Abstract
Objectives
This study aimed to elucidate effective methodologies for utilizing the generative artificial intelligence (AI) system, namely the Chat Generative Pre-trained Transformer (ChatGPT), in improving clinical reasoning abilities among clinicians.
Methods
We conducted a comprehensive exploration of the capabilities of ChatGPT, emphasizing two main areas: (1) efficient utilization of ChatGPT, with a focus on application and language selection, input methodology, and output verification; and (2) specific strategies to bolster clinical reasoning using ChatGPT, including self-learning via simulated clinical case creation and engagement with published case reports.
Results
Effective AI-based clinical reasoning development requires a clear delineation of both system roles and user needs. All outputs from the system necessitate rigorous verification against credible medical resources. When used in self-learning scenarios, capabilities of ChatGPT in clinical case creation notably enhanced disease comprehension.
Conclusions
The efficient use of generative AIs, as exemplified by ChatGPT, can impressively enhance clinical reasoning among medical professionals. Adopting these cutting-edge tools promises a bright future for continuous advancements in clinicians’ diagnostic skills, heralding a transformative era in digital healthcare.
Acknowledgments
This study was made possible using the resources from the Department of Diagnostic and Generalist Medicine, Dokkyo Medical University.
-
Research ethics: Not applicable.
-
Informed consent: Not applicable.
-
Author contributions: The authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: The authors state no conflict of interest.
-
Research funding: None declared.
-
Data availability: Not applicable.
References
1. Patrizio, A. Google bard: TechTarget; 2023. Available from: https://www.techtarget.com/searchenterpriseai/definition/Google-Bard.Search in Google Scholar
2. Aydın, Ö. Google bard generated literature review: metaverse. Available from: SSRN. 2023.10.61969/jai.1311271Search in Google Scholar
3. Touvron, H, Lavril, T, Izacard, G, Martinet, X, Lachaux, M-A, Lacroix, T, et al.. Llama: open and efficient foundation language models. arXiv preprint arXiv:230213971 2023.Search in Google Scholar
4. OpenAI. GPT-4 Technical report 2023 March 01, 2023. [arXiv:2303.08774 p.].Search in Google Scholar
5. Kassirer, JR. Clinical problem-solving—a new feature in the Journal. N Engl J Med 1992;326:60–1. https://doi.org/10.1056/nejm199201023260112.Search in Google Scholar PubMed
6. Pile, JC, Baudendistel, TE, Harte, B. Clinical Care Conundrums: challenging diagnoses in hospital medicine. Hoboken, New Jersey: John Wiley & Sons; 2013.10.1002/9781118483206Search in Google Scholar
7. Elstein, AS, Schwarz, A. Clinical problem solving and diagnostic decision making: selective review of the cognitive literature. BMJ 2002;324:729–32. https://doi.org/10.1136/bmj.324.7339.729.Search in Google Scholar PubMed PubMed Central
8. Hirosawa, T, Harada, Y, Yokose, M, Sakamoto, T, Kawamura, R, Shimizu, T. Diagnostic accuracy of differential-diagnosis lists generated by generative pretrained transformer 3 Chatbot for clinical Vignettes with common chief complaints: a pilot study. Int J Environ Res Publ Health 2023;20:3378. https://doi.org/10.3390/ijerph20043378.Search in Google Scholar PubMed PubMed Central
9. Kanjee, Z, Crowe, B, Rodman, A. Accuracy of a generative artificial intelligence model in a complex diagnostic Challenge. JAMA 2023;330:78–80. https://doi.org/10.1001/jama.2023.8288.Search in Google Scholar PubMed PubMed Central
10. Shimizu, T. Twelve tips for physicians’ mastering expertise in diagnostic excellence. MedEdPublish 2023;13:21. https://doi.org/10.12688/mep.19618.1.Search in Google Scholar
© 2023 Walter de Gruyter GmbH, Berlin/Boston
