sa (1)

segAdapter: Apply corrections to segmentations produced by the host segmentation method. Usage: ./sa input_Segmentation AdaBoost_OutPutPrefix output_Segmentation [options]

• Meanings of the parameters: input_Segmentation: Segmentation produced by the host segmentation method for the image AdaBoost_OutPutPrefix: Path and prefix to the AdaBoost files specified in ./bl output_Segmentation: Path and file name of the output corrected segmentation. options:

-f feature1.nii feature2.nii ...

Feature images for the target subject. The name pattern could be in C printf format, e.g. feature%04d.nii Then feature0000.nii will be used for label 0 and feature0001.nii for label 1, etc.

-m mask:

Specify ROI for the training images. Should be in C printf format, e.g. mask%04d.nii ROI will be derived by performing dilation on this mask.

-x label image.nii

Specify an exclusion region for the given label. If a voxel has non-zero value in an exclusion image, the corresponding label is not allowed at that voxel.

-p filenamePattern

Save the posterior maps (probability that each voxel belongs to each label) as images. The number of images saved equals the number of labels. The filename pattern must be in C printf format, e.g. posterior%04d.nii.gz

-mrf <method> [parameters]

Apply Markov Random Field prior to derive the segmentation. Options: ICM (Iterated Conditional Modes) May be followed by optional parameters in brackets, e.g., -mrf ICM[beta,iter]. beta: weight for the MRF prior, must be a non-negative number. Default: 0.1 iter: max iteration for ICM optimization. Default: 10