Leave k-out cross-vaiidation, stratified by diagnosis, study arm, and age, will be used by the groups conducting voxel-based analyses. A document describing the details of the implementation of the scheme and a table of the randomized assignments can be found below. The table contains a column of participant identifiers (RID), strata (GROUP), randomized set assignements (SET), and a training set indicator (TRAINING).
A dataset with variables as follows:
RID -4 -4 Participant roster ID
GROUP -4 -4 Strata
SET -4 -4 Randomized set assignments
TRAINING -4 -4 Training set indicator
A data frame with 822 rows and 5 variables