Use of artificial intelligence large language models as a clinical tool in rehabilitation medicine: a comparative test case
DOI:
https://doi.org/10.2340/jrm.v55.13373Keywords:
Artificial Intelligence, Large Language Models, Rehabilitation Prescriptions, International Classification of Functioning, Disability, and Health (ICF) codesAbstract
Objective: To explore the potential use of artificial intelligence language models in formulating rehabilitation prescriptions and International Classification of Functioning, Disability and Health (ICF) codes.
Design: Comparative study based on a single case report compared to standard answers from a textbook.
Subjects: A stroke case from textbook.
Methods: Chat Generative Pre-Trained Transformer-4 (ChatGPT-4)was used to generate comprehensive medical and rehabilitation prescription information and ICF codes pertaining to the stroke case. This information was compared with standard answers from textbook, and 2 licensed Physical Medicine and Rehabilitation (PMR) clinicians reviewed the artificial intelligence recommendations for further discussion.
Results: ChatGPT-4 effectively formulated rehabilitation prescriptions and ICF codes for a typical stroke case, together with a rationale to support its recommendations. This information was generated in seconds. Compared with standard answers, the large language model generated broader and more general prescriptions in terms of medical problems and management plans, rehabilitation problems and management plans, as well as rehabilitation goals. It also demonstrated the ability to propose specified approaches for each rehabilitation therapy. The language model made an error regarding the ICF category for the stroke case, but no mistakes were identified in the ICF codes assigned.
Conclusion: This test case suggests that artificial intelligence language models have potential use in facilitating clinical practice and education in the field of rehabilitation medicine.
Downloads
References
OpenAI. OpenAI homepage. 2023. Available from: https://openai.com/product/gpt-4
Lee P, Bubeck S, Petro J. Benefits, limits, and risks of GPT-4 as an AI chatbot for medicine. N Engl J Med 2023; 388: 1233–1239. DOI: 10.1056/NEJMsr2214184. DOI: https://doi.org/10.1056/NEJMsr2214184
Moor M, Banerjee O, Abad ZSH, Krumholz HM, Leskovec J, Topol EJ, et al. Foundation models for generalist medical artificial intelligence. Nature 2023; 616: 259–265. DOI: 10.1038/s41586-023-05881-4. DOI: https://doi.org/10.1038/s41586-023-05881-4
Patel SB, Lam K. ChatGPT: the future of discharge summaries? Lancet Digit Health 2023; 5: e107–e108. DOI: 10.1016/S2589-7500(23)00021-3. DOI: https://doi.org/10.1016/S2589-7500(23)00021-3
Zhou Z, Wang X, Li X, Liao L. Is ChatGPT an evidence-based doctor? Eur Urol 2023; 10.1016/j.eururo.2023.03.037. DOI: 10.1016/j.eururo.2023.03.037. DOI: https://doi.org/10.1016/j.eururo.2023.03.037
Grunebaum A, Chervenak J, Pollet SL, Katz A, Chervenak FA. The exciting potential for ChatGPT in obstetrics and gynecology. Am J Obstet Gynecol 2023; 10.1016/j.ajog.2023.03.009. DOI: 10.1016/j.ajog.2023.03.009. DOI: https://doi.org/10.1016/j.ajog.2023.03.009
Yamada S. Tips in assessment of rehabilitation learn from 100 cases; 2013.
Tsuji T, Liu M, Sonoda S, Domen K, Chino N. The stroke impairment assessment set: its internal consistency and predictive validity. Arch Phys Med Rehabil 2000; 81: 863–868. DOI: 10.1053/apmr.2000.6275. DOI: https://doi.org/10.1053/apmr.2000.6275
Mukaino M, Prodinger B, Yamada S, Senju Y, Izumi SI, Sonoda S, et al. Supporting the clinical use of the ICF in Japan – development of the Japanese version of the simple, intuitive descriptions for the ICF Generic-30 set, its operationalization through a rating reference guide, and interrater reliability study. BMC Health Serv Res 2020; 20: 66. DOI: 10.1186/s12913-020-4911-6. DOI: https://doi.org/10.1186/s12913-020-4911-6
Senju Y, Mukaino M, Prodinger B, Selb M, Okouchi Y, Mizutani K, et al. Development of a clinical tool for rating the body function categories of the ICF generic-30/rehabilitation set in Japanese rehabilitation practice and examination of its interrater reliability. BMC Med Res Methodol 2021; 21: 121. DOI: 10.1186/s12874-021-01302-0. DOI: https://doi.org/10.1186/s12874-021-01302-0
Downloads
Additional Files
Published
How to Cite
License
Copyright (c) 2023 Liang Zhang, Syoichi Tashiro, Masahiko Mukaino, Shin Yamada
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
All digitalized JRM contents is available freely online. The Foundation for Rehabilitation Medicine owns the copyright for all material published until volume 40 (2008), as from volume 41 (2009) authors retain copyright to their work and as from volume 49 (2017) the journal has been published Open Access, under CC-BY-NC licences (unless otherwise specified). The CC-BY-NC licenses allow third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for non-commercial purposes, provided proper attribution to the original work.
From 2024, articles are published under the CC-BY licence. This license permits sharing, adapting, and using the material for any purpose, including commercial use, with the condition of providing full attribution to the original publication.
