arXiv Analytics

Sign in

arXiv:2107.08274 [cs.CV]AbstractReferencesReviewsResources

Lesion-based Contrastive Learning for Diabetic Retinopathy Grading from Fundus Images

Yijin Huang, Li Lin, Pujin Cheng, Junyan Lyu, Xiaoying Tang

Published 2021-07-17Version 1

Manually annotating medical images is extremely expensive, especially for large-scale datasets. Self-supervised contrastive learning has been explored to learn feature representations from unlabeled images. However, unlike natural images, the application of contrastive learning to medical images is relatively limited. In this work, we propose a self-supervised framework, namely lesion-based contrastive learning for automated diabetic retinopathy (DR) grading. Instead of taking entire images as the input in the common contrastive learning scheme, lesion patches are employed to encourage the feature extractor to learn representations that are highly discriminative for DR grading. We also investigate different data augmentation operations in defining our contrastive prediction task. Extensive experiments are conducted on the publicly-accessible dataset EyePACS, demonstrating that our proposed framework performs outstandingly on DR grading in terms of both linear evaluation and transfer capacity evaluation.

Comments: 10 pages, 2 figures, MICCAI2021 early accepted
Categories: cs.CV, eess.IV
Related articles: Most relevant | Search more
arXiv:2101.09864 [cs.CV] (Published 2021-01-25)
Applications of Deep Learning in Fundus Images: A Review
arXiv:1410.8577 [cs.CV] (Published 2014-10-30)
An Ensemble-based System for Microaneurysm Detection and Diabetic Retinopathy Grading
arXiv:1709.02270 [cs.CV] (Published 2017-09-06)
Adaptive Real-Time Removal of Impulse Noise in Medical Images