Can we Adopt Self-supervised Pretraining for Chest X-Rays?

TL;DR

This paper investigates the effectiveness of self-supervised pretraining methods for chest X-ray image analysis, comparing it to traditional supervised pretraining on ImageNet and exploring combined approaches for improved performance.

Contribution

It provides a comprehensive analysis of self-supervised pretraining on both ImageNet and CXR datasets, highlighting when and how these methods can enhance medical image classification.

Findings

Supervised ImageNet pretraining yields strong representations.

Self-supervised pretraining on ImageNet is comparable to CXR datasets.

Combining supervised ImageNet training with self-supervised CXR fine-tuning improves results.

Abstract

Chest radiograph (or Chest X-Ray, CXR) is a popular medical imaging modality that is used by radiologists across the world to diagnose heart or lung conditions. Over the last decade, Convolutional Neural Networks (CNN), have seen success in identifying pathologies in CXR images. Typically, these CNNs are pretrained on the standard ImageNet classification task, but this assumes availability of large-scale annotated datasets. In this work, we analyze the utility of pretraining on unlabeled ImageNet or Chest X-Ray (CXR) datasets using various algorithms and in multiple settings. Some findings of our work include: (i) supervised training with labeled ImageNet learns strong representations that are hard to beat; (ii) self-supervised pretraining on ImageNet (~1M images) shows performance similar to self-supervised pretraining on a CXR dataset (~100K images); and (iii) the CNN trained on supervised ImageNet can be trained further with self-supervised CXR images leading to improvements, especially when the downstream dataset is on the order of a few thousand images.