RetinaRegen: A Hybrid Model for Readability and Detail Restoration in Fundus Images

TL;DR

RetinaRegen is a hybrid model that improves fundus image clarity by restoring details and readability, aiding clinical diagnosis with superior image quality metrics.

Contribution

The paper introduces RetinaRegen, a novel hybrid model combining classification, diffusion, and VAE techniques for retinal image restoration.

Findings

Achieved PSNR of 27.45, SSIM of 0.956, LPIPS of 0.191 on SynFundus-1M.

Demonstrated superior restoration of key retinal regions.

Enhanced fundus image quality for better clinical diagnosis.

Abstract

Fundus image quality is crucial for diagnosing eye diseases, but real-world conditions often result in blurred or unreadable images, increasing diagnostic uncertainty. To address these challenges, this study proposes RetinaRegen, a hybrid model for retinal image restoration that integrates a readability classifi-cation model, a Diffusion Model, and a Variational Autoencoder (VAE). Ex-periments on the SynFundus-1M dataset show that the proposed method achieves a PSNR of 27.4521, an SSIM of 0.9556, and an LPIPS of 0.1911 for the readability labels of the optic disc (RO) region. These results demonstrate superior performance in restoring key regions, offering an effective solution to enhance fundus image quality and support clinical diagnosis.