Abstract
Tractography atlas-based analysis (TABS) is a new diffusion tensor image (DTI) statistical analysis method for detecting and understanding voxel-wise white matter properties along a fiber tract. An important requisite for accurate and sensitive TABS is the availability of a deformation field that is able to register DTI in native space to standard space. Here, three different feature images including the fractional anisotropy (FA) image, T1 weighted image, and the maximum eigenvalue of the Hessian of the FA (hFA) image were used to calculate the deformation fields between individual space and population space. Our results showed that when the FA image was a feature image, the tensor template had the highest consistency with each subject for scalar and vector information. Additionally, to demonstrate the sensitivity and specificity of the TABS method with different feature images, we detected a gender difference along the corpus callosum. A significant difference between the male and female group in diffusion measurement appeared predominantly in the right corpus callosum only when FA was the feature image. Our results demonstrated that the FA image as a feature image was more accurate with respect to the underlying tensor information and had more accurate analysis results with the TABS method.
Highlights
Clinical neuroimaging studies increasingly rely on diffusion tensor imaging (DTI), which is unique in providing rich information regarding the properties and structure of brain white matter (WM) in vivo[1,2,3,4]
We tested the effect of different feature images on the sensitivity of the statistical analysis in the pipeline of the Tractography atlas-based analysis (TABS) method
For the results of the tract analysis, only when the fractional anisotropy (FA) image was applied to the TABS pipeline to be a feature image, the WM morphology difference in the FA value could be detected in the right corpus callosum
Summary
Clinical neuroimaging studies increasingly rely on diffusion tensor imaging (DTI), which is unique in providing rich information regarding the properties and structure of brain white matter (WM) in vivo[1,2,3,4]. The TABS method contains two major steps, the construction of a diffusion tensor (DT) template and a statistical model for each voxel along the fiber tracts[12,13]. The basic role of a feature image, which should have ultra-sensitivity to identify the corresponding regions of white matter geometry, is transformed into standard space to obtain the deformation field[6,19]. We assumed that the feature image will affect the sensitivity of the statistical analysis in the pipeline of the TABS method by affecting the spatial consistency between each subject. In order to verify the hypothesis, we attempted to construct three study-specific white matter diffusion tensor templates by using different feature images (T1 weighted image, FA image, and hFA image) based on simulation data and experimental data. Based on the difference between the two groups, the receiver operating characteristic (ROC) curve was used to calculate the accuracy of gender classification
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
