Big Self-Supervised Models Advance Medical Image Classification

TL;DR

This paper demonstrates that large self-supervised models, especially with the novel MICLe method, significantly improve medical image classification accuracy and robustness, even with limited labeled data.

Contribution

It introduces a new Multi-Instance Contrastive Learning (MICLe) method and shows that self-supervised pretraining on ImageNet and medical images enhances medical image classification.

Findings

6.7% top-1 accuracy improvement in dermatology classification

1.1% mean AUC improvement in chest X-ray classification

Models are robust to distribution shifts and data scarcity

Abstract

Self-supervised pretraining followed by supervised fine-tuning has seen success in image recognition, especially when labeled examples are scarce, but has received limited attention in medical image analysis. This paper studies the effectiveness of self-supervised learning as a pretraining strategy for medical image classification. We conduct experiments on two distinct tasks: dermatology skin condition classification from digital camera images and multi-label chest X-ray classification, and demonstrate that self-supervised learning on ImageNet, followed by additional self-supervised learning on unlabeled domain-specific medical images significantly improves the accuracy of medical image classifiers. We introduce a novel Multi-Instance Contrastive Learning (MICLe) method that uses multiple images of the underlying pathology per patient case, when available, to construct more informative positive pairs for self-supervised learning. Combining our contributions, we achieve an improvement of 6.7% in top-1 accuracy and an improvement of 1.1% in mean AUC on dermatology and chest X-ray classification respectively, outperforming strong supervised baselines pretrained on ImageNet. In addition, we show that big self-supervised models are robust to distribution shift and can learn efficiently with a small number of labeled medical images.