In studies comparing young and elders, the already challenging issue of spatial normalization becomes even more problematic due to the atrophied brains in older adults, rendering the nonlinear registration step more BIX 01294 clinical trial difficult. To address this issue, we substituted the nonlinear registration and smoothing steps by analyzing the fMRI data in the subjects’ native space. We used the structural T1 image acquired

at the same time as the fMRI data to perform the Inhibitors,research,lifescience,medical localization by reconstructing the T1 image through FreeSurfer. FreeSurfer’s regional masks were then used to extract the regional fMRI image from the whole-brain fMRI data. Spatial averaging of the fMRI data was performed within each Inhibitors,research,lifescience,medical region to get a single time series (signal) for each region. This approach increases power because it ensures that brain regions under examination are identical for each subject. The spatial averaging of voxels’ data within a region should not be considered equivalent to spatial smoothing. Spatial smoothing is a weighted averaging of an area in the size of the smoothing filter’s kernel (FWHM = 6–8 mm) regardless of the spatial location. Spatial smoothing thus often blends the signals from different tissue types (white

matter, gray matter, and cerebro-spinal Inhibitors,research,lifescience,medical fluid) which has a drastically damaging effect on results as is demonstrated in a recent study (Smith et al. 2011). However, spatial averaging within a region in the proposed native space method only combines Inhibitors,research,lifescience,medical the signals from voxels that are carefully delineated and are supposed to belong to only one tissue type. We demonstrated that the native space method was able to detect age-related differences in the integrity of the DMN across Inhibitors,research,lifescience,medical regions reported in studies that used the prevailing spatial coregistration method on data from larger groups of subjects (Andrews-Hanna et al. 2007). Previous studies using standard analytic

techniques with sample sizes comparable to this study could not detect these differences (Bluhm et al. 2008; Beason-Held et al. 2009; Koch et al. 2010). A similar method was proposed by Seibert and Brewer (2011) which is based on the native surface of the brain cortex rather than native volumes. In the native surface method, the vertices in nearly the center of the gray matter of the cortex were considered as seed points, whereas in this work we averaged all voxels inside the ROI to obtain the regional time series. One advantage of our proposed method is that it is easily extendable to subcortical regions, whereas for native surface method this becomes challenging. However, a comprehensive comparison of the two methods is necessary to be able to thoroughly evaluate the relative effectiveness of the two methods. As shown in Figures 6, we found seven significant age-related differences in the functional connectivity of DMN regions.

No related posts.