Multi-modal Investigation of Cortical Connectivity at Multiple Scales

In neuroimaging, a great interest is currently being directed to diffusion magnetic resonance imaging (dMRI) which, in addition to functional magnetic resonance imaging (fMRI), magnetoencephalography (MEG), electroencephalography (EEG), functional near-infrared-spectroscopy (fNIRS) provides a large spectrum of measurements to enlighten the brain structure and function. The success of dMRI is deeply rooted in the powerful concept that during their random, diffusion-driven displacements molecules probe tissue structure at a microscopic scale well beyond the usual image resolution. Diffusion imaging opens several perspectives for what concerns the development of new non invasive techniques not only to optimize the diagnosis and therapy planning for oncological patients but also to discover the anatomical structure of the human cortex.Though, many issues still remains to be solved. Among the most striking are the reconstruction of the ODF (orientation distribution function) in noisy conditions, its reproducibility over time points acquisitions, the intra and inter-subject registration and the integration of functional information about the cortical activity within the reconstruction of the fiber network from raw data. This is of paramount importance as it would allow to link the functional information to the structural anatomical substrate. This thesis aims at investigating a subset of such issues in order to trace the path to the overall solution. In particular, it aims at integrating multiscale space-scale processing, diffusion imaging and cortical signals to (i) improve the orientation diffusion function (ODF) reconstruction, reproducibility and robustness to noise; (ii) contribute new methods for the registration of intra and inter-modality multidimensional data (tensors, probability distributions); (iii) explore the possibility of integrating functional signals in the processing pipeline in order to guide the fiber reconstruction and as a potential mean of validation of the proposed methods.From the clinical point of view, the goal of this thesis is to make tractography exploitable in daily practice for surgical planning and follow-up, assessment of degenerative pathologies as well as of pharmacological treatments.