Tasks
Task 1: automated classification of the circle of Willis configuration variants
The purpose of this task is to classify circle of Willis (CoW) configuration variants on TOF-MRA images. Desired classification outputs contain two classes per image: one for the anterior CoW variant and one for the posterior CoW variant.
Task 2: automated quantification of the Circle of Willis artery diameters and bifurcation angles
Background
Unruptured, intracranial aneurysms (UIAs) typically develop on bifurcations of the circle of Willis (CoW). Rupture of an UIA leads to aneurysmal subarachnoid hemorrhage, a severe subtype of stroke with significant socio-economic burden1. Variation in CoW configuration, bifurcation angles and diameters of CoW arteries are one of the few established morphological imaging markers predictive of UIA development2. However, accurate risk prediction models to identify individuals at the highest risk of UIAs are still lacking. Automated methods to assess CoW configurations and artery morphology would facilitate the search for new risk factors for UIA development in large patient cohorts.
A full description of the motivation behind this challenge and references can be found here.
Circle of Willis. Left: inferior view of the brain arteries. Right: schematic representation of the circle of Willis.
Method
Participants will containerize their algorithms with Docker and submit these to the organizers. Detailed instructions and easy-to-follow examples will be provided and, if needed, the organizers will help with containerization. The organizers will run the techniques on the test data. This guarantees that the test data remains secret and cannot be included in the training procedure of the techniques. The container / Docker infrastructure guarantees standardized testing in the coming years, and is supported by the UMC Utrecht. The main organizer and the UMC Utrecht agree to commit themselves to future support of this challenge.
Terms of participation
The terms of participation can be viewed here. This agreement needs to be signed on DataverseNL in order to activate your registration and get access to the data.
References