Abstract:Introduction: Arterial brain vessel assessment is crucial for the diagnostic process in patients with cerebrovascular disease. Non-invasive neuroimaging techniques, such as time-of-flight (TOF) magnetic resonance angiography (MRA) imaging are applied in the clinical routine to depict arteries. They are, however, only visually assessed. Fully automated vessel segmentation integrated into the clinical routine could facilitate the time-critical diagnosis of vessel abnormalities and might facilitate the identification of valuable biomarkers for cerebrovascular events. In the present work, we developed and validated a new deep learning model for vessel segmentation, coined BRAVE-NET, on a large aggregated dataset of patients with cerebrovascular diseases. Methods: BRAVE-NET is a multiscale 3-D convolutional neural network (CNN) model developed on a dataset of 264 patients from three different studies enrolling patients with cerebrovascular diseases. A context path, dually capturing high- and low-resolution volumes, and deep supervision were implemented. The BRAVE-NET model was compared to a baseline Unet model and variants with only context paths and deep supervision, respectively. The models were developed and validated using high-quality manual labels as ground truth. Next to precision and recall, the performance was assessed quantitatively by Dice coefficient (DSC); average Hausdorff distance (AVD); 95-percentile Hausdorff distance (95HD); and via visual qualitative rating. Results: The BRAVE-NET performance surpassed the other models for arterial brain vessel segmentation with a DSC = 0.931, AVD = 0.165, and 95HD = 29.153. The BRAVE-NET model was also the most resistant toward false labelings as revealed by the visual analysis. The performance improvement is primarily attributed to the integration of the multiscaling context path into the 3-D Unet and to a lesser extent to the deep supervision architectural component. Discussion: We present a new state-of-the-art of arterial brain vessel segmentation tailored to cerebrovascular pathology. We provide an extensive experimental validation of the model using a large aggregated dataset encompassing a large variability of cerebrovascular disease and an external set of healthy volunteers. The framework provides the technological foundation for improving the clinical workflow and can serve as a biomarker extraction tool in cerebrovascular diseases.

Integration- and Separation-Aware Adversarial Model for Cerebrovascular Segmentation from TOF-MRA

Generative Adversarial Network Based Cerebrovascular Segmentation for Time-of-flight Magnetic Resonance Angiography Image

Simultaneous Intracranial Artery Tracing and Segmentation from Magnetic Resonance Angiography by Joint Optimization from Multiplanar Reformation

Adapting the Segment Anything Model for Multi-modal Retinal Anomaly Detection and Localization

Semi-supervised region-connectivity-based cerebrovascular segmentation for time-of-flight magnetic resonance angiography image

Unsupervised Cerebrovascular Segmentation of TOF-MRA Images Based on Deep Neural Network and Hidden Markov Random Field Model

A Parallel Cerebrovascular Segmentation Algorithm Based on Focused Multi-Gaussians Model and Heterogeneous Markov Random Field

Deep Learning-based Segmentation of Cerebral Aneurysms in 3D TOF-MRA using Coarse-to-Fine Framework

Accurate and robust segmentation of cerebral vasculature on four-dimensional arterial spin labeling magnetic resonance angiography using machine-learning approach

Cerebrovascular Segmentation via Vessel Oriented Filtering Network

Segmentation of Cerebral Vascular Structures Using an Active Contour Model

Cerebrovascular segmentation in phase-contrast magnetic resonance angiography by multi-feature fusion and vessel completion

Contour attention network for cerebrovascular segmentation from TOF‐MRA volumetric images

Streamlining neuroradiology workflow with AI for improved cerebrovascular structure monitoring

Training Cross-Modality Cerebrovascular Segmentation Networks with Paired Images.

Automated Cerebrovascular Segmentation and Visualization of Intracranial Time-of-Flight Magnetic Resonance Angiography Based on Deep Learning

Cross-modality cerebrovascular segmentation based on pseudo-label generation via paired data

CV-Attention UNet: Attention-based UNet for 3D Cerebrovascular Segmentation of Enhanced TOF-MRA Images

COSTA: A Multi-center TOF-MRA Dataset and A Style Self-Consistency Network for Cerebrovascular Segmentation

BRAVE-NET: Fully Automated Arterial Brain Vessel Segmentation in Patients With Cerebrovascular Disease

An Effective Brain Vasculature Segmentation Algorithm for Time-of-Flight MRA Data