Hi AC/Figa, I just talked to David about this, and we can't answer you right now. Can you ping us in a couple of weeks on this? John Harwell will be back, and we're grappling with some of these very issues ourselves.
Donna On 07/12/2010 10:15 AM, Donna Dierker wrote: > You really do need to map to atlas or directly to either/both of these > files in the Caret distribution: > > data_files/fmri_mapping_files/Human.PALS_B12.LEFT_AVG_B1-12.FIDUCIAL_FLIRT.clean.73730.coord > > > data_files/fmri_mapping_files/Human.PALS_B12.RIGHT_AVG_B1-12.FIDUCIAL_FLIRT.clean.73730.coord > > > > But then it seems like it should be possible to downsample to one of > these simpler spheres in the distribution: > > ./data_files/CONSTRUCT.SPHERE/sphere.0.coord > ./data_files/CONSTRUCT.SPHERE/sphere.1.coord > ./data_files/CONSTRUCT.SPHERE/sphere.2.coord > ./data_files/CONSTRUCT.SPHERE/sphere.3.coord > ./data_files/CONSTRUCT.SPHERE/sphere.4.coord > ./data_files/CONSTRUCT.SPHERE/sphere.5.coord > ./data_files/CONSTRUCT.SPHERE/sphere.6.coord > ./data_files/REGISTER.LVD.SPHERE/sphere.1.LVD.74.coord > ./data_files/REGISTER.LVD.SPHERE/sphere.2.LVD.290.coord > ./data_files/REGISTER.LVD.SPHERE/sphere.3.LVD.1154.coord > ./data_files/REGISTER.LVD.SPHERE/sphere.4.LVD.4610.coord > ./data_files/REGISTER.LVD.SPHERE/sphere.5.LVD.18434.coord > ./data_files/REGISTER.LVD.SPHERE/sphere.6.LVD.73730.coord > ./data_files/REGISTER.SPHERE/sphere.0.coord > ./data_files/REGISTER.SPHERE/sphere.1.coord > ./data_files/REGISTER.SPHERE/sphere.2.coord > ./data_files/REGISTER.SPHERE/sphere.3.coord > ./data_files/REGISTER.SPHERE/sphere.4.coord > ./data_files/REGISTER.SPHERE/sphere.5.coord > ./data_files/REGISTER.SPHERE/sphere.6.coord > > How exactly to do that, I'm not sure. John Harwell is on vacation, > but perhaps David or Matt knows how. > > On 07/12/2010 09:31 AM, Figa Pelosa wrote: >> hello, >> I want to map a fMRI volume created with FLS onto a simplified >> version of the PALS surface. >> I tried to map it with "map to atlas" and then reduce the number of >> the nodes in the mesh, but that doesn't work. >> >> Will it be OK if I choose "map to caret" and use the simplified mesh? >> Or do I really need to use "map tu atlas" to obtain the best fit >> between FSL data and the caret mesh? >> >> Thanks all >> AC >> >> >> >> --- Ven 9/7/10, Donna Dierker <[email protected]> ha scritto: >> >> >>> Da: Donna Dierker <[email protected]> >>> Oggetto: Re: [caret-users] caret-users Digest, Vol 82, Issue 6 >>> A: "Caret, SureFit, and SuMS software users" >>> <[email protected]> >>> Data: Venerdì 9 luglio 2010, 21:27 >>> Hi Yan, >>> >>> It's tough, and I don't think I have a good solution. You have two >>> problems working against you, both illustrated in the >>> attached capture: >>> >>> * I can't make out by eye where the cerebellum ends and the >>> ventral occipital cortex begins, so Caret is also having a tough >>> time. This is keeping the hindbrain from disconnecting. >>> >>> * While cranking up the WMpeak might help cause the >>> cerebellum to fall off, it causes the dark white matter in motor >>> cortex to >>> threshold out of the segmentation. You have serious intensity >>> inhomogeneities, and I'm not sure how well tools like FSL's fast, >>> AFNI's >>> 3Duniformize, and other bias correction utilities will be able to >>> help you. It looks severe. >>> >>> With the first point, it is conceivable that you could use >>> MRIcron, Analyze, or something like that to zero out voxels in the >>> anatomical volume at the cerebellar-occipital junction, thereby making >>> it easier for the hindbrain to disconnect. (Caret