Print

Print


Hi,

The filtering here is temporal filtering, so you cannot check it visually by looking at a single timepoint.  Your fslmaths command looks right, but there is no mean intensity left in your image, so that also makes the difference stronger visually.  However, your FEAT output seems to have been _spatially_ filtered.  Are you sure that you are selecting the correct settings for what you want?

All the best,
Mark


On 17 Aug 2017, at 16:18, Usman Sheikh <[log in to unmask]<mailto:[log in to unmask]>> wrote:


Hi FSL people,

I am running an event-related study with 8 scanning sessions per subject (i.e. 8 4D functional images). The TR is 0.85 seconds. In preprocessing, I am using ICA Aroma and then high pass filter of 50s. The problem is, when I do high-pass filtering using fslmaths, I get a different result from when I use FEAT's HPF (with everything else turned OFF).

The command I am using for filtering using fslmaths is:

fslmaths denoised_func_data -bptf 29.41 -1 filtered_func_data

And the outputs that I am getting with fslmaths as well as feat are both attached herewith.

Thanks, Usman.


<feat.png><fslmaths.png>