Image data used in this challenge were acquired on a 3 T Philips MRI system over the past 20 years at the UMC Utrecht, the Netherlands. Data consists of 3D time-of-flight Magnetic Resonance Angiography (TOF-MRA) images from subjects who have been preventively screened for the presence of unruptured, intracranial aneurysms (UIAs) because of a positive family history for aneurysmal subarachnoid hemorrhage and/or UIA.

Participants who have registered their team and signed the confidentiality agreement are eligible to download the data on DataverseNL.

Details

The dataset is equally split into training (N=300) and test (N=300) cases, to include sufficient anatomical variation in
both subsets. Participating teams will only have access to the training cases.
 
Each case consists of one TOF-MRA image and three annotation files:
 
FileDetails
/3D_TOF_MRA.nii.gzThe 3D TOF-MRA image.
/Lippert_classes.jsonJSON file containing a dictionary of key/value pairs with the Lippert class variant for the anterior and posterior part of the circle of Willis. More details can be found here.
/artery_diameters.jsonJSON file containing a dictionary of key/value pairs with diameter annotations. More details can be found here.
/bifurcation_angles.jsonJSON file containing a dictionary of key/value pairs with angle annotations. More details can be found here.

Example image

/3D_TOF_MRA.nii.gz

Example annotations

{'Anterior class': 'a', 'Posterior class': 'e'}

/Lippert_classes.json

{'A2(L)': 2.228,  'A1(L)': 1.936,  'M1(L)': 2.183, 'ICA(L)': nan, 'Pcom(L)': nan, 'P1(L)': 2.163, 'BA': 2.832,  'VA(L)': 2.39,  'VA(R)': 2.931, 'P1(R)': 2.386, 'Pcom(R)': nan, 'ICA(R)': 2.924, 'M1(R)': 2.528, 'A1(R)': 1.916, 'A2(R)': nan}

/diameter_annotations.json

{‘A2_A1(L)’: 126.0, ‘ICA top(L)’: 99.0, ‘Pcom_ICA(L)’: nan, ‘M2_M2(L)’: nan, ‘BA top’: 96.0,

 ‘A2_A1(R)’: 135.0, ‘ICA top(R)’: 115.0, ‘Pcom_ICA(R)’: nan, ‘M2_M2(R)’: 86.0, ‘VBJ’: 113.0}

/angle_bifurcations.json

Additional data

Participants are allowed to use additional data to train their method. This must be mentioned in the description that is submitted with the method. We encourage participants to use open data or make the additional data open access.