isn't really >>> good for this purpose, because Edit Voxels requires the volume to be >>> loaded as a segmentation volume, and when you save it, the volume will >>> be binarized into 0 and 255. You need to preserve the grayscale >>> features of the anatomical volume to put it through segmentation, >>> but just >>> zero out a strip of voxels along several sagittal slices, to wipe out >>> that cerebellar bridge.) >>> >>> But I don't know how you get around the second problem, >>> because it's going to cause trouble beyond the hindbrain >>> disconnection. You won't be able to find peaks that prevent fusing >>> in brighter regions, >>> like the occipital and frontal lobes, without thresholding out the >>> parietal lobe. >>> >>> Donna >>> >>> On 07/09/2010 01:08 PM, Ma, Yan wrote: >>> >>>> Hi Donna, >>>> >>>> I uploaded both the BRIK and Head file. Their name is >>>> >>> CAZspgraxss1mmLPI_ACORIGIN_LH+orig. >>> >>>> Thanks a lot, >>>> Yan ________________________________________ >>>> From: [email protected] >>>> >>> [[email protected]] >>> on behalf of [email protected] >>> [[email protected]] >>> >>>> Sent: Friday, July 09, 2010 12:00 PM >>>> To: [email protected] >>>> Subject: caret-users Digest, Vol 82, Issue 6 >>>> >>>> Send caret-users mailing list submissions to >>>> [email protected] >>>> >>>> To subscribe or unsubscribe via the World Wide Web, >>>> >>> visit >>> >>>> http://brainvis.wustl.edu/mailman/listinfo/caret-users >>>> or, via email, send a message with subject or body >>>> >>> 'help' to >>> >>>> [email protected] >>>> >>>> You can reach the person managing the list at >>>> [email protected] >>>> >>>> When replying, please edit your Subject line so it is >>>> >>> more specific >>> >>>> than "Re: Contents of caret-users digest..." >>>> >>>> >>>> Today's Topics: >>>> >>>> 1. Re: Problem with Anatomical Volume >>>> >>> Segmentation (Donna Dierker) >>> >>>> >>>> >>> ---------------------------------------------------------------------- >>> >>>> Message: 1 >>>> Date: Thu, 8 Jul 2010 16:13:54 -0700 (PDT) >>>> From: Donna Dierker <[email protected]> >>>> Subject: Re: [caret-users] Problem with Anatomical >>>> >>> Volume Segmentation >>> >>>> To: "Caret, SureFit, and SuMS software users" >>>> <[email protected]> >>>> Message-ID: <[email protected]> >>>> Content-Type: text/plain; charset=us-ascii >>>> >>>> Hi Yan, >>>> >>>> Can you upload your cropped anatomical volume here: >>>> >>>> http://brainvis.wustl.edu/cgi-bin/upload.cgi >>>> >>>> I'll try to have a look at it in the morning, but I'll >>>> >>> be tied up all day >>> >>>> beginning at 8:30am. >>>> >>>> Donna >>>> >>>> >>>> ----- Original Message ---- >>>> From: "Ma, Yan" <[email protected]> >>>> To: "[email protected]" >>>> >>> <[email protected]> >>> >>>> Sent: Thu, July 8, 2010 2:43:52 PM >>>> Subject: [caret-users] Problem with Anatomical Volume >>>> >>> Segmentation >>> >>>> Hi there, >>>> >>>> I tried several times to segment this volume with >>>> >>> different gray and white >>> >>>> matter peak values to avoid 'Error in hind brain >>>> >>> segmentation'. When I finally >>> >>>> went through the segmentation process, the resulting >>>> >>> volume is missing large >>> >>>> chunk of brain. Attached is a snapshot of the >>>> >>> segmentation volume overlapped on >>> >>>> the approximate volume appeared when setting the gray >>>> >>> and white matter peaks. >>> >>>> Current settings are gray: 104, white: 144. >>>> >>>> Thank you, >>>> Yan >>>> _______________________________________________ caret-users mailing list [email protected] http://brainvis.wustl.edu/mailman/listinfo/caret-users
