arXiv 2605.113042026-05-11cs.CV

CheXTemporal：用于胸部X光影像中时间感知推理的数据集

CheXTemporal: A Dataset for Temporally-Grounded Reasoning in Chest Radiography

Stanford University（斯坦福大学）；University of Oxford（牛津大学）；University of California, Berkeley（加州大学伯克利分校）；HOPPR；University Hospital Zurich（苏黎世大学医院）

Eva Prakash, Yunhe Gao, Chong Wang, Justin Xu, Neal Prakash, Arne Michalson, Seena Dehkharghani, Eun Kyoung Hong, Julie Bauml, Roger Boodoo, Jean-Benoit Delbrou… 展开作者

Eva Prakash, Yunhe Gao, Chong Wang, Justin Xu, Neal Prakash, Arne Michalson, Seena Dehkharghani, Eun Kyoung Hong, Julie Bauml, Roger Boodoo, Jean-Benoit Delbrouck, Sophie Ostmeier, Curtis Langlotz

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

该研究提出CheXTemporal数据集，用于胸部X光影像中时间感知推理，包含前后X光片及时间空间标注，评估了多种视觉语言模型在零样本设定下的接地和进展分类任务，发现模型在空间接地和时间推理方面存在局限。

摘要：

胸部X光影像解读需要对先前和当前研究进行时间推理，但大多数视觉语言模型仅在静态图像-报告配对上训练，缺乏对纵向变化的显式监督。我们引入CheXTemporal数据集，用于胸部X光影像中时间感知推理，包含配对的先前-当前胸部X光片（CXR）及其在发现层面的时间和空间标注。该数据集包含五类进展分类法（新、更差、稳定、改善、解决），局部空间监督病理学，配对研究间的显式空间-时间对齐，以及多源覆盖用于跨领域评估。我们还构建了一个包含280,000对银数据集，通过自动衍生的时间和解剖学监督进行大规模评估。使用这些资源，我们评估了多种最先进的视觉语言X光影像模型，在接地和进展分类任务中进行零样本设置下的评估。在黄金和银评价中，当前模型在空间接地、细粒度时间推理和分布偏移下的鲁棒性方面表现出一致的局限。特别是，模型在显著进展类别（如更差）上表现明显优于在时间细微状态（如稳定和解决）上，表明在胸部X光影像中对纵向疾病演变的建模有限。

英文摘要：

Chest radiograph interpretation requires temporal reasoning over prior and current studies, yet most vision-language models are trained on static image-report pairs and lack explicit supervision for modeling longitudinal change. We introduce CheXTemporal, a dataset for temporally grounded reasoning in chest radiography consisting of paired prior-current chest X-rays (CXR) with finding-level temporal and spatial annotations. The dataset includes a five-class progression taxonomy (new, worse, stable, improved, resolved), localized spatial supervision of pathology, explicit spatial-temporal alignment across paired studies, and multi-source coverage for cross-domain evaluation. We additionally construct a 280K-pair silver dataset with automatically derived temporal and anatomical supervision for large-scale evaluation under weaker supervision. Using these resources, we evaluate multiple state-of-the-art vision-language CXR models on grounding and progression-classification tasks in a zero-shot setting. Across both gold and silver evaluations, current models exhibit consistent limitations in spatial grounding, fine-grained temporal reasoning, and robustness under distribution shift. In particular, models perform substantially better on salient progression categories such as worse than on temporally subtle states such as stable and resolved, suggesting limited modeling of longitudinal disease evolution in chest radiography.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。