Differences

This shows you the differences between two versions of the page.

--- creating_and_combining_masks_from_vtcs [2010/06/23 04:10] – two typos... jochen
+++ creating_and_combining_masks_from_vtcs [2010/06/25 18:48] (current) – changed comparison operator to greater or equal jochen
@@ Line 26: / Line 26: @@
   * clear mask object
-===== Code =====
+==== Code ====
 <code matlab create_msk_from_vtcs.m>% settings
 % - intensity threshold: values < 10 are treated as relative threshold!
@@ Line 121: / Line 121: @@
 mdm.ClearObject;</code>
+===== Alternative  algorithm =====
+In case the MSK files/objects have already been created (e.g. as gray matter mask per subject), this is a way to average them. **This will only work if the masks are in the same space, e.g. large TAL box!**
+  * making settings (i.e. percentage threshold for combined mask)
+  * locating all mask files to be averaged
+  * loading all masks and adding the data
+  * thresholding the data
+  * saving new mask
+==== Code ====
+<code matlab combine_existing_msks.m>% settings (80% of masks must have a set voxel)
+gthresh = 0.8;
+% locate masks
+%  - this could be enhanced by changing the pattern, e.g. using
+%    'SK*_TAL*.msk' to locate only masks of subjects beginning with SK in TAL space
+%  - additionally, the startfolder and depth could be altered
+mskfiles = findfiles(pwd, '*.msk');
+% alternative: mskfiles = findfiles([pwd '/SK*/VTC*/RUN*'], 'SK*TAL*.msk', 'depth=1');
+% loop over masks
+msk = [];
+for mc = 1:numel(mskfiles)
+    % clear old mask
+    if ~isempty(msk)
+        msk.ClearObject;
+    end
+    % load mask
+    msk = xff(mskfiles{mc});
+    % for first mask
+    if mc == 1
+        % copy data
+        mask = uint16(msk.Mask);
+    % otherwise
+    else
+        % add data
+        mask = mask + uint16(msk.Mask);
+    end
+end
+% threshold mask
+mask = uint8(mask >= uint16(ceil(gthresh * numel(mskfiles))));
+% store
+msk.Mask = mask;
+% save
+msk.SaveAs;
+% clear object
+msk.ClearObject;</code>