Dear Erin,
> Hello,
>
> While running topup, I'm having a strange issue where the output files are identical regardless of the valence of the vectors in acq_params.txt. For example, using the same input images, I'm getting identical output with these two different acq_params files:
>
> 0 1 0 0.06
> 0 -1 0 0.06
>
> and
>
> 0 -1 0 0.06
> 0 1 0 0.06
>
> The EPI distortion is clearly in the Y direction, but just to check, I tried also using x vectors (e.g. 1 0 0 0.06 & -1 0 0 0.06) and also tried reducing and enlarging the number in the 4th column. If the acq_params (incorrectly) indicate distortion along the x-axis, we do get different output than if it's indicated along the y-axis, but the output images are still not different for (1 0 0 0.06 ; -1 0 0 0.06) vs (-1 0 0 0.06 ; 1 0 0 0.06). Output images do look different from the input image, they just don't look different from each other when topup is run with EPI distortion in the +x vs -x or +y vs -y direction. I hope that makes sense.
what you describe is perfectly normal. One nice thing about topup is that normally one can get the acquisition parameters completely wrong, and yet the results that matter will be identical.
If you look at the estimated fieldmap (--fout output) you will see that it will be sign reversed for the cases you describe. But unless you have a particular interest in the field itself, that will not matter to you.
Similarly, if you get the readout time wrong by a factor of X, your fieldmap will be wrong by a factor of 1/X, but again the corrected images will be identical.
Jesper
>
> Any help would be much appreciated.
>
> Thanks,
> Erin
|