Meeting Banner
Abstract #2079

Small Scale Functional Activations Using a Non-Linear, Population-Specific Brain Model and 3D EPI at 7 Tesla

Gnther Grabner Grabner1, 2, Benedikt Andreas Poser2, Siegfried Trattnig1, Markus Barth2

1Department of Radiology, Medical University Vienna, Vienna, Austria; 2Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands

Due to the availability of high field scanners and novel imaging methods, high resolution, whole brain fMRI becomes feasible. However, for performing fMRI group analyses spatial smoothing is necessary to account for inter-individual anatomical variation. Here, we investigate the possibility to build a high resolution, group specific anatomical template (model) directly from the functional T2* weighted data acquired at 7 Tesla. The purpose of this model is two fold; first, spatial smoothing can be kept at a low level and second, misregistration between distorted functional and anatomical data is avoided.

Abstract PDF

Keywords

accelerated accordance account accurate acquisition activated activation activations analyses anatomical applied assist atlas authors availability avoided axial becomes blocks brain build carried channel clusters cognition coil consisted contrast corrected correction cortex course create created defined directly distances distorted embedded enables ethics euclidean evolving excellent feasible feat field filtering finger flipped fold functional generate generation gravity hand head improves index individual institute inter investigate involves isotropic kept labeling length library linear little local localization maps materials matrix medical model models motion motor necessary normalization normalized novel orientation original oversampling part population poser possibility preprocessing processing providing radiology registration regulations repeated resolution resting scale scanner scanners sigma slice slices smoothing software spatial stepwise still structure structures subject subjects support symmetric tapping task tasks template temporal thank transformation transforming variation version versus whole