Self-Supervised Domain Adaptation for Diabetic Retinopathy Grading using Vessel Image Reconstruction
This addresses the problem of domain shift in medical imaging for diabetic retinopathy grading, which is incremental as it builds on existing domain adaptation strategies.
The paper tackles domain adaptation for diabetic retinopathy grading by learning invariant features through a self-supervised task based on retinal vessel image reconstruction, outperforming existing methods and competing with state-of-the-art approaches in classification accuracy.
This paper investigates the problem of domain adaptation for diabetic retinopathy (DR) grading. We learn invariant target-domain features by defining a novel self-supervised task based on retinal vessel image reconstructions, inspired by medical domain knowledge. Then, a benchmark of current state-of-the-art unsupervised domain adaptation methods on the DR problem is provided. It can be shown that our approach outperforms existing domain adaption strategies. Furthermore, when utilizing entire training data in the target domain, we are able to compete with several state-of-the-art approaches in final classification accuracy just by applying standard network architectures and using image-level labels.