- Each submission should be a single compressed archive (zip) containing the segmentations of all images. Segmentation files should be directly in the root of the archive, and not nested in a folder structure. Each segmentation should be a hdr/img file (e.g. subject-11-label.hdr and subject-11-label.img ) of type 8 bit unsigned char.
- The resolution, dimensions, and orientation of the segmentation results should be the same as the T1- and T2-weighted scans (voxel size: 1mm x 1mm x 1mm).
Results should be named as subject-11-label to subject-23-label. Within these files the segmented tissue should be labeled as follows:
0: Background (everything outside the brain)
10: Cerebrospinal fluid
150: Gray matter
250: White matter
- With each submission, a short description of the segmentation algorithm (1-2 pages) should be provided. Generally, maximal two submissions are allowed. For the second submission, please first send an email to (firstname.lastname@example.org) before your submission and state the differences with the first submission.
- Please consider the following guidelines for the content of the method description:
- Is your algorithm automatic or semi-automatic? Describe the required user inputs, and the average time spent per scan, for semi-automatic algorithms.
- Which sequences are used by your algorithm? Only T1 or T2, or both?
- List the overall structure of the algorithm in a step-wise fashion and describe each step of the algorithm in detail. Include pre- or post-processing steps, when required.
- Is the algorithm trained with example data other than the training data provided by the iSeg-2017 challenge? If so, describe the characteristics of the training data.
- If the algorithm has been tested on other datasets, you could consider including those results.
- What is the average runtime of your algorithm, and on which system is this runtime achieved?