Multiscale characterization of white matter microstructural geometry and organization to improve diffusion MRI modelling

Early diagnosis and noninvasive monitoring of neurological disorders require sensitivity to cellular-level alterations in brain tissue. Among millimeter-resolution medical imaging modalities, diffusion MRI (dMRI) stands out due to its unique sensitivity to tissue microgeometry, orders of magnitude below its nominal resolution. To uncover the signature of tissue microstructure in dMRI, we analytically identify key degrees of freedom from tissue geometry that survive massive averaging over millimeter-sized voxels. Identifying such
parameters requires accessing tissue microgeometry in its realistic state, instrumental in deriving and validating theoretical models. In this project, we plan to use X-ray holographic nanotomography for 3D imaging of millimeter-sized white matter tissue from rat brains at sub-100 nm resolution. We will study tissue morphology and organizational properties via machine learning-based image analysis and investigate their links to in-vivo and ex-vivo dMRI signals.