# Self-Supervised Equivariant Regularization Reconciles Multiple Instance Learning: Joint Referable Diabetic Retinopathy Classification and Lesion Segmentation

arXiv: https://arxiv.org/abs/2210.05946 · HTML: https://mlsanity.com/paper/2210.05946/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 21 references that are in the map.
- First posted on arXiv 12 Oct 2022.

## Builds on (most cited first, 12 of 12)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/1803.10464/) (2018)
- [Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2004.04581/) (2020)
- [Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation](https://mlsanity.com/paper/1603.06098/) (2016)
- [Wider or Deeper: Revisiting the ResNet Model for Visual Recognition](https://mlsanity.com/paper/1611.10080/) (2016)
- [CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/1811.10842/) (2018)
- [Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification](https://mlsanity.com/paper/1612.05968/) (2016)
- [Deep image mining for diabetic retinopathy screening](https://mlsanity.com/paper/1610.07086/) (2016)
- [Attention based Multiple Instance Learning for Classification of Blood Cell Disorders](https://mlsanity.com/paper/2007.11641/) (2020)
- [Identifying the key components in ResNet-50 for diabetic retinopathy grading from fundus images: a systematic investigation](https://mlsanity.com/paper/2110.14160/) (2021)

## Built on by (most cited first, 10 of 10)

- [OTRE: Where Optimal Transport Guided Unpaired Image-to-Image Translation Meets Regularization by Enhancing](https://mlsanity.com/paper/2302.03003/) (2023)
- [Optimal Transport Guided Unsupervised Learning for Enhancing low-quality Retinal Images](https://mlsanity.com/paper/2302.02991/) (2023)
- [CUNSB-RFIE: Context-aware Unpaired Neural Schr"odinger Bridge in Retinal Fundus Image Enhancement](https://mlsanity.com/paper/2409.10966/) (2024)
- [nnMobileNet: Rethinking CNN for Retinopathy Research](https://mlsanity.com/paper/2306.01289/) (2023)
- [How Effective Can Dropout Be in Multiple Instance Learning ?](https://mlsanity.com/paper/2504.14783/) (2025)
- [Attention2Minority: A salient instance inference-based multiple instance learning for classifying small lesions in whole slide images](https://mlsanity.com/paper/2301.07700/) (2023)
- [RBAD: A Dataset and Benchmark for Retinal Vessels Branching Angle Detection](https://mlsanity.com/paper/2407.12271/) (2024)
- [Learning on Bandwidth Constrained Multi-Source Data with MIMO-inspired DPP MAP Inference](https://mlsanity.com/paper/2306.02497/) (2023)
- [SGW-GAN: Sliced Gromov-Wasserstein Guided GANs for Retinal Fundus Image Enhancement](https://mlsanity.com/paper/2601.13417/) (2026)
- [Bridging Restoration and Diagnosis: A Comprehensive Benchmark for Retinal Fundus Enhancement](https://mlsanity.com/paper/2604.03806/) (2026)

## Abstract (arXiv)

Lesion appearance is a crucial clue for medical providers to distinguish referable diabetic retinopathy (rDR) from non-referable DR. Most existing large-scale DR datasets contain only image-level labels rather than pixel-based annotations. This motivates us to develop algorithms to classify rDR and segment lesions via image-level labels. This paper leverages self-supervised equivariant learning and attention-based multi-instance learning (MIL) to tackle this problem. MIL is an effective strategy to differentiate positive and negative instances, helping us discard background regions (negative instances) while localizing lesion regions (positive ones). However, MIL only provides coarse lesion localization and cannot distinguish lesions located across adjacent patches. Conversely, a self-supervised equivariant attention mechanism (SEAM) generates a segmentation-level class activation map (CAM) that can guide patch extraction of lesions more accurately. Our work aims at integrating both methods to improve rDR classification accuracy. We conduct extensive validation experiments on the Eyepacs dataset, achieving an area under the receiver operating characteristic curve (AU ROC) of 0.958, outperforming current state-of-the-art algorithms.