Temporal Inversion for Learning Interval Change in Chest X-Rays

TL;DR

This paper introduces TILA, a framework that uses temporal inversion as a supervisory signal to improve the sensitivity of vision-language models to changes over time in chest X-ray images.

Contribution

TILA is a novel approach that explicitly incorporates temporal order learning into vision-language models for chest radiographs, enhancing interval change detection.

Findings

TILA improves progression classification accuracy.

TILA enhances temporal embedding alignment.

TILA shows consistent gains across multiple architectures.

Abstract

Recent advances in vision--language pretraining have enabled strong medical foundation models, yet most analyze radiographs in isolation, overlooking the key clinical task of comparing prior and current images to assess interval change. For chest radiographs (CXRs), capturing interval change is essential, as radiologists must evaluate not only the static appearance of findings but also how they evolve over time. We introduce TILA (Temporal Inversion-aware Learning and Alignment), a simple yet effective framework that uses temporal inversion, reversing image pairs, as a supervisory signal to enhance the sensitivity of existing temporal vision-language models to directional change. TILA integrates inversion-aware objectives across pretraining, fine-tuning, and inference, complementing conventional appearance modeling with explicit learning of temporal order. We also propose a unified evaluation protocol to assess order sensitivity and consistency under temporal inversion, and introduce MS-CXR-Tretrieval, a retrieval evaluation set constructed through a general protocol that can be applied to any temporal CXR dataset. Experiments on public datasets and real-world hospital cohorts demonstrate that TILA consistently improves progression classification and temporal embedding alignment when applied to multiple existing architectures.