构建大规模医学影像问答数据集，推动CT/MRI精准诊断研究
RadImageNet-VQA: A Large-Scale CT and MRI Dataset for Radiologic Visual Question Answering
- 基于专家标注构建750万图像-问题对，覆盖97种病灶和8个解剖区域
- 顶尖视觉语言模型在细粒度病理识别上仍表现不佳，尤其开放问答场景
- 实验证明模型依赖图像而非文字线索，适合医疗AI可解释性研究
本文提出RadImageNet-VQA，一个面向CT和MRI影像的大型放射学视觉问答数据集。现有医学VQA数据集规模有限，多集中于X光或生物图示，且易受文本线索干扰。本数据集基于专家标注，包含750K张图像与750万组问题-答案样本，涵盖异常检测、解剖识别、病理识别三类任务，覆盖8个解剖区域及97种病理类型，支持开放式、封闭式和多选题。大量实验表明，当前先进视觉语言模型在细粒度病理识别上仍表现欠佳，尤其在开放式问答中，即使微调后也难突破。纯文本分析显示，无图像输入时模型性能接近随机，证实该数据集避免了语言捷径。完整数据集与基准已公开于https://huggingface.co/datasets/raidium/RadImageNet-VQA。
原文摘要 · Abstract (English)
In this work, we introduce RadImageNet-VQA, a large-scale dataset designed to advance radiologic visual question answering (VQA) on CT and MRI exams. Existing medical VQA datasets are limited in scale, dominated by X-ray imaging or biomedical illustrations, and often prone to text-based shortcuts. RadImageNet-VQA is built from expert-curated annotations and provides 750K images paired with 7.5M question-answer samples. It covers three key tasks - abnormality detection, anatomy recognition, and pathology identification - spanning eight anatomical regions and 97 pathology categories, and supports open-ended, closed-ended, and multiple-choice questions. Extensive experiments show that state-of-the-art vision-language models still struggle with fine-grained pathology identification, particularly in open-ended settings and even after fine-tuning. Text-only analysis further reveals that model performance collapses to near-random without image inputs, confirming that RadImageNet-VQA is free from linguistic shortcuts. The full dataset and benchmark are publicly available at https://huggingface.co/datasets/raidium/RadImageNet-VQA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。