CompareSegmentations[ref,pred,vox]
compares the segmentations ref and pred for all labels that are present in both, using the voxel size vox.
CompareSegmentations[ref,pred,labs,vox]
compares only the labels labs.
CompareSegmentations[segs,vox]
with segs an association of named segmentations compares every pair of segmentations for the labels present in all of them. The output is a Dataset with a row per label and a column per measure, for the association input it is an association with an entry "name1 vs name2" per pair. The measures are set with the Method option, which can be "Dice", "Jaccard" or any method of SurfaceDistance ("Mean", "Median", "RMS", "HD", "HD95", "Std"), or a list of these.
CompareSegmentations
CompareSegmentations[ref,pred,vox]
compares the segmentations ref and pred for all labels that are present in both, using the voxel size vox.
CompareSegmentations[ref,pred,labs,vox]
compares only the labels labs.
CompareSegmentations[segs,vox]
with segs an association of named segmentations compares every pair of segmentations for the labels present in all of them. The output is a Dataset with a row per label and a column per measure, for the association input it is an association with an entry "name1 vs name2" per pair. The measures are set with the Method option, which can be "Dice", "Jaccard" or any method of SurfaceDistance ("Mean", "Median", "RMS", "HD", "HD95", "Std"), or a list of these.
Details
- The following options can be given:
-
Method {Dice, HD95} Method is an option for various algorithm-intensive functions that specifies what internal methods they should use.