 |
 |
Badreddine: Strangely, I know of no papers that assess the quality of SPM99 segmentation to other segmentation methods. In fact I know of no papers comparing the segmentation of MRI data to actual grey and white matter tissue in the brain. This always struck me as strange, considering how much is written about loss of grey matter tissue in schiophrenia. Most MRI segmentation algorithms assign voxels to a tissue class based on location or intensity. SPM99 uses location and intensity to assign each voxel a probability of being in a tissue class between 0 and 1. The output is a probability map for grey, white and csf. A fellow in our group wrote code in IDL that assigns each voxel in the map to a certain tissue class based on those probablity maps. He modelled this based on a program written by another group (see Wilke et al, Neuroimage, 13, 814-824) Another "off the shelf" MRI segmentation program is in FSL. (see http://www.fmrib.ox.ac.uk/fsl/index.html). I would defer any comments on the "quality" of coregistration in SPM99 the experts. Rob McClure -----Original Message----- From: Badreddine Bencherif [mailto:[log in to unmask]] Sent: Monday, March 11, 2002 9:23 AM To: [log in to unmask] Subject: SPM99 quality of segmentation and coregistration Doea anyone knows about papers published where authors assessed in SPM99 the quality of: 1) brain MR segmentation 2) coregistration What is the gold standard ? Is it visual comparison of data ? Thanks for your help -- Badreddine Bencherif,MD Department of Radiology Division of Nuclear Medicine Johns Hopkins University School of Medicine 601 North Caroline Street / JHOC 4230 Baltimore, MD 21287-0855 USA Voice: (410) 614-2787 Pager: (410) 283-2050 Fax: (410) 614-1977