Print

Print


Hi Shaquanda,

> As is often the case, I can find all sorts of theoretical information about how to do VBM (e.g. formulas, etc.) but I cannot find any examples or workshop data sets of how actually do to VBM statistics on using SPM?

Sometimes finding out exactly ho to actually do things is hardest 
(partly because there are many different ways to do it :) Generally, 
what you do after you processed your images (which has become one hell 
of a lot easier in spm5, so I really recommend you use that version) you 
do a statistical comparison by entering your data into a second-level 
statistical model. For example, if you want to compare two groups 
(patients vs. controls) you could

click on segment, select all subject

Output files: unmodulated normalized GM

run segment

click on smooth, select all wc1*.img

smoothing width, say, 12 x 12 x 12 mm

run smooth

click on second level statistical model

select 2-sample-t-test

select patients as one group, controls as the other

run stats

click on results, define your contrast (e.g., 1 -1 or so)

correct for multiple comparisons (recommended unless you have VERY 
convincing arguments to the contrary)

see group differences

publish

I admit that this is not the whole story and there may be some more 
bumps in the road but perhaps it gets you started. Also try the spm 
manual, it is very helpful to understand, for example, what options to 
choose during segmentation, or smoothing, or stats.

Good luck :)
Marko
-- 
=====================================================================
Marko Wilke                                            (Dr.med./M.D.)
                 [log in to unmask]

Universitäts-Kinderklinik              University Children's Hospital
Abt. III (Neuropädiatrie)             Dept. III (Pediatric neurology)
             Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416                   Fax: (+49) 07071 29-5473
=====================================================================