Browse by author
Lookup NU author(s): Dr Judith HarrisonORCiD, Alex Robertson, Dr Huizhi LiangORCiD, Professor John-Paul TaylorORCiD
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
© 2025 MA Healthcare Ltd. All rights reserved.Dementia poses major challenges to healthcare worldwide. Traditional diagnostics rely on lengthy assessments, and access to specialist clinicians is limited. Large language models (LLMs), like Generative Pre-trained Transformer 4 (GPT-4) present new avenues for enhancing dementia diagnosis and care through advanced language processing. Whilst research into their applications is in its infancy, LLMs can harness vast datasets and powerful algorithms, with significant potential to enhance diagnostic accuracy in dementia, monitor symptom progression, and provide personalised care recommendations. While dementia serves as the primary example, the ethical and practical considerations discussed are applicable to the wider use of LLMs across different areas of medicine. This review explores the prospect of LLMs transforming dementia management and addresses the ethical and practical considerations involved.
Author(s): Harrison JR, Tang SL, Burston B, Robertson A, Liang H, Taylor JP
Publication type: Review
Publication status: Published
Journal: British Journal of Hospital Medicine
Year: 2025
Volume: 86
Issue: 10
Pages: epub ahead of print
Print publication date: 25/10/2025
Online publication date: 09/10/2025
Acceptance date: 04/02/2025
ISSN (print): 1750-8460
ISSN (electronic): 1759-7390
Publisher: MA Healthcare Ltd
URL: https://doi.org/10.12968/hmed.2024.0666
DOI: 10.12968/hmed.2024.0666
PubMed id: 41134177
Data Access Statement: All the data of this study are included in this article.
