Dear SPMers,
By reading other postings I understood the segmentation in VBM8 is purely intensity-based.
And the classic SPM8 segmentation (New segment) is only used for removing non-brain parts of the T1 image (skull-stripping). (Does this process need a prior information about tissue probabilities?)
So that means MAP technique is done by using only image intensity?
Any suggestion would be highly appreciated.
Thank you so much,
Naoki Hirabayashi
>>>>>>Subject:
MAP technique
From:
Naoki Hirabayashi <[log in to unmask]>
Date:
Thu, 12 Dec 2013 11:43:34 +0000
Dear VBMers,
I have a question about MAP technique used in segmentation of VBM8 toolbox.
It says in the technical information of VBM8 that this technique doesn't need a prior information about tissue probabilities.
But I think MAP technique which uses bayes' rule needs prior probability.
They use prior information in segmentation(http://dbm.neuro.uni-jena.de/vbm/segmentation/).
Does this website explains about VBM8 segmentaion or SPM8 New Segmetation?
Doesn't VBM8 segmentation need tissue probability maps?
Please tell me about MAP technique.
Any suggestion would be highly appreciated.
Thank you so much,
Naoki Hirabayashi
|