SGW-GAN: Sliced Gromov-Wasserstein Guided GANs for Retinal Fundus Image Enhancement

TL;DR

SGW-GAN introduces a novel retinal image enhancement framework that preserves intra-class structure using Sliced Gromov-Wasserstein, leading to improved visual quality and clinical task performance.

Contribution

It is the first to incorporate Sliced Gromov-Wasserstein into GANs for retinal image enhancement, balancing relational fidelity and computational efficiency.

Findings

Produces visually compelling retinal images

Achieves superior diabetic retinopathy grading accuracy

Reports lowest GW discrepancy across disease labels

Abstract

Retinal fundus photography is indispensable for ophthalmic screening and diagnosis, yet image quality is often degraded by noise, artifacts, and uneven illumination. Recent GAN- and diffusion-based enhancement methods improve perceptual quality by aligning degraded images with high-quality distributions, but our analysis shows that this focus can distort intra-class geometry: clinically related samples become dispersed, disease-class boundaries blur, and downstream tasks such as grading or lesion detection are harmed. The Gromov Wasserstein (GW) discrepancy offers a principled solution by aligning distributions through internal pairwise distances, naturally preserving intra-class structure, but its high computational cost restricts practical use. To overcome this, we propose SGW-GAN, the first framework to incorporate Sliced GW (SGW) into retinal image enhancement. SGW approximates GW via random projections, retaining relational fidelity while greatly reducing cost. Experiments on public datasets show that SGW-GAN produces visually compelling enhancements, achieves superior diabetic retinopathy grading, and reports the lowest GW discrepancy across disease labels, demonstrating both efficiency and clinical fidelity for unpaired medical image enhancement.