Particle Swarm Optimization for Great Enhancement in Semi-Supervised Retinal Vessel Segmentation with Generative Adversarial Networks

TL;DR

This paper introduces a semi-supervised retinal vessel segmentation framework that combines deep learning, GANs, self-training, and particle swarm optimization to achieve high accuracy with minimal labeled data.

Contribution

It presents the first integration of particle swarm optimization with semi-supervised learning and GANs for retinal vessel segmentation, reducing the need for extensive labeled data.

Findings

Achieves segmentation performance comparable to supervised methods with only 10% labeled data.

Demonstrates effective hyper-parameter tuning using PSO enhances semi-supervised learning.

First to combine intelligent optimization with semi-supervised learning in this domain.

Abstract

Retinal vessel segmentation based on deep learning requires a lot of manual labeled data. That is time-consuming, laborious and professional. What is worse, the acquisition of abundant fundus images is difficult. These problems are more serious due to the presence of abnormalities, varying size and shape of the vessels, non-uniform illumination and anatomical changes. In this paper, we propose a data-efficient semi-supervised learning framework, which effectively combines the existing deep learning network with GAN and self-training ideas. In view of the difficulty of tuning hyper-parameters of semi-supervised learning, we propose a method for hyper-parameters selection based on particle swarm optimization algorithm. To the best of our knowledge, this work is the first demonstration that combines intelligent optimization with semi-supervised learning for achieving the best performance. Under the collaboration of adversarial learning, self-training and PSO to select optimal hyper-parameters, we obtain the performance of retinal vessel segmentation approximate to or even better than representative supervised learning using only one tenth of the labeled data from DRIVE.