Knebel Dominik, Priglinger Siegfried, Scherer Nicolas, Klaas Julian, Siedlecki Jakob, Schworm Benedikt
Department of Ophthalmology, University Hospital, Ludwigs-Maximilians-Universität München, München, Germany.
Klin Monbl Augenheilkd. 2024 May;241(5):675-681. doi: 10.1055/a-2149-0447. Epub 2023 Oct 27.
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.
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.
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.
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.
基于人工智能（AI）的平台ChatGPT（聊天生成预训练变换器，美国加利福尼亚州旧金山的OpenAI LP）在最近几个月获得了令人瞩目的人气。其在一般医疗（非眼科）紧急情况的病例 vignettes 上的表现已经过评估——结果非常令人鼓舞。本研究的目的是根据主要结局指标分诊准确性、推荐的院前措施的适当性以及对用户/患者造成伤害的总体可能性，评估ChatGPT在眼科紧急情况病例 vignettes 上的表现。
我们撰写了十个简短的虚构病例 vignettes，描述不同的急性眼科症状。每个 vignette 以相同的措辞并按照标准化的交互路径输入ChatGPT五次。对答案进行系统分析。
我们观察到分诊准确率为93.6%。大多数答案仅包含对院前措施的适当建议。然而，32%的答案中存在对用户/患者造成伤害的总体可能性。
目前ChatGPT不应被用作关于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紧急情况院前管理中的安全性和有效性。
