Connectometry: A statistical approach harnessing the analytical potential of the local connectome

• Published in: NeuroImage (Orlando, Fla.) Vol. 125; pp. 162 - 171
• Main Authors: Yeh, Fang-Cheng, Badre, David, Verstynen, Timothy
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.01.2016; Elsevier Limited
• Subjects: Brain - physiology; Connectome; Connectome - methods; Connectometry; Consortia; Diffusion Magnetic Resonance Imaging; Diffusion MRI; Diffusion spectrum imaging; Generalized q-sampling imaging; Humans; Image Processing, Computer-Assisted; Laboratories; Mental disorders; Models, Neurological; Neural Pathways - physiology; q-Space diffeomorphic reconstruction; Quantitative anisotropy; Studies; White Matter - physiology; Diffusion spectrum imaging; Diffusion MRI; Quantitative anisotropy; Connectome; Connectometry; Generalized q-sampling imaging; q-Space diffeomorphic reconstruction
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2015.10.053

Here we introduce the concept of the local connectome: the degree of connectivity between adjacent voxels within a white matter fascicle defined by the density of the diffusing spins. While most human structural connectomic analyses can be summarized as finding global connectivity patterns at either end of anatomical pathways, the analysis of local connectomes, termed connectometry, tracks the local connectivity patterns along the fiber pathways themselves in order to identify the subcomponents of the pathways that express significant associations with a study variable. This bottom-up analytical approach is made possible by reconstructing diffusion MRI data into a common stereotaxic space that allows for associating local connectomes across subjects. The substantial associations can then be tracked along the white matter pathways, and statistical inference is obtained using permutation tests on the length of coherent associations and corrected for multiple comparisons. Using two separate samples, with different acquisition parameters, we show how connectometry can capture variability within core white matter pathways in a statistically efficient manner and extract meaningful variability from white matter pathways, complements graph-theoretic connectomic measures, and is more sensitive than region-of-interest approaches. •Here we introduce the concept of the local connectome.•Connectometry “tracks-difference” in local connectomes.•It avoids the limitations of fiber tracking in mapping end-to-end connectivity.•It can be combined with any statistical model to study feature-related variance.