A Comparison of Self-Supervised Pretraining Approaches for Predicting Disease Risk from Chest Radiograph Images
This work addresses label efficiency for disease risk prediction in medical imaging, but it is incremental as it compares existing methods on a specific task.
The paper tackled the problem of predicting disease risk from chest radiograph images by comparing self-supervised pretraining approaches, finding that a semi-supervised autoencoder outperformed contrastive and transfer learning in internal and external validation.
Deep learning is the state-of-the-art for medical imaging tasks, but requires large, labeled datasets. For risk prediction, large datasets are rare since they require both imaging and follow-up (e.g., diagnosis codes). However, the release of publicly available imaging data with diagnostic labels presents an opportunity for self and semi-supervised approaches to improve label efficiency for risk prediction. Though several studies have compared self-supervised approaches in natural image classification, object detection, and medical image interpretation, there is limited data on which approaches learn robust representations for risk prediction. We present a comparison of semi- and self-supervised learning to predict mortality risk using chest x-ray images. We find that a semi-supervised autoencoder outperforms contrastive and transfer learning in internal and external validation.