CHALLENGE DATASETS
An extended description of the challenge dataset can be found here.
Data for Task 2
A training case consists of an image dataset showing a contrast-enhanced cerebrovascular vessel tree and one or more aneurysm seed points. Furthermore, segmentation masks are provided as both, NIFTI (.nii.gz) mask image and STL (.stl) geometry files, and are considered as ground truth.
The test cases for generating results to be uploaded automatically contain only image data.
For the generation of semi-automatic and interactive segmentation results, coordinates of a seed (the aneurysm voxel closest to its center of gravity) will be provided for the test cases (and released later), to enable local initialization.
The subset of the data to be used as training data comprises 110 datasets with 128 annotated aneurysms. It is released to the participants and can be freely partitioned e.g. for cross validation.
