← back to statistic_image methods ·
Object methods index
Apply a sequence of thresholds (default: p < .001, p < .01, p < .05) to a
statistic_image, prune to clusters that survive the most stringent
threshold and grow them out at progressively more lenient ones, and
render the result as a layered montage with a registered fmridisplay
handle. Lets you see "where the strong effects are" and "what they grow
into at more permissive thresholds" in a single figure.
imgs = load_image_set('emotionreg');
t = ttest(imgs);
[o2, sig, poscl, negcl] = multi_threshold(t);
% Make a table of the regions at the most stringent threshold:
r = table(poscl{1});
statistic_image.threshold— single-threshold workhorseregion—poscl/negclare region objects, one per thresholdcanlab_results_fmridisplay— pre-built montage scaffolds the layered render uses