Page 22 - AIH-1-2

Artificial Intelligence in Health

REVIEW ARTICLE
LLMs-Healthcare: Current applications and

challenges of large language models in various
medical specialties

Ummara Mumtaz , Awais Ahmed , and Summaya Mumtaz *
1
2
1
1 Department of Information Technology, University of the Cumberlands, Williamsburg, Kentucky,
United States of America
2 Department of Gynecology and Obstetrics, University of Concepción, Concepción, Chile

Abstract

The purpose of this review is to provide a comprehensive overview of the latest
advancements in utilizing large language models (LLMs) in the health-care sector,
emphasizing their transformative impact across various medical domains. LLMs
have become pivotal in supporting healthcare, including physicians, health-care
providers, and patients. Our review provides insight into the applications of LLMs in
healthcare, specifically focusing on diagnostic and treatment-related functionalities.
We shed light on how LLMs are applied in cancer care, dermatology, dental care,
neurodegenerative disorders, and mental health, highlighting their innovative
contributions to medical diagnostics and patient care. Throughout our analysis,
we explore the challenges and opportunities associated with integrating LLMs in
healthcare, recognizing their potential across various medical specialties despite
existing limitations. In addition, we offer an overview of handling diverse data types
*Corresponding author:
Summaya Mumtaz within the medical field.
(summaya.mumtaz@gmail.com)
Citation: Mumtaz U, Ahmed A, Keywords: Large language models; Medical specialties; Cancer; Mental health;
Mumtaz S. LLMs-Healthcare: Healthcare; Diagnosis and treatments; Clinical notes; Dermatology
Current applications and challenges
of large language models in various
medical specialties. Artif Intell
Health. 2024;1(2): 16-28.
doi: 10.36922/aih.2558 1. Introduction
Received: December 28, 2023
The field of artificial intelligence (AI) has undergone a remarkable evolution in recent
Accepted: February 23, 2024 years, with significant advancements, particularly noticeable in natural language
Published Online: April 2, 2024 processing (NLP) and the development of large language models (LLMs). These models
represent a paradigm shift in AI’s capability to understand, generate, and interact using
Copyright: © 2024 Author(s).
This is an Open-Access article human language. At their foundation, LLMs are complex algorithms trained on vast,
1
distributed under the terms of the text-based documents and datasets. Such extensive training allows them to recognize
Creative Commons Attribution patterns adeptly, predict subsequent words in a sentence, and generate coherent,
License, permitting distribution,
and reproduction in any medium, contextually relevant text for the specified inputs, often called prompts within the NLP
provided the original work is community. This ability demonstrates the technical prowess of LLMs and signifies their
properly cited. potential to revolutionize how machines understand and process human language.
Publisher’s Note: AccScience One of the most prominent features of LLMs is their proficiency in processing and
Publishing remains neutral with analyzing large volumes of text rapidly and accurately, a capability that far surpasses
regard to jurisdictional claims in 2
published maps and institutional human potential in speed and efficiency. This quality makes them indispensable in
affiliations. areas requiring the analysis of extensive data sets. They are also known as “few-shot”

Volume 1 Issue 2 (2024) 16 doi: 10.36922/aih.2558