A Multi-task Network to Detect Junctions in Retinal Vasculature

TL;DR

This paper introduces a multi-task deep learning approach for detecting retinal vessel junctions, leveraging auxiliary labels to improve accuracy and generalization across datasets.

Contribution

The proposed multi-task network uses auxiliary labels for vessel features to enhance junction detection, outperforming previous deep learning methods and generalizing well to unseen datasets.

Findings

Outperforms recent deep learning junction detection methods.

Effective across multiple datasets after single training.

Utilizes auxiliary labels to improve detection accuracy.

Abstract

Junctions in the retinal vasculature are key points to be able to extract its topology, but they vary in appearance, depending on vessel density, width and branching/crossing angles. The complexity of junction patterns is usually accompanied by a scarcity of labels, which discourages the usage of very deep networks for their detection. We propose a multi-task network, generating labels for vessel interior, centerline, edges and junction patterns, to provide additional information to facilitate junction detection. After the initial detection of potential junctions in junction-selective probability maps, candidate locations are re-examined in centerline probability maps to verify if they connect at least 3 branches. The experiments on the DRIVE and IOSTAR showed that our method outperformed a recent study in which a popular deep network was trained as a classifier to find junctions. Moreover, the proposed approach is applicable to unseen datasets with the same degree of success, after training it only once.