Assessment of ChatGPT in the Prehospital Management of Ophthalmological Emergencies - An Analysis of 10 Fictional Case Vignettes.

IF 0.8 4区医学 Q4 OPHTHALMOLOGY

Klinische Monatsblatter fur Augenheilkunde Pub Date : 2024-05-01 Epub Date: 2023-10-27 DOI:10.1055/a-2149-0447

Dominik Knebel, Siegfried Priglinger, Nicolas Scherer, Julian Klaas, Jakob Siedlecki, Benedikt Schworm

{"title":"Assessment of ChatGPT in the Prehospital Management of Ophthalmological Emergencies - An Analysis of 10 Fictional Case Vignettes.","authors":"Dominik Knebel, Siegfried Priglinger, Nicolas Scherer, Julian Klaas, Jakob Siedlecki, Benedikt Schworm","doi":"10.1055/a-2149-0447","DOIUrl":null,"url":null,"abstract":"Background: The artificial intelligence (AI)-based platform ChatGPT (Chat Generative Pre-Trained Transformer, OpenAI LP, San Francisco, CA, USA) has gained impressive popularity in recent months. Its performance on case vignettes of general medical (non-ophthalmological) emergencies has been assessed - with very encouraging results. The purpose of this study was to assess the performance of ChatGPT on ophthalmological emergency case vignettes in terms of the main outcome measures triage accuracy, appropriateness of recommended prehospital measures, and overall potential to inflict harm to the user/patient.Methods: We wrote ten short, fictional case vignettes describing different acute ophthalmological symptoms. Each vignette was entered into ChatGPT five times with the same wording and following a standardized interaction pathway. The answers were analyzed following a systematic approach.Results: We observed a triage accuracy of 93.6%. Most answers contained only appropriate recommendations for prehospital measures. However, an overall potential to inflict harm to users/patients was present in 32% of answers.Conclusion: ChatGPT should presently not be used as a stand-alone primary source of information about acute ophthalmological symptoms. As AI continues to evolve, its safety and efficacy in the prehospital management of ophthalmological emergencies has to be reassessed regularly.","PeriodicalId":17904,"journal":{"name":"Klinische Monatsblatter fur Augenheilkunde","volume":" ","pages":"675-681"},"PeriodicalIF":0.8000,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Klinische Monatsblatter fur Augenheilkunde","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1055/a-2149-0447","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2023/10/27 0:00:00","PubModel":"Epub","JCR":"Q4","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: The artificial intelligence (AI)-based platform ChatGPT (Chat Generative Pre-Trained Transformer, OpenAI LP, San Francisco, CA, USA) has gained impressive popularity in recent months. Its performance on case vignettes of general medical (non-ophthalmological) emergencies has been assessed - with very encouraging results. The purpose of this study was to assess the performance of ChatGPT on ophthalmological emergency case vignettes in terms of the main outcome measures triage accuracy, appropriateness of recommended prehospital measures, and overall potential to inflict harm to the user/patient.

Methods: We wrote ten short, fictional case vignettes describing different acute ophthalmological symptoms. Each vignette was entered into ChatGPT five times with the same wording and following a standardized interaction pathway. The answers were analyzed following a systematic approach.

Results: We observed a triage accuracy of 93.6%. Most answers contained only appropriate recommendations for prehospital measures. However, an overall potential to inflict harm to users/patients was present in 32% of answers.

Conclusion: ChatGPT should presently not be used as a stand-alone primary source of information about acute ophthalmological symptoms. As AI continues to evolve, its safety and efficacy in the prehospital management of ophthalmological emergencies has to be reassessed regularly.

查看原文本刊更多论文

ChatGPT在眼科急诊院前管理中的评估——对10例虚构病例的分析。

背景：最近几个月，基于人工智能（AI）的平台ChatGPT（Chat Generative Pre-Trained Transformer，OpenAI LP，旧金山，CA，USA）获得了令人印象深刻的人气。它在一般医疗（非眼科）紧急情况下的表现已经得到了评估，取得了非常令人鼓舞的结果。本研究的目的是评估ChatGPT在眼科急诊病例小插曲中的表现，包括主要结果指标分诊的准确性、推荐的院前措施的适当性以及对使用者/患者造成伤害的总体可能性。方法：我们写了十个简短的虚构病例小插曲，描述了不同的急性眼科症状。每个小插曲都以相同的措辞进入ChatGPT五次，并遵循标准化的交互路径。按照系统的方法对答案进行了分析。结果：我们观察到分诊的准确率为93.6%。大多数答案只包含适当的院前措施建议。然而，32%的回答中存在对使用者/患者造成伤害的总体可能性。结论：目前不应将ChatGPT作为有关急性眼科症状的独立主要信息来源。随着人工智能的不断发展，必须定期重新评估其在眼科急诊院前管理中的安全性和有效性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Klinische Monatsblatter fur Augenheilkunde 医学-眼科学

CiteScore

1.30

自引率

0.00%

发文量

235

审稿时长

4-8 weeks

期刊介绍： -Konzentriertes Fachwissen aus Klinik und Praxis: Die entscheidenden Ergebnisse der internationalen Forschung - für Sie auf den Punkt gebracht und kritisch kommentiert, Übersichtsarbeiten zu den maßgeblichen Themen der täglichen Praxis, Top informiert - breite klinische Berichterstattung. -CME-Punkte sammeln mit dem Refresher: Effiziente, CME-zertifizierte Fortbildung, mit dem Refresher, 3 CME-Punkte pro Ausgabe - bis zu 36 CME-Punkte im Jahr!. -Aktuelle Rubriken mit echtem Nutzwert: Kurzreferate zu den wichtigsten Artikeln internationaler Zeitschriften, Schwerpunktthema in jedem Heft: Ausführliche Übersichtsarbeiten zu den wichtigsten Themen der Ophthalmologie – so behalten Sie das gesamte Fach im Blick!, Originalien mit den neuesten Entwicklungen, Übersichten zu den relevanten Themen.