This process aims at extracting volume measures and descriptive statistics from quantitative images and/or atlases. Feature images can be 2, 3 or 4 dimensional and supervising ROIs can be graphs, volume of labels (2D/3D) or probability maps (4D).
If a hierarchy is provided, the analysis is performed, in a hierarchical fashion, for each node of the hierarchy. This hierarchy can be lateralized (with a different label for left and right strucutres), or a lateralization mask (Input_ListOf_Side) can be provided. In both cases, full, left and right statistics can be computed. Else, only full statistics are computed.
Each input can receive single entries or lists. If several inputs are provided with lists, they must be the same size. Single entries are used for all the entries of list inputs.
The process shows on the standard output a table compiling the results. This table can be saved as a CSV file if Output_CSV_File is filled.
Volume of features for which descriptive statistics must be computed. Available statistics are mean, minimum, and maximum values as well as standard deviation. If a ROI volume is provided, theses statistics are computed in each region (each node of a hierarchy is also provided).
If the volume is 4-dimensional, statistics are computed at each time step and for the aggregation of all time steps.
If no volume of features is provided, only volume measurements computed from the ROI volume are computed.
Volumes of labels or graph volumes (.arg). If volumes of features are provided, ROI volumes are resampled in their resolution before any computation (of statistics or volumes) is performed. Volume measures might thus differ from those obtained with AimsRoiFeatures.
Input_ListOf_Side: ListOf( 3D Volume ) ( optional, input )
Volumes of side labels allow to lateralize ROIs when labels are not natively different for the left and right hemispheres. They should contain two labels :
Affine motions transforming ROI volumes towards feature volumes (or the opposite).
Motion_Direction: Choice ( optional, input )
Direction of the provided motions.
Automatic_Names: Boolean ( optional, input )
This option only works if the input images are stored in a BrainVISA database. If true, volumes are automatically named in the result table based on their attributes : [subject]_[acquistion]_[analysis]
[Analysis] Hierarchical ROI Analysis