Uses the chronological visitation order of medical scans as a self-supervised signal for disease progression modeling.
March 24, 2026
Original Paper
Chronological Contrastive Learning: Few-Shot Progression Assessment in Irreversible Diseases
arXiv · 2603.21935
The Takeaway
By assuming monotonic progression in irreversible diseases, it bypasses the need for expensive expert-annotated severity scores. The method achieves high diagnostic accuracy with as few as five labeled patients, unlocking the ability to train on vast, previously unusable longitudinal clinical archives.
From the abstract
Quantitative disease severity scoring in medical imaging is costly, time-consuming, and subject to inter-reader variability. At the same time, clinical archives contain far more longitudinal imaging data than expert-annotated severity scores. Existing self-supervised methods typically ignore this chronological structure. We introduce ChronoCon, a contrastive learning approach that replaces label-based ranking losses with rankings derived solely from the visitation order of a patient's longitudin