Published April 18, 2024 | Version v1
Other Open

TopCoW 2024 (2nd Edition): Topology-Aware Anatomical Segmentation of the Circle of Willis for CTA and MRA

Creators

  • 1. University of Zurich, Switzerland
  • 2. Zurich University of Applied Sciences, Switzerland
  • 3. Imperial College London, UK
  • 4. Harvard Medical School, USA
  • 5. University Hospital of Zurich, Switzerland
  • 6. Zhongnan Hospital of Wuhan University, China
  • 7. Geneva University Hospitals, Switzerland
  • 8. Helmholtz Munich, Germany
  • 9. Technical University of Munich, Germany
  • 10. University Hospital Berne and University of Berne, Switzerland
  • 11. German Cancer Research Center (DKFZ) Heidelberg, Germany
  • 12. Harbin Institute of Technology (Shenzhen), China
  • 13. National University of Singapore, Singapore
  • 14. University of Toronto, Canada
  • 15. University of Chicago, USA

Description

The Circle of Willis (CoW) is an important network of arteries connecting major circulations of the brain. Its vascular architecture is believed to affect the risk, severity, and clinical outcome of serious neuro-vascular diseases. However, characterizing the highly variable CoW anatomy is still a manual and time-consuming expert task. The CoW is usually imaged by two angiographic imaging modalities, magnetic resonance angiography (MRA) and computed tomography angiography (CTA), but there exist limited public datasets with annotations on CoW anatomy, especially for CTA. Therefore, we organized the TopCoW challenge in 2023 with the release of an annotated CoW dataset. We invited submissions worldwide for the CoW segmentation task, which attracted over 140 registered participants from four continents and resulted in 18 high-quality algorithms. Based on the results in 2023, for the second edition of the TopCoW challenge in 2024, we propose to expand and improve the challenge in the following three aspects: Firstly, we will increase the dataset both in terms of size and site. Secondly, we will also expand the quantitative evaluation for the anatomical segmentation task. Thirdly, we will introduce two new tasks with added annotations. To our knowledge, the newly proposed multiclass vascular tree segmentation task has never been benchmarked and has limited public annotated datasets. The newly proposed CoW variant graph classification task can help bridge vessel segmentation tasks to further exciting graph problems in brain vasculature. For the 2024 edition, TopCoW will build on the experience from 2023 with increased paired CTA and MRA dataset size, enriched annotations, expanded metrics, and two new tasks. For our new 2024 organization team, we have involved more clinical partners for data annotation and some of the participants from 2023 for pre-annotation. We look forward to organizing the second edition of the TopCoW challenge and welcoming submissions for an improved and more comprehensive CoW benchmark.

Files

TopCoW 2024 (2nd Edition)_ Topology-Aware Anatomic.pdf

Files (159.1 kB)