Deep Neural Ensemble for Retinal Vessel Segmentation in Fundus Images towards Achieving Label-free Angiography

TL;DR

This paper introduces an ensemble of autoencoders for retinal vessel segmentation in fundus images, achieving high accuracy and robustness by leveraging unsupervised feature learning and diverse network architectures.

Contribution

It presents a novel ensemble autoencoder approach with two-level training and fusion strategies for improved vessel segmentation in label-free fundus images.

Findings

Maximum average accuracy of 95.33% on DRIVE dataset

Low standard deviation of 0.003 indicates consistent performance

High Kappa coefficient of 0.708 demonstrates strong agreement with ground truth

Abstract

Automated segmentation of retinal blood vessels in label-free fundus images entails a pivotal role in computed aided diagnosis of ophthalmic pathologies, viz., diabetic retinopathy, hypertensive disorders and cardiovascular diseases. The challenge remains active in medical image analysis research due to varied distribution of blood vessels, which manifest variations in their dimensions of physical appearance against a noisy background. In this paper we formulate the segmentation challenge as a classification task. Specifically, we employ unsupervised hierarchical feature learning using ensemble of two level of sparsely trained denoised stacked autoencoder. First level training with bootstrap samples ensures decoupling and second level ensemble formed by different network architectures ensures architectural revision. We show that ensemble training of auto-encoders fosters diversity in learning dictionary of visual kernels for vessel segmentation. SoftMax classifier is used for fine tuning each member auto-encoder and multiple strategies are explored for 2-level fusion of ensemble members. On DRIVE dataset, we achieve maximum average accuracy of 95.33\% with an impressively low standard deviation of 0.003 and Kappa agreement coefficient of 0.708 . Comparison with other major algorithms substantiates the high efficacy of our model.