Hi all, The option does not exist in ft_read_cifti yet, but has been requested by others as well. I have filed it as feature request on https://github.com/oostenveld/cifti-matlab/issues/1 and will follow up there.
Removing the nans and the corresponding positions should be fine (and possibly triangles that connect those positions), as long as you don’t want to write the results back to a cifti file (with ft_write_cifti) and expect it to still be interpretable as a functional representation on the full set of brainordinates. Note that the requested functionality is also strongly desired for sparse subcortical structures (volumetric, non-surface), as the volume in that case is defined such that it spans the whole brain and not only the area with actual data. best Robert On 31 Aug 2015, at 15:20, Harms, Michael <[email protected]> wrote: > > Hi, > No such option in ft_read_cifti currently. > I've added Robert in case he wants to comment. > > You should be able to remove the NaN and get a standard ordering. If you > want to check, you can use the 'ciftiopen.m' alternative for loading CIFTI > and compare that you get the same ordering. > > cheers, > -MH > > -- > Michael Harms, Ph.D. > ----------------------------------------------------------- > Conte Center for the Neuroscience of Mental Disorders > Washington University School of Medicine > Department of Psychiatry, Box 8134 > 660 South Euclid Ave. Tel: 314-747-6173 > St. Louis, MO 63110 Email: [email protected] > > From: Matthew George Liptrot <[email protected]> > Date: Monday, August 31, 2015 6:51 AM > To: "[email protected]" <[email protected]> > Subject: Re: [HCP-Users] ft_read_cifti too many grayordinates > > Hiya, > > Is there such an option with ft_read_cifti to omit these "non-standard” > grayordinates? > > If not, and instead we use Matlab to remove the NaN rows/columns and > ‘squeeze’ the connectivity matrix, can anyone confirm that we will always get > a “standard”, correctly ordered, dense connectivity matrix that we can > directly compare across subjects? (I’m assuming that I always get a 96854^2 > matrix from ft_read_cifti and then find and remove 5572 rows/columns of > NaNs). No other gotchas, pitfalls etc? > > Cheers, > > M@ > > On 21/7/15 10:33 , "Stephen Smith" <[email protected]> wrote: > >> HI Matt - thanks - so is there an option with ft_read_cifti for excluding >> those grayordinates that are not part of the "standard" 91k set, upon >> read-in? >> Cheers >> >> >>> On 20 Jul 2015, at 17:06, Glasser, Matthew <[email protected]> wrote: >>> >>> Some of us don’t think that code should do that. You can manually remove >>> the NaNs, or use alternative software (option B): >>> >>> https://wiki.humanconnectome.org/display/PublicData/HCP+Users+FAQ >>> >>> Peace, >>> >>> Matt. >>> >>> From: <[email protected]> on behalf of Janine >>> Bijsterbosch <[email protected]> >>> Date: Monday, July 20, 2015 at 9:52 AM >>> To: "[email protected]" <[email protected]> >>> Subject: [HCP-Users] ft_read_cifti too many grayordinates >>> >>> Hi, >>> >>> When I use ft_read_cifti to load the dense connectome I end up with >>> 96854*96854 grayordinates and lots of NaN's. This appears to be because the >>> centre of the brain is included in the cortical surfaces, which I can >>> remove manually to end up with the expected 91282 grayordinates and no >>> NaN's. However, I was just wondering if this behaviour is expected and >>> fixing it manually is ok, or if this is indicative of other associated bugs >>> please? >>> >>> Thanks!! >>> >>> Best wishes, >>> >>> Janine >>> >>> ----- >>> Dr Janine Bijsterbosch >>> Postdoctoral Researcher >>> FMRIB Centre, University of Oxford >>> John Radcliffe Hospital >>> Oxford, United Kingdom >>> [email protected] >>> > >> >> > -- > Matthew George Liptrot > > Department of Computer Science > University of Copenhagen > & > Section for Cognitive Systems > Department of Applied Mathematics and Computer Science > Technical University of Denmark > > http://about.me/matthewliptrot > > _______________________________________________ > HCP-Users mailing list > [email protected] > http://lists.humanconnectome.org/mailman/listinfo/hcp-users > > > The materials in this message are private and may contain Protected > Healthcare Information or other information of a sensitive nature. If you are > not the intended recipient, be advised that any unauthorized use, disclosure, > copying or the taking of any action in reliance on the contents of this > information is strictly prohibited. If you have received this email in error, > please immediately notify the sender via telephone or return mail. _______________________________________________ HCP-Users mailing list [email protected] http://lists.humanconnectome.org/mailman/listinfo/hcp-users
