Abstract
Class imbalance in diabetic retinopathy (DR) datasets, particularly the underrepresentation of advanced stages such as proliferative DR, limits classifier performance and necessitates specialized augmentation strategies. This study presents a data augmentation approach using StyleGAN2-ADAto generate stage-specific synthetic fundus images. Unlike conditional approaches requiring labels or segmentation masks, our method trains separate unconditional models for each DR severity stage, producing anatomically realistic images while preserving pathological characteristics. We propose a comprehensive evaluation framework comprising three components: quantitative assessment using distribution-based metrics (FID, KID, IS) and nearest-neighbor-based metrics (NN-SSIM, NN-PSNR); task-based evaluation through DR classification experiments with ConvNeXt Tiny, MobileNet V3 Small, and Xception architectures; and expert validation by three ophthalmologists assessing clinical plausibility. Classification experiments demonstrated improvements in overall accuracy, with ConvNeXt Tiny achieving 73.48% accuracy (+1.82% over baseline) when combining synthetic and classical augmentation. Inter-rater reliability analysis (κ = 0.723 between two evaluators) confirmed the clinical realism of generated images. Our unconditional StyleGAN2-ADA approach provides a scalable, mask-free augmentation solution addressing class imbalance in fundus imaging datasets, potentially enhancing DR screening systems in resource-constrained clinical settings.
Funding Information
  • Ph.D. Students and Young Researchers Call of Slovak University of Technology in Bratislava to start a Research Career (23-03-09-A)
  • Recovery and Resilience Plan of Slovak Republic (RRP SR) through a call (09I03-03-V05)
  • Agentúra na Podporu Výskumu a Vývoja (APVV-22-606-AI4EYE)
  • Internal Faculty of Electrical Engineering and Information Technology (FEI) Slovak University of Technology (STU) [TRUST-DIAG-Artificial Intelligence (AI)] (2026-01-24)

This publication has 48 references indexed in Scilit: