Image Magnification Network for Vessel Segmentation in OCTA Images

TL;DR

This paper introduces a novel image magnification network (IMN) for improved vessel segmentation in OCTA images, effectively capturing small microvascular structures and outperforming existing methods across multiple datasets.

Contribution

The proposed IMN employs a unique up-sampling encoding and down-sampling decoding design, enhancing the detection of fine retinal vessels in OCTA images.

Findings

IMN achieves an average dice score of 90.2% on three datasets

IMN outperforms existing segmentation methods

IMN demonstrates superior cross-field vessel segmentation and skeleton extraction

Abstract

Optical coherence tomography angiography (OCTA) is a novel non-invasive imaging modality that allows micron-level resolution to visualize the retinal microvasculature. The retinal vessel segmentation in OCTA images is still an open problem, and especially the thin and dense structure of the capillary plexus is an important challenge of this problem. In this work, we propose a novel image magnification network (IMN) for vessel segmentation in OCTA images. Contrary to the U-Net structure with a down-sampling encoder and up-sampling decoder, the proposed IMN adopts the design of up-sampling encoding and then down-sampling decoding. This design is to capture more low-level image details to reduce the omission of small structures. The experimental results on three open OCTA datasets show that the proposed IMN with an average dice score of 90.2% achieves the best performance in vessel segmentation of OCTA images. Besides, we also demonstrate the superior performance of IMN in cross-field image vessel segmentation and vessel skeleton extraction.