Cerebral artery segmentation Challenge

Stroke is one of the leading causes of death worldwide. The major etiology of ischemic stroke is cerebrovascular diseases characterized by disorders in the cerebral vasculature, such as stenosis and occlusion of arteries. Accurate assessment of cerebrovascular disorders is important for the diagnosis, treatment, and intervention of
cerebrovascular diseases. Magnetic resonance angiography (MRA) is widely used to visualize the cerebral arterial tree for disease diagnosis. Accurate cerebral artery segmentation of MRA is significant for quantitative analysis of cerebrovascular diseases, such as estimation of degree of luminal stenosis. However, manual segmentation is
challenging even for experts given the complex network of cerebral arteries with substantial inter-individual variations, and weak signals in small vessels due to slow or in-plane blood flow.

Time-of-flight (TOF) MRA is the most widely used non-invasive imaging technique to depict the anatomy of the cerebrovascular tree without use of contrast agents. Private datasets and annotations are commonly used in recent cerebral artery segmentation studies, whereas open-accessible large-scale TOF-MRA data with well-labeled cerebral arteries are rare, hindering the development and validation of reliable automatic cerebral artery segmentation algorithms. Hence, we attempt to host the first cerebral artery segmentation challenge in MICCAI 2023.

In this challenge, the task is to segment cerebral arteries from 3D TOF-MRA images acquired from patients with symptomatic intracranial artery stenosis. Precise segmentation of cerebral arteries will be helpful for identification and quantitative characterization of stenosis and help the diagnosis of vascular diseases such as arteriosclerosis, dissection, arteritis, moyamoya disease, and reversible cerebral vasoconstriction syndrome.