Self-supervised pretraining for label-efficient medical imaging

Resumen

Annotating medical images is costly and requires expert radiologists. We evaluate a contrastive self-supervised pretraining scheme on chest radiographs and demonstrate that downstream classifiers reach expert-level AUROC with only 10% of the labels required by supervised baselines. We analyze failure modes on rare pathologies and propose a curriculum that mitigates them.

Palabras clave