Deep learning-based 3D cerebrovascular segmentation workflow on bright and black blood sequences MRA

Cerebrovascular diseases are seen as a significant threat to human life and health, and the segmentation of brain blood vessels has become a scientific challenge. Therefore, the authors of this study aimed to develop a fully automated deep learning workflow capable of accurate 3D segmentation of cerebral blood vessels using convolutional neural networks (CNNs) and transformer models. The study, conducted by the first researchers to segment black blood sequence MRI cerebrovascular images and explore the ability of deep learning for the segmentation of the smallest cerebral vessels, found that the workflow demonstrated excellent performance, allowing doctors to better visualize cerebrovascular structures. Key points: The proposed deep learning-based workflow performs well in cerebrovascular segmentation tasks. Among comparison models, SwinUNETR achieved the best DSC, ASD, PRE, and SPE values in lenticulostriate artery segmentation. The proposed workflow can be used for different MR sequences, such as bright and black blood imaging. Article: Deep learning-based 3D cerebrovascular segmentation workflow on bright and black blood sequences magnetic resonance angiography Authors: Langtao Zhou, Huiting Wu, Guanghua Luo & Hong Zhou