Hidden flaws behind expert-level accuracy of multimodal GPT-4 vision in medicine

• URL: http://arxiv.org/abs/2401.08396v4
• Date: Sat, 31 Aug 2024 23:51:14 GMT
• Title: Hidden flaws behind expert-level accuracy of multimodal GPT-4 vision in medicine
• Authors: Qiao Jin, Fangyuan Chen, Yiliang Zhou, Ziyang Xu, Justin M. Cheung, Robert Chen, Ronald M. Summers, Justin F. Rousseau, Peiyun Ni, Marc J Landsman, Sally L. Baxter, Subhi J. Al'Aref, Yijia Li, Alex Chen, Josef A. Brejt, Michael F. Chiang, Yifan Peng, Zhiyong Lu,
• Abstract summary: Generative Pre-trained Transformer 4 with Vision (GPT-4V) outperforms human physicians in medical challenge tasks. 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.
• Score: 15.491432387608112
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• GMAI-MMBench: A Comprehensive Multimodal Evaluation Benchmark Towards General Medical AI [67.09501109871351]
  Large Vision-Language Models (LVLMs) are capable of handling diverse data types such as imaging, text, and physiological signals. GMAI-MMBench is the most comprehensive general medical AI benchmark with well-categorized data structure and multi-perceptual granularity to date. It is constructed from 284 datasets across 38 medical image modalities, 18 clinical-related tasks, 18 departments, and 4 perceptual granularities in a Visual Question Answering (VQA) format.
  arXiv  Detail & Related papers  (2024-08-06T17:59:21Z)
• Enhancing Medical Task Performance in GPT-4V: A Comprehensive Study on Prompt Engineering Strategies [28.98518677093905]
  GPT-4V, OpenAI's latest large vision-language model, has piqued considerable interest for its potential in medical applications. Recent studies and internal reviews highlight its underperformance in specialized medical tasks. This paper explores the boundary of GPT-4V's capabilities in medicine, particularly in processing complex imaging data from endoscopies, CT scans, and MRIs etc.
  arXiv  Detail & Related papers  (2023-12-07T15:05:59Z)
• Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine [89.46836590149883]
  We build on a prior study of GPT-4's capabilities on medical challenge benchmarks in the absence of special training. We find that prompting innovation can unlock deeper specialist capabilities and show that GPT-4 easily tops prior leading results for medical benchmarks. With Medprompt, GPT-4 achieves state-of-the-art results on all nine of the benchmark datasets in the MultiMedQA suite.
  arXiv  Detail & Related papers  (2023-11-28T03:16:12Z)
• GPT-4V(ision) Unsuitable for Clinical Care and Education: A Clinician-Evaluated Assessment [6.321623278767821]
  GPT-4V was recently developed for general image interpretation. Board-certified physicians and senior residents assessed GPT-4V's proficiency across a range of medical conditions. GPT-4V's diagnostic accuracy and clinical decision-making abilities are poor, posing risks to patient safety.
  arXiv  Detail & Related papers  (2023-11-14T17:06:09Z)
• Holistic Evaluation of GPT-4V for Biomedical Imaging [113.46226609088194]
  GPT-4V represents a breakthrough in artificial general intelligence for computer vision. We assess GPT-4V's performance across 16 medical imaging categories, including radiology, oncology, ophthalmology, pathology, and more. Results show GPT-4V's proficiency in modality and anatomy recognition but difficulty with disease diagnosis and localization.
  arXiv  Detail & Related papers  (2023-11-10T18:40:44Z)
• Accuracy of a Vision-Language Model on Challenging Medical Cases [1.7726473251723847]
  General-purpose large language models that utilize both text and images have not been evaluated on a diverse array of challenging medical cases. We evaluated the accuracy of the recently released Generative Pre-trained Transformer 4 with Vision model (GPT-4V) compared to human respondents. We also conducted a physician evaluation of GPT-4V on 69 NEJM clinicopathological conferences.
  arXiv  Detail & Related papers  (2023-11-09T18:48:02Z)
• A Systematic Evaluation of GPT-4V's Multimodal Capability for Medical Image Analysis [87.25494411021066]
  GPT-4V's multimodal capability for medical image analysis is evaluated. It is found that GPT-4V excels in understanding medical images and generates high-quality radiology reports. It is found that its performance for medical visual grounding needs to be substantially improved.
  arXiv  Detail & Related papers  (2023-10-31T11:39:09Z)
• Multimodal ChatGPT for Medical Applications: an Experimental Study of GPT-4V [20.84152508192388]
  We critically evaluate the capabilities of the state-of-the-art multimodal large language model, GPT-4 with Vision (GPT-4V) Our experiments thoroughly assess GPT-4V's proficiency in answering questions paired with images using both pathology and radiology datasets. The experiments with accuracy score conclude that the current version of GPT-4V is not recommended for real-world diagnostics.
  arXiv  Detail & Related papers  (2023-10-29T16:26:28Z)
• Can GPT-4V(ision) Serve Medical Applications? Case Studies on GPT-4V for Multimodal Medical Diagnosis [59.35504779947686]
  GPT-4V is OpenAI's newest model for multimodal medical diagnosis. Our evaluation encompasses 17 human body systems. GPT-4V demonstrates proficiency in distinguishing between medical image modalities and anatomy. It faces significant challenges in disease diagnosis and generating comprehensive reports.
  arXiv  Detail & Related papers  (2023-10-15T18:32:27Z)
• COVID-Net USPro: An Open-Source Explainable Few-Shot Deep Prototypical Network to Monitor and Detect COVID-19 Infection from Point-of-Care Ultrasound Images [66.63200823918429]
  COVID-Net USPro monitors and detects COVID-19 positive cases with high precision and recall from minimal ultrasound images. The network achieves 99.65% overall accuracy, 99.7% recall and 99.67% precision for COVID-19 positive cases when trained with only 5 shots.
  arXiv  Detail & Related papers  (2023-01-04T16:05:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.