ISSN (Print): 3007-6706 ISSN (Online): 3007-6706
International IT Journal of Research Official Publication of Octopus Publication, Hong Kong
Cover of April-June, 2025
research article

Generative AI in Health IT: Building Trustworthy LLMs for Clinical and Administrative Workflows

  • Srinivas Raghu Chilakamarri

Vol. 3 , Issue 2 (2025) · pp. 22-30

DOI: 10.64180/iitjr.322522

Abstract

Generative artificial intelligence, particularly large language models (LLMs), is fundamentally transforming healthcare delivery through enhanced clinical decision-making, medical documentation automation, and administrative workflow optimization. The global market for generative AI in healthcare reached USD 2.17 billion in 2024 and is projected to reach USD 23.56 billion by 2033, reflecting a compound annual growth rate of 35.17% through 2034. Medical-domain-specific models such as Med-PaLM 2 have achieved 86.5% accuracy on USMLE-style examinations, demonstrating clinical competency approaching specialist-level performance. Clinical documentation automation has reduced administrative burden by 72% while decreasing error rates by approximately 70%, with potential annual savings of USD 200–360 billion in the United States healthcare system. However, significant challenges persist, including hallucination rates of 1.47% in clinical documentation, demographic bias perpetuating health disparities across racial and ethnic groups, and incomplete regulatory frameworks governing AI-enabled medical devices.

Keywords: large language models generative artificial intelligence clinical decision support trustworthiness healthcare information technology medical artificial intelligence regulatory frameworks natural language processing hallucination mitigation health equity
0 views 0 downloads

How to Cite

Cite this article