1. Distributional Shifts In Automated Diabetic Retinopathy Screening
- Author
-
Mong Li Lee, Jay Nandy, and Wynne Hsu
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Computer Science - Artificial Intelligence ,Computer science ,business.industry ,Computer Vision and Pattern Recognition (cs.CV) ,Deep learning ,Diabetic retinopathy screening ,Computer Science - Computer Vision and Pattern Recognition ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Pattern recognition ,Diabetic retinopathy ,medicine.disease ,Dirichlet distribution ,Machine Learning (cs.LG) ,Image (mathematics) ,symbols.namesake ,Artificial Intelligence (cs.AI) ,Classifier (linguistics) ,medicine ,symbols ,Generalizability theory ,Artificial intelligence ,business - Abstract
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., Accepted at IEEE ICIP 2021
- Published
- 2021