Jun-Cheng Weng1,2, Wen-Yih Isaac Tseng3,4
1School of Medical Imaging
& Radiological Sciences, Chung Shan Medical University, Taichung, Taiwan;
2Department of Medical Imaging, Chung Shan Medical University Hospital,
Taichung, Taiwan; 3Center for Optoelectronic Biomedicine, National
Taiwan University College of Medicine, Taipei, Taiwan; 4Department
of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan
The corpus callosum (CC) is the main fiber tract connecting bilateral cerebral hemispheres, serving information transfer and processing in various cognitive functions. Different CC regions might be affected differently in the development of disease, and their structural parameters such as size and shape might associate with cognitive or functional tests involved in different modes of interhemispheric interactions. Previously we proposed a novel magnetic resonance imaging method called q-planar imaging (QPI), which could in vivo map the relative axonal diameters and density of CC in human brain. We also studied the optimum parameters, cutoff values of diffusion sensitivity b and sampling number, to apply this technique to clinical study. In the study, to further visualize the difference in the computed axonal diameter and density distribution for each voxel, we used cluster analysis to segment the CC based on the QPI parameters, displacement and probability. Correlation analysis was also performed between diffusion spectrum imaging (DSI) and QPI derived parameters. Our cluster results demonstrated that QPI produced reasonable segmentation of relative axonal diameters and density of CC in normal human brain. Poor to moderate correlations between the DSI indices and the parameters derived from QPI implied the incompatibility of the two methods.