Exploring the Potential of Large Language Models in Addressing Challenges of Hospital Information Systems

Abstract

This study aims to examine the pain points encountered by healthcare providers in utilizing HIS and to explore the potential of LLMs in addressing these challenges through a structured proof-of-concept evaluation. The research objectives include identifying specific challenges through qualitative interviews with 6 healthcare professionals, assessing the potential utility of LLMs based on both literature and empirical findings, testing LLM performance on predefined patient case scenarios, and formulating recommendations for the integration of LLM technology into HIS. By addressing these research questions, this study seeks to contribute to the discourse on AI-driven advancements in healthcare informatics, offering insights into how LLMs may enhance data interoperability, accuracy, and clinical decision support within HIS frameworks.