My (sometimes imperfect) understanding is that the paint file, Macaque.F99UA1.BOTH.COMPOSITE.73730.paint, covers both the left and right hemispheres, except for the last two columns, which identify the medial wall for the right and left separately. Note that almost all of the other column-names have Macaque.F99UA1.R in them, which suggests they were developed for the right hem in particular; however, David told me the macaque is far more symmetric than humans are, and the "BOTH" in the filename, together with lack of a left-hem counterpart, suggest that this paint file can be used for both hemispheres.
I can tell you that both left and right hems have been registered to a common target, so the nodes are in register across hemispheres, within the limits of our surface-based registration methods. I'd use this paint file/column for both hemispheres. But you'll have to run the steps below separately using Macaque.F99UA1.LEFT.FIDUCIAL.Std-MESH.73730.coord for the left, where you would use Macaque.F99UA1.RIGHT.FIDUCIAL.Std-MESH.73730.coord for the right. Hope that isn't too confusing, but your question can be taken multiple ways. On 05/12/2011 05:14 PM, David Grayson wrote: > Wow, thanks Donna! That was awesome. I did that and then split up the ROIs > into individual files. Worked perfectly. > > So one question about that: do those ROIs cover both hemispheres, or just the > right? Is there a way to check? > > Thanks again, > > David > > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of Donna Dierker > Sent: Thursday, May 12, 2011 7:19 AM > To: Caret, SureFit, and SuMS software users > Subject: Re: [caret-users] Generating lots of ROIs at once? > > Hi David, > > I'm going to assume you want a 3D label/ROI/paint volume, rather than > writing the node:scalar pairs to a node_numx1x1 nii volume (or as I > sometimes see it written as mgz). > > I used this dataset to do it: > > CARET_TUTORIAL_SEPT-06 > http://sumsdb.wustl.edu/sums/directory.do?id=6585200 > > Specifically the MACAQUE/Macaque.F99.RIGHT.DEMO.73730.spec spec file. > > Then I did: > > Load scenes > Select scene 8. Lewis-VE areas, borders on Flat, Fiducial, VeryInflated > (Macaque F99 RIGHT > > Attributes: Paint: Convert Paint Column to Paint Volume > Volume Selection: Space: MACAQUE-F99 > OK > D/C Overlay/Underlay Volume: Paint primary overlay > File: Save Data File: Volume Paint File: > Macaque.F99UA1.BOTH.LewisVE00.4dfp.ifh > > Normally, I'd choose NIFTI over 4dfp.ifh, because the latter is > generally only readable within wustl.edu, while the formerly is a > broadly recognized standard; however, to my knowledge, the NIFTI > standards body has not adopted a standard way to represent ROI index to > structure/meaning lookup, much less structure/meaning to color. So > until that happens, we are stuck with the two work-arounds we know: > > * AFNI head into which Caret has inserted a LUT_NAMES tag (but other > software like AFNI won't recognize) > * ifh, which already had a mechanism for representing index to > structure/meaning lookup > > Since you said nifti or 4dfp.ifh, in this case I recommend ifh, because > it stores the lookup. Note that the intensities are off by 2, e.g. > where the header says this: > > region names := 0 LVE00_VOT > region names := 1 LVE00_V4 > region names := 2 LVE00_TEa_m > ... > > The actuall intensity of the voxels intersecting the LVE00_VOT nodes is > 2 -- not 0. So add 2 to all the intensities in the header, and direct > all questions to Avi Snyder. ;-) > > Donna > > On 05/11/2011 08:58 PM, David Grayson wrote: > >> Hey everybody, >> >> >> >> I am working on a project generating macaque surface ROIs in caret. I >> was wondering if there is a way to generate ROIs in nifti format (or >> 4dfp.ifh format) for each node within a paint category. For instance, >> I have started generating individual ROIs corresponding to each node >> in the LVE00 surface category. However, since I have to do this one at >> a time, it really takes a while L. Ultimately we would like to make >> comparisons of all the ROIs of many different surface categories. So I >> am wondering, is there a way to produce all the ROIs of a paint >> category at once, instead of one by one? >> >> >> >> Thank you, >> >> >> >> David >> >> > > _______________________________________________ > caret-users mailing list > [email protected] > http://brainvis.wustl.edu/mailman/listinfo/caret-users > > _______________________________________________ > caret-users mailing list > [email protected] > http://brainvis.wustl.edu/mailman/listinfo/caret-users > _______________________________________________ caret-users mailing list [email protected] http://brainvis.wustl.edu/mailman/listinfo/caret-users
