"Vertigo, likely peripheral": the dizzying rise of ChatGPT.
Artificial intelligence
Dizziness
Natural language processing
Vertigo
Journal
European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgery
ISSN: 1434-4726
Titre abrégé: Eur Arch Otorhinolaryngol
Pays: Germany
ID NLM: 9002937
Informations de publication
Date de publication:
Oct 2023
Oct 2023
Historique:
received:
17
06
2023
accepted:
13
07
2023
medline:
5
9
2023
pubmed:
26
7
2023
entrez:
26
7
2023
Statut:
ppublish
Résumé
The use of artificial intelligence (AI) in medical decision-making has once again come to the forefront with the prevalence of Natural Language Processing (NLP). In this exploratory article, we tested one such model, ChatGPT, for its ability to identify vestibular causes of dizziness METHODS: Eight hypothetical scenarios were presented to ChatGPT, which included varying clinical pictures and types of prompts. The responses given by ChatGPT were evaluated for coherence, clarity, consistency, accuracy, appropriateness, and recognition of limitations. ChatGPT provided coherent and logical responses. The model accurately provided differentials for both vestibular and non-vestibular causes of dizziness, with the correct diagnosis presented first in six of the cases, with important limitations CONCLUSION: Being an AI tool, ChatGPT lacks the ability to process certain nuances in clinical decision making, in both identifying atypical dizziness, as well as in recommending further examination steps to elucidate a clearer diagnosis. We believe that AI will continue to forge ahead in the medical field. Merging the immense knowledge base of AI programming with the nuances of clinical assessment and knowledge integration will surely enhance patient care in the years to come.
Identifiants
pubmed: 37493845
doi: 10.1007/s00405-023-08135-1
pii: 10.1007/s00405-023-08135-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
4687-4689Informations de copyright
© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.
Références
Kerber KA, Newman-Toker DE. Misdiagnosing Dizzy Patients: Common Pitfalls in Clinical Practice. Neurol Clin. 2015;33(3):565–75. https://doi.org/10.1016/j.ncl.2015.04.009
Hybels RL. History taking in dizziness. The most important diagnostic step. Postgrad Med. 1984;75(3):41–3. https://doi.org/10.1080/00325481.1984.11698590
Anh DT, Takakura H, Asai M, Ueda N, Shojaku H (2022) Application of machine learning in the diagnosis of vestibular disease. Sci Rep 12(1):20805. https://doi.org/10.1038/s41598-022-24979-9
doi: 10.1038/s41598-022-24979-9
pubmed: 36460741
pmcid: 9718758
Howard A, Hope W, Gerada A (2023) ChatGPT and antimicrobial advice: the end of the consulting infection doctor? Lancet Infect Dis 23(4):405–406. https://doi.org/10.1016/S1473-3099(23)00113-5
doi: 10.1016/S1473-3099(23)00113-5
pubmed: 36822213
Dmitriew C, Regis A, Bodunde O, Lepage R, Turgeon Z, McIsaac S et al (2021) Diagnostic accuracy of the HINTS exam in an emergency department: a retrospective chart review. Acad Emerg Med 28(4):387–393. https://doi.org/10.1111/acem.14171
doi: 10.1111/acem.14171
pubmed: 33171003
Sallam M. ChatGPT Utility in Healthcare Education, Research, and Practice: Systematic Review on the Promising Perspectives and Valid Concerns. Healthcare (Basel). 2023; https://doi.org/10.3390/healthcare11060887
OpenAI. OpenAI Blog [Internet]. OpenAI, editor. https://openai.com/blog/chatgpt2022 . [cited 2023].