【论文精度-大模型结果评测】A Systematic Evaluation of GPT-4V’s Multimodal Capability for Medical Image Analysis

A Systematic Evaluation of GPT-4V’s Multimodal Capability for Medical Image Analysis

1.Keyword

GPT-4V - 多模态模型
Medical Image - 医学影像图片
Radiology Report Generation - 放射性报告生成
Medical Visual Question Answering - 医学视觉问答
Medical Visual Grounding - 找到文本所指的内容在图像中的位置
Large Language Model Evaluation - 大模型评估

2. Research Problem

1. 本文对GPT-4V在医学图像分析上的能力进行评测，主要是聚焦在3个任务: 1. Radiology Report Generation 2. Medical Visual Question Answering 3. Medical Visual Grounding
2. For the evaluation, a set of prompts is designed for each task to induce the corresponding capability of GPT4V to produce sufficiently good outputs
3. Three evaluation ways including quantitative analysis, human evaluation, and case study are employed to achieve an in-depth and extensive evalu