This sounds very odd and not anything like I've seen before.
Is your image a 3D or a 4D image.
I can imagine that with a 4D image you could get strange things apparently happening if you did not look at all timepoints.
If you still cannot figure out what is going on, please upload the image to:
All the best,
On 16 Apr 2013, at 14:40, William Pettersson-Yeo <[log in to unmask]> wrote:
> Dear FSL experts,
> I have an image in which i'm trying to find the voxels with the minimum, and maximum, signal intensities respectively.
> The command, 'fslstats input_file -R' shows the min/max intensities in the image to be -0.0181 and 0.0178.
> However, the same image set at a 70% threshold of these minimum/maximum values (i.e. only showing the either the top, or bottom, 30% of voxel intensities) is completely blank.
> Furthermore, the coordinates of the maximum voxel returned by the command 'fslstats input_image -x', refer to a voxel of intensity 0.0076.
> Lastly, using the 'cluster' command, no voxels of any value are found when the threshold (-t) is set to 0.01205 and above. This also corresponds with a manual search through the image, combined with gradually increasing the lower threshold until no voxels are visible.
> If anyone could help me understand the discrepancy between the outputs of the different commands, it would be much appreciated. Should I simply use the maximum/minimum signal intensity suggested by the cluster command?
> Many thanks,