CheXTemporal: A Dataset for Temporally-Grounded Reasoning in Chest Radiography
For researchers developing medical vision-language models, this work highlights the need for explicit temporal supervision to improve longitudinal reasoning in chest radiography.
CheXTemporal introduces a dataset with paired prior-current chest X-rays and fine-grained temporal annotations to evaluate vision-language models on temporal reasoning. Current models show significant limitations in spatial grounding and fine-grained progression classification, performing worse on subtle states like stable and resolved compared to salient categories like worse.
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.