Dear Ming-Ting,
I found some similiar effects and raised a similiar query on the SPM mailbase:
http://www.mailbase.ac.uk/lists/spm/2000-01/0107.html
I received some useful advice/comments from:
Steve Smith:
http://www.mailbase.ac.uk/lists/spm/2000-01/0108.html
and Kent Kiehl
http://www.mailbase.ac.uk/lists/spm/2000-01/0113.html
See also Jesper Andersson's posting:
http://www.mailbase.ac.uk/lists/spm/1999-11/0152.html
Good luck!
Krish
"Wu, Ming-Ting" wrote:
> Dear Group:
> I have basic questions about the option of global normalization.
> 1. what is the indication for use of none, scaling for global normalization
> in SPM99?
> 2. Why does the "ancova" in SPM96 be removed?
> 3. I compared the result of using scaling, ancova and none for a
> somatosensory stimulation, with a realigned-only dataset for references;
> please see the attached files
> Fig.1 : the result of realigned, non-smoothed, non-normalized data with a
> very simple correlation map
> Fig.2: same of fig.1, representive slice showing right parietal activation
> and its time-course.
> Fig.3: render map with ancova normalization, the parietal activation is
> there.
> Fig.4: render map with scaling global normalization, the parietal activation
> removed a lot, more deactivation spots now.
> Fig.5 render map without global normalization. large activations noted.
>
> >From figure 1, 2, my feeling is that there really have strong activation,
> corresponding to somatosensory stimulation of left leg. it seems to be close
> to that shown on the map with Ancova. The map with scaling seems to be too
> restrictive, and have more deactivation (blue color) than other maps; Is
> the deactivation true or not?
>
> Thank you in advance
>
> Sincerely
>
> Ming-Ting
>
> ............................................................................
> ...............................................................
> Ming-Ting Wu, MD Tel: (O)886-7-3422121 ext 6205,
> 6235, 6212, 6201
> Dept. of Radiology Tel: (H) 886-7-3461096 Fax:(O)
> 886-7-3468301
> Kaohsiung Veterans General Hospital
> 386 Ta-chung 1st Rd. Kaohsiung 813, Taiwan
>
> ------------------------------------------------------------------------
> Name: Fig5_nonegobal.tif
> Fig5_nonegobal.tif Type: TIFF Image (image/tiff)
> Encoding: base64
>
> Name: Fig2_realign_correlate.tif
> Fig2_realign_correlate.tif Type: TIFF Image (image/tiff)
> Encoding: base64
>
> Name: Fig3_ancova.tif
> Fig3_ancova.tif Type: TIFF Image (image/tiff)
> Encoding: base64
>
> Name: Fig4_scaling.tif
> Fig4_scaling.tif Type: TIFF Image (image/tiff)
> Encoding: base64
>
> Name: Fig1_correlate_whole.tif
> Fig1_correlate_whole.tif Type: TIFF Image (image/tiff)
> Encoding: base64
--
Dr Krish Singh, ([log in to unmask])
Magnetic Resonance and Image Analysis Research Centre, Liverpool University
Pembroke Place, Liverpool, L69 3BX, UK. Tel 0151 7945645. Fax 0151 794 5635
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|