelia Averitt, MPH, MA, PhD - Regeneron Pharmaceuticals; Humayera Islam, PhD - The University of Chicago; Carl Yang, Ph.D.;

---