Dear Simon,

Niels is absolutely right that you will need to supply a config file (or specify the different options on the command line).

Did I understand you right that you have only three slices and that you have 20mm gaps between them? If so, you will have to turn off motion detection as any interpolation between slices would be meaningless. You do that by setting --estmov=0.

Let me know how that works for you.

Jesper


On 26 Sep 2017, at 15:54, Niels Bergsland <[log in to unmask]> wrote:

Hi Simon,
I don't have experience working with data gapped like that.However, I do see that you're missing the "--config=b02b0.cnf" parameter on the command line. So I'd add that and see if you get any further along the way. Also, just to double check, the dwi_rl_lr file is just the b=0 images and does not actually have diffusion-weighted volumes in it as well, right?

Best,
Niels

On Tue, Sep 26, 2017 at 4:21 PM, Simon Levy <[log in to unmask]> wrote:
Dear all,

Thank you very much for sharing such useful processing tools, it is of a great help!

However, I am having trouble running topup on an axial diffusion dataset although I think I respected all necessary input parameters. My gut feeling is that the large gap (20mm) between the 3 axial slices making up the dataset is causing trouble to the script. However topup should be able to correct EPI readout-related distortions on such a dataset, either slice-by-slice or interpolating in-between slices.

I am ready to share the data, I think it might be the easiest and most efficient way of fixing this issue, as the error message is not very specific.

Here is the command line used:
topup --imain=dwi_rl_lr.nii.gz --datain=acqparams.txt --out=topup_results --fout=field_estimate --iout=dwi_topup_corrected
And here is the output:
Segmentation fault: 11

Many thanks for your help.

Best regards,
Simon Levy



--
Niels Bergsland
Integration Director / Research Assistant Professor of Neurology
Buffalo Neuroimaging Analysis Center / University at Buffalo
100 High St. Buffalo NY 14203
[log in to unmask]