I do not have a good explanation for the splitting of the bumps,
but I can say that there is no filter applied to enhance the image
grey/white contrast. If you don't have many images in the time
series, then the adjustment step may have a slight averaging effect
that may produce an apparent increase in your signal to noise, thus
enhancing the image contrast.
Regards,
-John
| In a time series study, I only acquired one slice. I used spm99 to
| realign images of the slice and plotted histograms of signal intensity
| of the images. I noticed that the intensity distribution changed quite a
| bit comparing a pair of pre- and post-realignment images. The histogram
| of the pre-realignment image showed a 'bump' around 1000 (intensity
| unit). This bump 'splitted' in the histogram of the post-realignment
| image to two 'bumps', one centered at about 600 and the other one at
| 2000 with no overlap. That seems to imply that in addition to the
| slightly smoothing effect introduced by the realignment procedure, a
| certain filter had been applied to enhanced the image contrast between
| gray and white matters. Is this true? How can I turn the filter off so
| that I can use the original contrast for further quantitative
| processing?
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|