Re: [HCP-Users] Creating Group Average Resting State fMRI

[Glasser, Matthew]

Demean and normalize functionality might be useful things to add to wb_command -cifti-merge, unless there is a simple wb_command way to do this already (we generally do this in matlab now). Matt. From: Timothy Coalson <tsc...@mst.edu> Date: Tuesday, May 20, 2014 at 12:19 PM To: Stephen Smith <st...@fmrib.ox.ac.uk> Cc: Ausaf Bari <aus...@gmail.com>, "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] Creating Group Average Resting State fMRI Should you also normalize so that the standard deviations of each timeseries match? Tim On Tue, May 20, 2014 at 12:24 PM, Stephen Smith <st...@fmrib.ox.ac.uk> wrote: Hi - One important point - you should never temporally concatenate without first demeaning the individual timeseries. Cheers. On 20 May 2014, at 10:14, Ausaf Bari <aus...@gmail.com> wrote: Ok that is something I tried recently actually - to merge all the runs across all the subjects and then do a correlation with fisher-z transform. The results didn't "feel" right in that they didn't look like typical resting state images. That is why I wanted to double check I was doing it the right way.  As an alternative could I merge the within subject runs together with -merge and do a within-subject correlation and then average across subjects? Is there any downside to this other than the huge amount of storage space that would be required (32GB per subject)? -Ausaf On Mon, May 19, 2014 at 10:04 PM, Glasser, Matthew <glass...@wusm.wustl.edu> wrote: The simplest way is to concatenate the timeseries with wb_command -cifti-merge.  You could do this for all runs of all subjects, or all runs for each subject (and then z-transform and average the connectomes across subjects).  There will be a better way in the future. Peace, Matt. From: Timothy Coalson <tsc...@mst.edu> Date: Monday, May 19, 2014 at 10:48 PM To: Ausaf Bari <aus...@gmail.com> Cc: "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] Creating Group Average Resting State fMRI Averaging together timepoints does not make sense for resting state data, it does not use stimuli, so there is no temporal correspondence between runs or across subjects.  You should do the correlation before any averaging. I'm not sure what method our pipelines use, so I'll let someone else respond as to what the recommended method is (you probably don't need to correlate each file by itself, which would use up lots of disk space). Tim On Mon, May 19, 2014 at 8:22 PM, Ausaf Bari <aus...@gmail.com> wrote: I am trying to create a resting state fMRI average over 10 subjects to view in connectome workbench. For each HCP subject, I found the following 4 files: rfMRI_REST1_RL_Atlas_hp2000_clean.dtseries.nii rfMRI_REST1_LR_Atlas_hp2000_clean.dtseries.nii rfMRI_REST2_RL_Atlas_hp2000_clean.dtseries.nii rfMRI_REST2_LR_Atlas_hp2000_clean.dtseries.nii So for 10 subjects, I will have 40 total files. Am I correct to assume that I can use the command line tool "wb_command -cifti-average" to average over all 40 files and then "wb_command -cifti-correlation" to create a dense connectome? -Ausaf -- Ausaf A. Bari MD PhD Resident Physician UCLA Medical Center Department of Neurosurgery Email: aus...@gmail.com _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org 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. -- Ausaf A. Bari MD PhD Resident Physician UCLA Medical Center Department of Neurosurgery Phone: 617-642-1929 Email: aus...@gmail.com _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users --------------------------------------------------------------------------- Stephen M. Smith, Professor of Biomedical Engineering Associate Director,  Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK +44 (0) 1865 222726  (fax 222717) st...@fmrib.ox.ac.uk    http://www.fmrib.ox.ac.uk/~steve --------------------------------------------------------------------------- Stop the cultural destruction of Tibet _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org 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 HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users