Optimal Transport Guided Unsupervised Learning for Enhancing low-quality Retinal Images

Zhu W, Qiu P, Farazi M, Nandakumar K, Dumitrascu O, Wang Y


Real-world non-mydriatic retinal fundus photography is prone to artifacts, imperfections and low-quality when cer- tain ocular or systemic co-morbidities exist. Artifacts may result in inaccuracy or ambiguity in clinical diagnoses. In this paper, we proposed a simple but effective end-to-end framework for enhancing poor-quality retinal fundus images. Leveraging the optimal transport theory, we proposed an unpaired image-to-image translation scheme for transport- ing low-quality images to their high-quality counterparts. We theoretically proved that a Generative Adversarial Net- works (GAN) model with a generator and discriminator is sufficient for this task. Furthermore, to mitigate the incon- sistency of information between the low-quality images and their enhancements, an information consistency mechanism was proposed to maximally maintain structural consistency (optical discs, blood vessels, lesions) between the source and enhanced domains. Extensive experiments were conducted on the EyeQ dataset to demonstrate the superiority of our proposed method perceptually and quantitatively.

paper link: https://arxiv.org/abs/2302.02991

April 2023

Figures (click on each for a larger version):