Hi Rob,
Can you confirm which version of FSL you are using - as this is an issue that should have been fixed in the latest version ( 5.0.9 )?
Kind Regards
Matthew
> Hi,
>
> I performed a psychophysical interaction (PPI) analysis on task data and have some significant results that correlated with a regressor of interest (age). I wanted to extract the PPI betas from that area (from an analysis without that regressor) in order to plot the betas against age in a scatterplot. However, I get some strange beta values.
>
> This was the function I used (example of one subject):
> fslmeants -i ${homedir}/sub005/sub005_FH_PPI_bilateral_PPA_block_ss.feat/stats/cope3.nii.gz -o ${homedir}/sub005/cope_meants_PPI_lfO_mask_bilateral_PPA_attendH_ss.txt -m ${homedir}/sub005/lfO_sph5_indv.nii.gz
>
> (input is the cope, output is a textfile, mask is a binarised spherical mask in individual subject space. The output gives one value).
>
> Values extracted for a number of the subjects:
>
> -0x1.d91d98518f9c2p-4
> -0x1.bbc0e5314c1bbp-3
> -0x1.2a93426b7baf7p-1
> -0x1.8dd2bb754e25bp-4
> -0x1.806a2ad2p-5
> -0x1.c83f877dc4771p-1
> -0x1.eadcd70dd9767p-3
> -0x1.62e01a2a8d9dfp-4
> -0x1.5867051ea2577p-2
> 0x1.47c63977e327bp-3
> -0x1.1b2e248c565c9p-1
> -0x1.d28a31dd3dcbp-2
> -0x1.9a15068841d42p-1
> -0x1.6e3cf178e38e3p-3
> -0x1.4a43e02cp-2
> 0x1.35f4ebe8eep-2
>
>
> Note that when I load up the individual PPI images in FSLview the numbers are fine (values such as: -0.2 to 0.2). I have extracted zstats and tstats and those seem to be fine. I have done this with non-PPI data and the function also works fine.
>
> Any ideas? Is this just some strange engineering notation?
>
> Thanks!
> Rob
|