Dear all,

I am happy to announce the beta version of the VBM8 Toolbox which can be downloaded at:
  http://dbm.neuro.uni-jena.de/vbm8/vbm8_r330.zip

This toolbox is a an extension of the "New Segment Toolbox" in SPM8 (Wellcome Department of Cognitive Neurology) to provide voxel-based morphometry (VBM). 

You can also download a VBM8 manual, written by Florian Kurth and Eileen Lüders:
  http://dbm.neuro.uni-jena.de/vbm8/VBM8-Manual.pdf

Some general information about VBM can be found here, but the sites are not yet updated for VBM8:
  http://dbm.neuro.uni-jena.de/vbm/

New features in VBM8:
- segmentation without tissue priors
- Partial Volume Estimation (PVE)
- de-noising with non-local means filter
- integration of Dartel normalization
- processing of longitudinal data

Here is a longer (and more technical) description of the differences to SPM8/New Segment:

1.	The segmentation approach is based on an adaptive Maximum A Posterior (MAP) technique without the need for a priori information about tissue probabilities. That is, the Tissue Probability Maps are not used constantly in the sense of the classical unified segmentation approach, but just for spatial normalization3. The following MAP estimation is adaptive in the sense that local variations of the parameters (i.e., means and variance) are modelled as slowly varying spatial functions (Rajapakse et al. 1997). This not only accounts for intensity inhomogeneities but also for other local variations of intensity. 
2.	Additionally, the segmentation approach uses a Partial Volume Estimation (PVE) with a simplified mixed model of at most two tissue types (Tohka et al. 2004). We start with an initial segmentation into three pure classes: gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) based on the above described MAP estimation. The initial segmentation is followed by a PVE of two additional mixed classes: GM-WM and GM-CSF. This results in an estimation of the amount (or fraction) of each pure tissue type present in every voxel (as single voxels – given by their size – probably contain more than one tissue type) and thus provides a more accurate segmentation.
3.	Furthermore, we apply two denoising methods. The first method is an optimized block-wise non-local means (NLM) denoising filter (Coupe et al. 2008). This filter will remove noise while preserving edges and is implemented as preprocessing step. The second method is a classical Markov Random Field (MRF) approach, which incorporates spatial prior information of adjacent voxels into the segmentation estimation (Rajapakse et al. 1997).
4.	Another important extension to the SPM8 segmentation is the integration of the Dartel normalisation (Ashburner 2007) into the toolbox. If high-dimensional spatial normalisation is chosen, an already existing Dartel template in MNI space will be used. This template was derived from 550 healthy control subjects of the IXI-database (http://www.brain-development.org) and is provided in MNI space  for six different iteration steps of Dartel normalisation. Thus, for the majority of studies the creation of sample-specific Dartel templates is not necessary anymore. 

Best regards,

Christian

____________________________________________________________________________

Christian Gaser, Ph.D.
Assistant Professor of Computational Neuroscience
Department of Psychiatry
Friedrich-Schiller-University of Jena
Jahnstrasse 3, D-07743 Jena, Germany
Tel: ++49-3641-934752	Fax:   ++49-3641-934755
e-mail: [log in to unmask]
http://dbm.neuro.uni-jena.de