Connection Sensitive Attention U-NET for Accurate Retinal Vessel Segmentation

TL;DR

This paper introduces a connection sensitive attention U-Net (CSA-U-Net) that enhances retinal vessel segmentation accuracy by modeling structural properties, improving attention mechanisms, and effectively handling thin and complex vessels.

Contribution

The paper proposes a novel connection sensitive loss and an improved attention gate with DOWN-Link, integrating them into U-Net for superior retinal vessel segmentation.

Findings

Achieves state-of-the-art results on DRIVE, STARE, and HRF datasets.

Effectively preserves connectivity of thin vessels and handles abnormalities.

Outperforms existing methods in boundary and microvascular segmentation.

Abstract

We develop a connection sensitive attention U-Net(CSAU) for accurate retinal vessel segmentation. This method improves the recent attention U-Net for semantic segmentation with four key improvements: (1) connection sensitive loss that models the structure properties to improve the accuracy of pixel-wise segmentation; (2) attention gate with novel neural network structure and concatenating DOWN-Link to effectively learn better attention weights on fine vessels; (3) integration of connection sensitive loss and attention gate to further improve the accuracy on detailed vessels by additionally concatenating attention weights to features before output; (4) metrics of connection sensitive accuracy to reflect the segmentation performance on boundaries and thin vessels. Our method can effectively improve state-of-the-art vessel segmentation methods that suffer from difficulties in presence of abnormalities, bifurcation and microvascular. This connection sensitive loss tightly integrates with the proposed attention U-Net to accurately (i) segment retinal vessels, and (ii) reserve the connectivity of thin vessels by modeling the structural properties. Our method achieves the leading position on DRIVE, STARE and HRF datasets among the state-of-the-art methods.