Harnessing LLMs in Healthcare: A Systematic Review of Performance, Suitability, and Implementation Risks
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The rapid integration of Large Language Models (LLMs) into healthcare is transforming clinical decision support, medical documentation, and patient interaction.This study systematically reviews recent advances in LLM architectures and their suitability for healthcare applications by comparing GPT-4, Med-PaLM 2, ClinicalBERT, and BioBERT.Literature published between 2022 and mid-2025 was analyzed from PubMed, IEEE Xplore, arXiv, and related repositories, focusing on model performance, benchmark results, and implementation challenges.Decoder-only models such as GPT-4 and Med-PaLM 2 exhibit superior performance in generative tasks like summarization and question-answering, while encoder-based models such as ClinicalBERT and BioBERT excel in extractive tasks, including classification and named-entity recognition.Multilingual extensions such as GPT-4o and mBERT demonstrate emerging potential for cross-lingual healthcare applications.Despite these advances, widespread clinical adoption faces hurdles related to computational scalability, transparency, bias mitigation, and compliance with regulatory frameworks such as HIPAA.The review concludes with recommendations for developing lightweight, interpretable, and ethically aligned LLMs that can support reliable, safe, and equitable healthcare delivery.
Fields of Study & Subjects
Artificial Intelligence in Healthcare and Education
Machine Learning in Healthcare
Topic Modeling
Publication details
- Journal / source
- International Research Journal of Modernization in Engineering Technology and Science
- Publication date
- 30 March 2026
- Language
- EN