The 2015 Longitudinal MS Lesion Segmentation Challenge: Guidelines
|2015 Longitudinal MS Lesion Segmentation Challenge|
|MS Challenge Overview||MS Challenge Data||MS Challenge Evaluation|
- Segmentation algorithms must be automated. User provided initial seed points or contours are not allowed.
- Numerical input parameters may be used and should be constant for a particular test data set.
- Algorithms may use some or all of the provided contrasts in the original or preprocessed data.
- Output lesion segmentations should be in the same space as the preprocessed data.
- Other publicly available data sets may be used within the algorithm, but lesion modeling should come primarily from the provided training data or using an unsupervised model.
- Only one set of segmentations may be provided per team.
- There are no restrictions on how the algorithm is implemented in regards to platform, programming language, or dependent software libraries. Algorithms will be executed solely by the competing team with the segmentation results provided to the organizers.
- Output segmentations should be saved in NIFTI format with a label of 1 assigned to lesions and 0 otherwise. Segmentations should be in 3-D. For 3D segmentations, filenames should be as follows: "subject_timepoint_teamname.nii" with the timepoints using 2 digits (example: "test01_03_jhu.nii"). Segmentation files may be gzipped.
- The initial training data set will consist of data from 5 subjects and will include original MR images, preprocessed images, and lesion masks from 2 raters. Subsequently a test data set of 10 subjects will be released, as well as a final test data set of 5 subjects.
- Population-based approaches are allowed, but should be applied separately for test data sets 1 and 2.