
==== Front
ArXiv
ArXiv
arxiv
ArXiv
2331-8422
Cornell University

arXiv:2401.08396v4
2401.08396
4
preprint
Article
Hidden flaws behind expert-level accuracy of multimodal GPT-4 vision in medicine
Jin Qiao
Chen Fangyuan
Zhou Yiliang
Xu Ziyang
Cheung Justin M.
Chen Robert
Summers Ronald M.
Rousseau Justin F.
Ni Peiyun
Landsman Marc J
Baxter Sally L.
Al'Aref Subhi J.
Li Yijia
Chen Alex
Brejt Josef A.
Chiang Michael F.
Peng Yifan
Lu Zhiyong
31 8 2024
arXiv:2401.08396v416 1 2024
https://creativecommons.org/licenses/by/4.0/ This work is licensed under a Creative Commons Attribution 4.0 International License, which allows reusers to distribute, remix, adapt, and build upon the material in any medium or format, so long as attribution is given to the creator. The license allows for commercial use.
http://arxiv.org/abs/2401.08396v4
nihpp-2401.08396v4.pdf
Recent studies indicate that Generative Pre-trained Transformer 4 with Vision (GPT-4V) outperforms human physicians in medical challenge tasks. However, these evaluations primarily focused on the accuracy of multi-choice questions alone. Our study extends the current scope by conducting a comprehensive analysis of GPT-4V's rationales of image comprehension, recall of medical knowledge, and step-by-step multimodal reasoning when solving New England Journal of Medicine (NEJM) Image Challenges - an imaging quiz designed to test the knowledge and diagnostic capabilities of medical professionals. Evaluation results confirmed that GPT-4V performs comparatively to human physicians regarding multi-choice accuracy (81.6% vs. 77.8%). GPT-4V also performs well in cases where physicians incorrectly answer, with over 78% accuracy. However, we discovered that GPT-4V frequently presents flawed rationales in cases where it makes the correct final choices (35.5%), most prominent in image comprehension (27.2%). Regardless of GPT-4V's high accuracy in multi-choice questions, our findings emphasize the necessity for further in-depth evaluations of its rationales before integrating such multimodal AI models into clinical workflows.
==== Body
pmc
