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

arXiv: https://arxiv.org/abs/2107.08274 · HTML: https://mlsanity.com/paper/2107.08274/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 22 references that are in the map.
- First posted on arXiv 17 Jul 2021.

## Builds on (most cited first, 13 of 13)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Improved Regularization of Convolutional Neural Networks with Cutout](https://mlsanity.com/paper/1708.04552/) (2017)
- [Unsupervised Representation Learning by Predicting Image Rotations](https://mlsanity.com/paper/1803.07728/) (2018)
- [Learning Representations by Maximizing Mutual Information Across Views](https://mlsanity.com/paper/1906.00910/) (2019)
- [Unsupervised Representation Learning by Sorting Sequences](https://mlsanity.com/paper/1708.01246/) (2017)
- [Self-supervised Feature Learning for 3D Medical Images by Playing a Rubik's Cube](https://mlsanity.com/paper/1910.02241/) (2019)
- [Self-supervised Feature Learning via Exploiting Multi-modal Data for Retinal Disease Diagnosis](https://mlsanity.com/paper/2007.11067/) (2020)
- [Improving Cytoarchitectonic Segmentation of Human Brain Areas with Self-supervised Siamese Networks](https://mlsanity.com/paper/1806.05104/) (2018)
- [Zoom-in-Net: Deep Mining Lesions for Diabetic Retinopathy Detection](https://mlsanity.com/paper/1706.04372/) (2017)
- [Self-supervised Contrastive Video-Speech Representation Learning for Ultrasound](https://mlsanity.com/paper/2008.06607/) (2020)

## Built on by (most cited first, 25 of 25)

- [Dissecting Self-Supervised Learning Methods for Surgical Computer Vision](https://mlsanity.com/paper/2207.00449/) (2022)
- [Uni4Eye: Unified 2D and 3D Self-supervised Pre-training via Masked Image Modeling Transformer for Ophthalmic Image Classification](https://mlsanity.com/paper/2203.04614/) (2022)
- [Identifying the key components in ResNet-50 for diabetic retinopathy grading from fundus images: a systematic investigation](https://mlsanity.com/paper/2110.14160/) (2021)
- [SSiT: Saliency-guided Self-supervised Image Transformer for Diabetic Retinopathy Grading](https://mlsanity.com/paper/2210.10969/) (2022)
- [Cross-Field Transformer for Diabetic Retinopathy Grading on Two-field Fundus Images](https://mlsanity.com/paper/2211.14552/) (2022)
- [Contrastive learning-based pretraining improves representation and transferability of diabetic retinopathy classification models](https://mlsanity.com/paper/2208.11563/) (2022)
- [ASLseg: Adapting SAM in the Loop for Semi-supervised Liver Tumor Segmentation](https://mlsanity.com/paper/2312.07969/) (2023)
- [OTCXR: Rethinking Self-supervised Alignment using Optimal Transport for Chest X-ray Analysis](https://mlsanity.com/paper/2404.11868/) (2024)
- [AFiRe: Anatomy-Driven Self-Supervised Learning for Fine-Grained Representation in Radiographic Images](https://mlsanity.com/paper/2504.10972/) (2025)
- [Learning Enhancement From Degradation: A Diffusion Model For Fundus Image Enhancement](https://mlsanity.com/paper/2303.04603/) (2023)
- [Self-supervised Learning of Dense Hierarchical Representations for Medical Image Segmentation](https://mlsanity.com/paper/2401.06473/) (2024)
- [L-MAE: Longitudinal masked auto-encoder with time and severity-aware encoding for diabetic retinopathy progression prediction](https://mlsanity.com/paper/2403.16272/) (2024)
- [LesionPaste: One-Shot Anomaly Detection for Medical Images](https://mlsanity.com/paper/2203.06354/) (2022)
- [Detection of diabetic retinopathy using longitudinal self-supervised learning](https://mlsanity.com/paper/2209.00915/) (2022)
- [A Clinical-oriented Multi-level Contrastive Learning Method for Disease Diagnosis in Low-quality Medical Images](https://mlsanity.com/paper/2404.04887/) (2024)
- [Dive into Self-Supervised Learning for Medical Image Analysis: Data, Models and Tasks](https://mlsanity.com/paper/2209.12157/) (2022)
- [Eye tracking guided deep multiple instance learning with dual cross-attention for fundus disease detection](https://mlsanity.com/paper/2304.12719/) (2023)
- [PoCo: A Self-Supervised Approach via Polar Transformation Based Progressive Contrastive Learning for Ophthalmic Disease Diagnosis](https://mlsanity.com/paper/2403.19124/) (2024)
- [Low-Rank Adaptive Structural Priors for Generalizable Diabetic Retinopathy Grading](https://mlsanity.com/paper/2504.19362/) (2025)
- [From Retinal Pixels to Patients: Evolution of Deep Learning Research in Diabetic Retinopathy Screening](https://mlsanity.com/paper/2511.11065/) (2025)
- [TWLR: Text-Guided Weakly-Supervised Lesion Localization and Severity Regression for Explainable Diabetic Retinopathy Grading](https://mlsanity.com/paper/2512.13008/) (2025)
- [Weakly Supervised Patch Annotation for Improved Screening of Diabetic Retinopathy](https://mlsanity.com/paper/2603.03991/) (2026)
- [Machine Learning Enhanced Laser Spectroscopy for Multi-Species Gas Detection in Complex and Harsh Environments](https://mlsanity.com/paper/2605.01306/) (2026)
- [ClinReadNet: A clinical reading-inspired network for low-dose abdominal CT image quality assessment](https://mlsanity.com/paper/2606.10372/) (2026)
- [An Event-Driven Framework for Fly-Inspired Visual Motion Detection](https://mlsanity.com/paper/2607.05205/) (2026)

## Abstract (arXiv)

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.