Meeting Banner
Abstract #3024

Structural Contrast Enhancements by Novel Way to Combine T1- And T2-Weighted MR Images

Masaya Misaki1, Jonathan Savitz1, 2, Vadim Zotev1, Raquel Phillips1, Han Yuan1, Kymberly D. Young1, Wayne C. Drevets1, Jerzy Bodurka1, 3

1Laureate Institute for Brain Research, Tulsa, OK, United States; 2Tulsa School of Community Medicine, University of Tulsa, Tulsa, OK, United States; 3College of Engineering, University of Oklahoma, Tulsa, OK, United States

We propose a new structural image contrast from a novel combination of T1- and T2-weighted images. A specific combination of T1 and T2 images, CI=(T1-T2)/(T1+T2), could reduce noise and enhance contrast between brain structures. The utility of the new contrast image was evaluated using automatic segmentation software: FAST in FSL for white and gray matter segmentation and FreeSurfer for segmenting subcortical regions. The combined image had significantly higher contrast between white and gray matters as well as between subcortical structures. The contrast enhancement offered by the combined image can improve accuracy of structural analysis of the human brain.

Abstract PDF Presentation Video

Keywords

acceleration accuracy accurate acquisition adjacent affects aims aligned analyses application approaches approved array assessed atlas audience automatic bias body border brain cerebral clinical clinicians college combination combine combined combining common community comparing conducted content contrast coronal corrected cover covering critically dependent differ differed discovery disorders divided electric engineering enhance enhanced enhancement enhancements enhancing equation especially estimations evaluate evaluated examine example excepting fast female field fisher form fusion general gray healthy highly holds homogeneity human improve improved in vivo independently indicating individual initialize institute intensities introduce isotropic laureate mapping matrix medical medicine necessary neuropsychiatric noise nova novel nuclei offered opposite participants performance plexuses plot potential probabilistic process propeller propose proposed proven reduce regional researchers resolution respect revealed scanner school score segment segmentation segmentations segmented sense sensitivity significantly since slice slices software spatial squared structural structures studies subjects surfer target targeted template tissue upon utility utilizing variances volume volumes warping white whole years young yuan