Generalized Radiograph Representation Learning via Cross-supervision between Images and Free-text Radiology Reports
This addresses the need for efficient pre-training in medical imaging by reducing annotation costs, though it is incremental as it builds on existing cross-supervision ideas.
The paper tackles the problem of labor-intensive annotation and limited performance in radiograph analysis by proposing REFERS, a cross-supervised method that uses free-text radiology reports for supervision, achieving superior performance on 4 X-ray datasets under limited supervision and even surpassing methods with human-assisted labels.
Pre-training lays the foundation for recent successes in radiograph analysis supported by deep learning. It learns transferable image representations by conducting large-scale fully-supervised or self-supervised learning on a source domain. However, supervised pre-training requires a complex and labor intensive two-stage human-assisted annotation process while self-supervised learning cannot compete with the supervised paradigm. To tackle these issues, we propose a cross-supervised methodology named REviewing FreE-text Reports for Supervision (REFERS), which acquires free supervision signals from original radiology reports accompanying the radiographs. The proposed approach employs a vision transformer and is designed to learn joint representations from multiple views within every patient study. REFERS outperforms its transfer learning and self-supervised learning counterparts on 4 well-known X-ray datasets under extremely limited supervision. Moreover, REFERS even surpasses methods based on a source domain of radiographs with human-assisted structured labels. Thus REFERS has the potential to replace canonical pre-training methodologies.