Distributional Shifts in Automated Diabetic Retinopathy Screening
This addresses reliability issues in automated medical screening for diabetic patients, but it is incremental as it builds on existing OOD detection methods.
The paper tackles the problem of deep learning models for diabetic retinopathy screening degrading in accuracy due to distributional shifts and misclassifying non-retina images as referable, presenting a Dirichlet Prior Network-based framework that uses an OOD detector to identify such images for human intervention.
Deep learning-based models are developed to automatically detect if a retina image is `referable' in diabetic retinopathy (DR) screening. However, their classification accuracy degrades as the input images distributionally shift from their training distribution. Further, even if the input is not a retina image, a standard DR classifier produces a high confident prediction that the image is `referable'. Our paper presents a Dirichlet Prior Network-based framework to address this issue. It utilizes an out-of-distribution (OOD) detector model and a DR classification model to improve generalizability by identifying OOD images. Experiments on real-world datasets indicate that the proposed framework can eliminate the unknown non-retina images and identify the distributionally shifted retina images for human intervention.