BV-GAN: 3D time-of-flight magnetic resonance angiography cerebrovascular vessel segmentation using adversarial CNNs

Abstract

Purpose: Cerebrovascular vessel segmentation is a key step in the detection of vessel pathology. Brain time-of-flight magnetic resonance angiography (TOF-MRA) is a main method used clinically for imaging of blood vessels using magnetic resonance imaging. This method is primarily used to detect narrowing, blockage of the arteries, and aneurysms. Despite its importance, TOF-MRA interpretation relies mostly on visual, subjective assessment performed by a neuroradiologist and is mostly based on maximum intensity projections reconstruction of the threedimensional (3D) scan, thus reducing the acquired spatial resolution. Works tackling the central problem of automatically segmenting brain blood vessels typically suffer from memory and imbalance related issues. To address these issues, the spatial context of the segmentation consider by neural networks is typically restricted (e.g., by resolution reduction or analysis of environments of lower dimensions). Although efficient, such solutions hinder the ability of the neural networks to understand the complex 3D structures typical of the cerebrovascular system and to leverage this understanding for decision making.Approach: We propose a brain-vessels generative-adversarial-network (BV-GAN) segmentation model, that better considers connectivity and structural integrity, using prior based attention and adversarial learning techniques.Results: For evaluations, fivefold cross-validation experiments were performed on two datasets. BV-GAN demonstrates consistent improvement of up to 10% in vessel Dice score with each additive designed component to the baseline state-of-the-art models.Conclusions: Potentially, this automated 3D-approach could shorten analysis time, allow for quantitative characterization of vascular structures, and reduce the need to decrease resolution, overall improving diagnosis cerebrovascular vessel disorders.

Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography

“…This methodology can be adapted to train segmentation networks, where a generator is tasked with generating segmented images, and the discriminator differentiates between the predicted segmentation maps and the authentic ones. This modification prompts the segmentation network to yield more anatomically 112 . Their BV-GAN model utilised attention techniques, allowing the generator to focus on voxels more likely to contain vessels.…”

“…The number of slices and size of each slice segmentation were as followed (Z, X, Y): kidney 1: (2279, 1303, 912); kidney 2: (2217, 1041, 1511); and kidney 3: (501, 1706, 1510). The network topologies automatically generated for 3D full resolution configuration for experiment 1 were a patch size (Z, X, Y)of [112,112,192], a batch size of 2 and the number of pool per axis (Z, X, Y) were [4,4,5].…”