Hi Fellow Listers,
I have a follow-up question on this issue. Currently, I am also running
into the same problem when attempting to run multiple session ICA with
temporal concatenation.
My question is two fold: 1) How difficult a problem is it to solve? 2)
What viable alternatives do I have meanwhile?
In regards to 1), what I'm interested in is whether the problem can be
circumvented via commandline melodic usage (i.e. run each step manually and
adjust for volume difference when needed) or is it something that must be
accounted for within MELODIC?
For 2), I'm wondering how valid is it to use fslmerge to concatenate all my
BOLD files, run MELODIC with mcflirt checked and hope that things will work
out? I have pure resting-state data (without tasks) in the 5, 10, and 15
minute interval, with larger amounts of 10 minute sessions. By truncating
the 15 minute and not using the 5 minute scans, I'm essentially losing 10
minutes of data. Before I do this, I would like to find out if there is a
better way?
Many thanks,
-Shawn
On Mon, 30 Nov 2009 09:23:15 -0800, www <[log in to unmask]> wrote:
>A new release will be great. As I asked in previous email, allowing
different time points will be very helpful for combining datasets acquired
from different projects. Like I have two different datasets that are 20 time
points different.
>
>Ze
>
>--- On Sun, 11/29/09, Stephen Smith <[log in to unmask]> wrote:
>
>From: Stephen Smith <[log in to unmask]>
>Subject: Re: [FSL] Temporal Concat in MELODIC with differing number of
timepoints?
>To: [log in to unmask]
>Date: Sunday, November 29, 2009, 6:42 PM
>
>Hi - we will try to address this for future releases but for now it may
just be best to cut down scans to match the shortest, using fslroi. It's
not obvious to me why this is a problem wrt your experimental design, if
it's the case that concat-ICA makes sense in the first place?
>Cheers, Steve.
>
>
>
>On 28 Nov 2009, at 10:45, Cornelius Werner wrote:
>Hello List,
>
>it seems that I am unable to run a MELODIC temporal concatenation run
>on several subjects, if they have a differing number of timepoints.
>Unfortunately, I don't think I can cut down any excess numbers ( I
>have a block design with some random elements lying underneath). Is
>there a way to do this nevertheless?
>
>In a similar vein: melodic from the command line does not accept an
>*.fsf-file as an argument (running on Ubuntu 9.10 64bit using the
>neuro-debian repository version 3.09). Is this normal behavior?
>
>Thanks a lot na d have a nice weekend!
>Cornelius
>
>--
>Dr. med. Cornelius J. Werner
>GnuPG Public Key ID: 0x9D64BC2C
>
>Department of Neurology
>RWTH Aachen University
>Pauwelsstr. 30
>52074 Aachen
>Germany
>
>Institute of Neuroscience and Medicine
>Cognitive Neuroscience - INM3
>Research Centre Juelich
>52425 Juelich
>Germany
>
>::: Please encrypt confidential data :::
>
>
>
>---------------------------------------------------------------------------
>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)
>[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
>---------------------------------------------------------------------------
>
>
>
>
|