Meeting Banner
Abstract #2077

Effect of Distortion Correction on Group Level Statistics in FMRI of Medial Temporal Lobe

Jarle Ladstein1, Hallvard Re Evensmoen2, Asta Kristine Hberg2, Anders Kristoffersen3, Dominic Holland4, Anders M. Dale4, 5, Pl Erik Goa3

1MI Lab and Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway; 2Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway; 3Clinic of Radiology, St. Olavs University Hospital, Trondheim, Norway; 4Department of Neurosciences, The University of California, San Diego, United States; 5Department of Radiology, The University of California, San Diego, United States

Geometric distortions in functional EPI scans cause local misregistration between functional and anatomical images. This can decrease the power of functional group level analysis in regions with much distortion. To evaluate the effect of applying a distortion correction to the functional data, 25 subjects were scanned during a special encoding paradigme and the data analyzed both with and without applying a correction estimated from reverse gradient scans. In regions of interest in the entorhinal cortex the degree of distortion was high and using distortion corrected data had a positive effect in terms of increased activation volume and peak z-volume.

Abstract PDF

Keywords

acceleration according accuracy activated activation addition additional affine alleviate analyses analyzed anatomical anders applying atlas away bandwidth blue body channel circulation clear clinic cluster coil comprised confirmed corrected correction cortex cost decreases degree details distortion distortions distributed encoding environment environments evaluate even extracted feat field final flirt function functional generated geometric give gives gradient green head healthy hemispheres hence hospital identical illustrated improved included indicate intensity intersection investigate investigated isotropic larger learned learning lobe local male maps mark mask material matrices matrix medial medical motion mutual normalized observations opposing optimized orange oxford paired paradigm particular peak period pixel population portion position positive post power problem proposed radiology rage real reduced reduction registered registration require resolution reverse rigid rise runs scanning science selected sets shot side space spatial spatially statistics steps stimulus subjects sufficient summarized survived susceptibility systematic table task technology threshold tool transform transformation trio unit variation vendor virtual volume warp