Chronological Contrastive Learning: Few-Shot Progression Assessment in Irreversible Diseases
This work addresses the annotation burden for irreversible disease progression assessment in medical imaging, offering a practical solution to reduce expert labeling needs, though it is incremental as it builds on existing contrastive learning ideas.
The paper tackles the problem of costly and variable expert annotation for disease severity scoring in medical imaging by introducing ChronoCon, a contrastive learning method that uses only the chronological order of longitudinal scans to learn disease-relevant representations without labels. It achieves an intraclass correlation coefficient of 86% for severity prediction with fine-tuning on expert scores from only five patients, significantly outperforming supervised baselines in low-label settings.
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 longitudinal scans. Under the clinically plausible assumption of monotonic progression in irreversible diseases, the method learns disease-relevant representations without using any expert labels. This generalizes the idea of Rank-N-Contrast from label distances to temporal ordering. Evaluated on rheumatoid arthritis radiographs for severity assessment, the learned representations substantially improve label efficiency. In low-label settings, ChronoCon significantly outperforms a fully supervised baseline initialized from ImageNet weights. In a few-shot learning experiment, fine-tuning ChronoCon on expert scores from only five patients yields an intraclass correlation coefficient of 86% for severity score prediction. These results demonstrate the potential of chronological contrastive learning to exploit routinely available imaging metadata to reduce annotation requirements in the irreversible disease domain. Code is available at https://github.com/cirmuw/ChronoCon.