Circle of Willis Centerline Graphs: A Dataset and Baseline Algorithm

Will this be enough for your competent? doctor to verify a complete Circle of Willis and thus prevent an unneeded endarterectomy and stents.

 I would never do a carotid endarterectomy with all its' risks. I'm still of the opinion that you determine if the Circle of Willis is complete then you just close up the artery. I'm not medically trained so that opinion is obviously not worth listening to. I had a completely closed carotid artery for over 10 years with no problems with blood supply to the brain and I didn't have to worry about clots breaking off. 

Here is why your doctor needs to guarantee NO complications from endarterectomy!

Talk to your doctor about the dangers of stroke due to the endarterectomy procedure and why you would want to put inflexible metal stents in flexible arteries.

• stenting (22 posts to May 2011)

• carotid stenting (21 posts to May 2016)

• carotid artery stenting (7 posts to November 2021) 

The latest here:

 Circle of Willis Centerline Graphs: A Dataset and Baseline Algorithm

Fabio Musioa,b, Norman Juchlera, Kaiyuan Yangb, Suprosanna Shitb, Chinmay Prabhakarb, Bjoern Menzeb, Sven Hirscha {fabio.musio, sven.hirsch}@zhaw.ch 

 a Institute of Computational Life Sciences, Zurich University of Applied Sciences, Waedenswil, Switzerland arXiv:2510.13720v1 [cs.CV] 15 Oct 2025 

  bDepartment of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland 

 Abstract 

 The Circle of Willis (CoW) is a critical network of arteries in the brain, often implicated in cerebrovascular pathologies. Voxel-level segmentation is an important first step toward an automated CoW assessment, but a full quantitative analysis requires centerline representations. However, conventional skeletonization techniques often struggle to extract reliable centerlines due to the CoW’s complex geometry, and publicly available centerline datasets remain scarce. To address these challenges, we used a thinning-based skeletonization algorithm to extract and curate centerline graphs and morphometric features from the TopCoW dataset, which includes 200 stroke patients, each imaged with magnetic resonance angiography (MRA) and computed tomography angiography (CTA). The curated graphs were used to develop a baseline algorithm for centerline and feature extraction, combining U-Net-based skeletonization with A* graph connection. Performance was evaluated on a held-out test set, focusing on anatomical accuracy and feature robustness. Further, we used the extracted features to predict the frequency of fetal PCA variants, confirm theoretical bifurcation optimality relations, and detect subtle modality differences between MRA and CTA. The baseline algorithm consistently reconstructed graph topology with high accuracy (F1 = 1), and the average Euclidean node distance between refer ence and predicted graphs was below one voxel. Features such as segment radius, length, and bifurcation ratios showed strong robustness, with median relative errors below 5% and Pearson correlations above 0.95. Our results demonstrate the utility of learning-based skeletonization combined with graph connection for anatomically plausible centerline extraction. We emphasize the importance of going beyond simple voxel-based measures by evaluating anatomical accuracy and feature robustness. The dataset and baseline algorithm have been released to support fur ther method development and clinical research.