PURPOSE: To investigate white matter heterogeneity using a multichannel segmentation of a large sample of structural and diffusion magnetic resonance imaging (MRI) data. MATERIALS AND METHODS: A sample of 50 subjects was segmented using channels comprising exclusively structural (longitudinal and transverse relaxation times T1 and T2 and transverse relaxation rate R2*) and diffusion-based MRI indices (mean diffusivity and fractional anisotropy). These data were analyzed using a data driven approach in which no prior information was used. RESULTS: The analysis revealed the splitting of white matter into two subclasses in which the longitudinal fasciculi were distinguished from inferior/superior ones. The distribution of the adopted indices in the obtained clusters showed that R2* was mainly responsible for this splitting. CONCLUSION: This result supports the observation, previously hypothesized in the literature, that R2* is influenced by the fiber orientation.
PURPOSE: To investigate white matter heterogeneity using a multichannel segmentation of a large sample of structural and diffusion magnetic resonance imaging (MRI) data. MATERIALS AND METHODS: A sample of 50 subjects was segmented using channels comprising exclusively structural (longitudinal and transverse relaxation times T1 and T2 and transverse relaxation rate R2*) and diffusion-based MRI indices (mean diffusivity and fractional anisotropy). These data were analyzed using a data driven approach in which no prior information was used. RESULTS: The analysis revealed the splitting of white matter into two subclasses in which the longitudinal fasciculi were distinguished from inferior/superior ones. The distribution of the adopted indices in the obtained clusters showed that R2* was mainly responsible for this splitting. CONCLUSION: This result supports the observation, previously hypothesized in the literature, that R2* is influenced by the fiber orientation.