Circle of Willis configurations in stroke patients

References: [1] Lars B. Hindenes et al. Variations in the circle of willis in a large population sample using 3D TOF angiography: The tromsø study. PLoS One, 15(11), 2020. [2] A. W. J. Hoksbergen et al. Absent collateral function of the circle of willis as risk factor for ischemic stroke. Cerebrovascular Diseases, 16(3):191–198, 2003. [3] T. van Seeters et al. Completeness of the circle of willis and risk of ischemic stroke in patients without cerebrovascular disease. Neuroradiology, 57:1247–1251, 2015. [4] Topcow: Topology-aware anatomical segmentation of the circle of willis for CTA and MRA. MICCAI 2023. https://topcow23.grand-challenge.org/topcow23/.
Aims: The Circle of Willis (CoW) is an arterial anastomotic system formed by the principal arteries in the brain. Anatomical variants of the CoW are frequent and occur in a large part of the general population [1]. Not only does the length and caliber of vessel segments change, also the topology of the vessel tree is subject to variation. It is hypothesized that CoW variants can have an impact on the severity of stroke [2; 3]. As a first step in studying this relation we present a pipeline for annotating and extracting the CoW vessels in 3D angiographies and show some preliminary results on the topological variability in a stroke cohort. Methods: Leveraging the power of virtual reality (VR) we developed an efficient protocol for the 3D anatomical annotation of the CoW in MRA and CTA. This way we carefully annotated the CoW in a cohort of 115 stroke patients with dual imaging modalities, allowing us to analyze the CoW topological variability. Results: The results are twofold: (1) We present our dataset containing complete CoW vessel annotations on two common angiographic imaging modalities. This dataset is made publicly available [4] to help facilitate future CoW research. (2) As part of our preliminary analysis on the CoW topological variability we find all types of variants in our stroke cohort. The most common one was the absence of the posterior communicating arteries (Pcom). A complete CoW was found in less than 30% of the cohort. Outlook: We believe that our dataset can be of benefit to both clinicians and medical imaging researchers. Good quality data is critical for large-scale CoW characterization and the study of the effect of the angioarchitecture on cerebrovascular diseases.