getLesionLoad.RdComputes lesion loads from a series of images. A parcellation image (or simple mask) is required to define the regions from which to compute the lesion load.
getLesionLoad(lesions.list, parcellation, label = NA, mask = NA, binaryCheck = F, keepAllLabels = F, minSubjectPerLabel = "10%")
| lesions.list | list of antsImages or filenames. Must be binary (0 and 1 values). |
|---|---|
| parcellation | ansImage or filename of the parcellated volumes. A parcellation is an image brain regions showned as with integer values (i.e. ,1,2,3,...). |
| label | (default=NA) you can ask to get output for a specific label in the parcellation volume (i.e., label=122). |
| mask | (default=NA) if this mask is specified (antsImage or filename) lesioned voxels outside the mask are ignored. This is not a good choice, but in case you need it its there. |
| binaryCheck | (default=FALSE) check whether lesion maps are binary (0/1). Will output an error if lesion files are not binary. |
| keepAllLabels | (default=FALSE) by default labels are removed if affected in just few subjects. Setting this to TRUE will keep all labels. |
| minSubjectPerLabel | minimum number of subjects a parcel must be lesioned to keep and return it. |
outputMatrix of lesion loads between 0 and 1. 1 means 100% lesioned. Each column is a single parcel and each row a single subject. Parcel numbers are placed as column names.
lesydata = file.path(find.package('LESYMAP'),'extdata') filenames = Sys.glob(file.path(lesydata, 'lesions', '*.nii.gz')) lesions = imageFileNames2ImageList(filenames[1:10]) parcellation = antsImageRead( file.path(lesydata,'template', 'Parcellation_403areas.nii.gz')) lesload = getLesionLoad(lesions, parcellation)