Register here: https://cosmos.vastdata.com/clemson-edu-palmetto-2/
In this webinar, Becky Ligon, Scott Groel, and Carl Ehrett, will go over the Palmetto 2 high-performance computing cluster at Clemson University. They will cover the Palmetto 2 infrastructure, how jobs are run, and how it’s being used for AI research.
Dixizi "Krystal" Liu will then share the results of her research into using LLMs for proactive learning in healthcare. Full abstract below:
Title: Large Language Models for Proactive Learning in Healthcare: A comparative study of Traditional Incident Reporting and Resilience Engineering Approaches
Abstract: In this study, we explore the effectiveness of Large Language Models (LLMs) in enhancing the analysis of Incident Reporting Systems (IRS) and the Resilience Engineering Tool to Improve Patient Safety (RETIPS). Traditional IRSs often face challenges, such as underreporting and limited follow-up, which constrain learning about day-to-day adaptive patterns in healthcare. RETIPS, on the other hand, focuses on how systems adapt through proactive learning. Our first objective is to leverage LLMs to scale the analysis of lessons learned from incident reports and compare the qualitative differences in LLM-generated labels for IRS and RETIPS.
