Hi,

OK, that command line looks OK.

When you load it into matlab, what size does the variable motpars have?

Also, can you send the par file output (just attach it to the email) as this would be helpful to see.

All the best,

Mark

On 2 Jul 2014, at 15:17, Damji Heo <[log in to unmask]> wrote:

When I ran mcflirt, I put:

mcflirt -in filename -out bold_mcf -plots -report

and I got the bold_mcf.par file.

I uploaded this file to matlab and read it as ascii, 

I put the command line as 

subplot (2,1,1)

plot (motpars (:, 1:3))

title ('Rotations', 'fontsize', 14)

leg=legend ('x', 'y', 'z');

set(leg, 'fontsize', 14)

it's working but, but when I try to run:

subplot (2,1,2)

plot (motpars (:, 4:6))

it keeps saying that 'Index exceeds matrix dimensions' Is it because I did not do plot the motion parameters correctly?

Damji

2014-07-02 1:48 GMT-05:00 Mark Jenkinson <[log in to unmask]>:

Dear Damji,

You still have not answered my original question:

  What exact command line are you running?

It is very difficult to help without this information.

All the best,

Mark

On 1 Jul 2014, at 16:35, Damji Heo <[log in to unmask]> wrote:

Yes, I useed mcflirt, not flirt, and the .par file has 4X4 volumes value. 

when I run: 

subplot (2,1,1)

plot (motpars (:, 1:3))

title ('Rotations', 'fontsize', 14)

leg=legend ('x', 'y', 'z');

set(leg, 'fontsize', 14)

it's working but, but when I try to run:

subplot (2,1,2)

plot (motpars (:, 4:6))

it keeps saying that 'Index exceeds matrix dimensions'

I checked the nii.gz file with  fslnvols, and it says 384, the number of original dicom files.

So I don't know why .par file doesn't have enough volumes.

Damji

2014-07-01 2:45 GMT-05:00 Mark Jenkinson <[log in to unmask]>:

Hi,

What exact command line are you running?

Are you using "mcflirt" and not "flirt"?

If you use mcflirt then it should work and output a file that ends with .par (not motpars.mat, which ends in .mat, but something like blahblah.par).  This .par file is the one you need.

All the best,

Mark

On 30 Jun 2014, at 23:27, Damji Heo <[log in to unmask]> wrote:

i checked the volumes, and it says it's 384. the original dicom files have 384 images. Based on what it says, I think this nii.gz file has all time points. I think this means that dicom to nifti conversion ran properly. But still motpars.mat file has <4x4double> value. 

Damji

2014-06-27 21:08 GMT-05:00 Jeanette Mumford <[log in to unmask]>:

Hi Damji,

If you type

fslnvols filename

does that say you have as many volumes as you think?  My guess is your 4D image doesn't have as many time points as you think it should.  You'll need to check that your dicom to nifti conversion ran properly if the number of volumes is off.

Cheers,

Jeanette

On Fri, Jun 27, 2014 at 12:52 PM, Damji Heo <[log in to unmask]> wrote:

Hello, 

I keep getting a problem whenever I try to run motion parameters using fsl.

The motpars I got is 4x4 volumes, but I need to get more volumes so that I could plot 4:6.

I kept running this command line mcflirt -in filename -out bold_mcf -plots -report

but its still same. 

So, I tried this time to use mcflirt tab, and modified the command lines as 

/Applications/fsl-5.0.4/bin/flirt -in /filename -ref /Applications/fsl-5.0.4/data/standard/MNI152_T1_2mm brain -out bold_mcf -plots -bins 256 -cost corratio searchrx -90 90 searchry -90 90 -searchrz -90 90 -dof 6 -interp trilinear

and I keep getting 'Unrecognised option b' 

What is the way to get more volumes for motpars?

Thanks in advance,

Damji