The link and password to download data will appear at the bottom of this page when you log in.
This page details information about the available training data and more information about contents of the files can be found here.
Each case consists of one TOF-MRA and one structural MR image.
Train Dataset: 113 cases
1.93 cases containing at least one untreated, unruptured intracranial aneurysm
- 35 baseline and 35 follow-up of the same subject
- 23 unique subjects
2. 20 scans of subjects without intracranial aneurysms
Test Dataset (not released): 141 cases
1.117 cases containing at least one untreated, unruptured intracranial aneurysm
- 43 baseline and 43 follow-up of the same subject
- 29 unique subjects
2. 26 cases of subjects without intracranial aneurysms
For each subject the following files are provided:
| File | Details |
|---|---|
| /orig/TOF.nii.gz | The original TOF-MRA image. This image is used to manually delineate the aneurysms and participants should provide the results in this image space. |
| /orig/struct.nii.gz | The original structural image. |
| /orig/reg_struct_to_TOF.txt | Transformation parameters used to align the structural image with the TOF image. Participants can use this to do the transformation themselves, in case they use the structural image for processing. Results will be evaluated in the TOF MRA space. Call: transformix -in IMAGE.EXT -out /output -tp /orig/reg_struct_to_TOF.txt |
| /orig/ScanParams_TOF.json | Scan parameters of original TOF-MRA image |
| /orig/ScanParams_struct.json | Scan parameters of original structural image |
| /orig/struct_aligned.nii.gz | The structural image aligned with the TOF MRA image. |
| /pre/struct.nii.gz | Bias field corrected structural image. |
| /pre/TOF.nii.gz | Bias field corrected TOF MRA image. |
| /pre/struct_aligned.nii.gz | Bias field corrected structural image, aligned with TOF MRA |
| /aneurysms.nii.gz | Manual reference standard, only available for training data. |
| /location.txt | Text file containing centre of massvoxel coordinates x,y,z and radius for all unruptured, untreated aneurysms in the TOF-MRA . Please note: this file does not contain the locations of the treated aneurysms as they will not be assessed for evaluation. |
Example images
Folder: orig
This folder contains all of the original images which have been anonymised. The structural image (T1, T2 or FLAIR) has been aligned to the TOF image using elastix with the following parameter file.
Scan parameters are saved as a dictionary in json files within the orig folder. These scan parameters related to the original images. Included scan parameters are Modality, MagneticFieldStrength, RepetitionTime and EchoTime. Modality was grouped as either T1, T2 or FLAIR based on the protocol. Please be aware that the protocols between images, even within the same group of modalities, may be different. There also may be some protocols in the test data that are not in the training data. An example json file is provided below:
{"Modality": "T2", "MagneticFieldStrength": 3.0, "RepetitionTime": 3849.05, "EchoTime": 80.0}Folder: pre
All images were preprocess using n4biasfieldcorrection to correct bias field inhomogeneities. This was performed using the algorithm from ANTs .
Text file
The text file /location.txt is a text file containing 3D centre of mass voxel coordinates x,y,z and the radius for all unruptured, untreated aneurysms in the TOF-MRA . Each line in the text file is the measures for a different aneurysm in the scan. Co-ordinates are provided in voxel coordinates of the original TOF-MRA. The radius, r, is provided in millimeters and is rounded up to two decimal places. An example location.txt file containing two aneurysms is provided below. Note that in the text file each aneurysm location is on a separate line and the format is x, y, z, r
377, 221, 89, 3.0
233, 241, 42, 2.13Note: for the images containing no aneurysms the text file will be blank.
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.
Download
Participants who have registered their team and signed the confidentiality agreement are eligible to download the data.
You do not have access to this post. Links to download the data will appear here after registration.
Please register and log-in to view password.