DRStageNet: Deep Learning for Diabetic Retinopathy Staging from Fundus Images

TL;DR

DRStageNet is a deep learning model that improves diabetic retinopathy staging from fundus images by enhancing generalization across diverse datasets and providing explainability, addressing distribution shift challenges.

Contribution

The paper introduces DRStageNet, a novel approach using multi-source domain fine-tuning of a pretrained vision transformer to improve generalization in DR staging across diverse datasets.

Findings

Outperforms two state-of-the-art benchmarks.

Demonstrates robustness across multiple datasets.

Provides high-resolution explainability heatmaps.

Abstract

Diabetic retinopathy (DR) is a prevalent complication of diabetes associated with a significant risk of vision loss. Timely identification is critical to curb vision impairment. Algorithms for DR staging from digital fundus images (DFIs) have been recently proposed. However, models often fail to generalize due to distribution shifts between the source domain on which the model was trained and the target domain where it is deployed. A common and particularly challenging shift is often encountered when the source- and target-domain supports do not fully overlap. In this research, we introduce DRStageNet, a deep learning model designed to mitigate this challenge. We used seven publicly available datasets, comprising a total of 93,534 DFIs that cover a variety of patient demographics, ethnicities, geographic origins and comorbidities. We fine-tune DINOv2, a pretrained model of self-supervised vision transformer, and implement a multi-source domain fine-tuning strategy to enhance generalization performance. We benchmark and demonstrate the superiority of our method to two state-of-the-art benchmarks, including a recently published foundation model. We adapted the grad-rollout method to our regression task in order to provide high-resolution explainability heatmaps. The error analysis showed that 59\% of the main errors had incorrect reference labels. DRStageNet is accessible at URL [upon acceptance of the manuscript].