Can we Adopt Self-supervised Pretraining for Chest X-Rays?
This work addresses the need for effective pretraining in medical imaging when labeled data is scarce, but it is incremental as it builds on existing self-supervised methods.
The study analyzed self-supervised pretraining for chest X-rays, finding that supervised ImageNet pretraining yields strong representations, self-supervised methods on ImageNet and CXR datasets perform similarly, and combining supervised ImageNet with self-supervised CXR improves results on small downstream datasets.
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.