Diffusion tensor imaging: study of white matter connection and advanced MRI sequence development

Magnetic resonance imaging (MRI) is an imaging technique able to produce high quality images of the human body without ionizing radiation. Over years, MRI potential has been continuously improved, and exploited for increasing the knowledge about many morphological and functional aspects of the human central nervous system. However, despite all recent MRI applications and improvements, there are still many unsolved questions about brain structure and physiology. This thesis was developed in the framework of a wider research activity carried out at the Brain Connectivity Center (I.R.C.C.S. Fondazione “C. Mondino”, Pavia), in collaboration with the Department of Electrical, Computer and Biomedical Engineering (University of Pavia, Pavia), aimed at casting further light on white matter (WM) connectivity and properties, with particular attention to cerebro-cerebellar pathways. The first project was focused on investigating WM differences between healthy subjects, patients affected by VaD and AD. Since VaD and AD neuropsychological profiles often overlap and anatomical changes do not always correlate with a specific type of dementia, differentiating between them results difficult. Therefore, in order to provide quantitative imaging biomarkers specifically sensitive to AD or VaD, a voxel wise analysis on the whole WM was performed on a total of 93 subjects (31 AD, 27 VaD and 35 healthy control). WM alteration patterns considered useful in distinguishing between VaD and AD emerged from the analysis. In particular, parahippocampal gyri and corpus callosum showed different anterior/posterior alteration patterns specific to pathology, while thalamic radiation impairment seemed to be characteristic of VaD, because these patients showed alterations in this area compared both to AD and HC but no alterations were found in AD. Altogether these findings revealed an impairment in WM bundles involved in cognition and in the cerebro-cerebellar connectivity. In the second project, cerebro-cerebellar connectivity was studied with the aim of clarifying cerebellar involvement in cognition and describing cerebro-cerebellar loop structure. Advanced MRI tractography was used to reconstruct and quantitatively describe the cortico-ponto-cerebellar (CPC) pathway in human in vivo. From the analysis emerged that the CPC pathway is mainly involved in the connection of cognitive cerebral and cerebellar areas. Furthermore, in order to study the whole cerebro-cerebellar loop, which is composed of the CPC and the cerebello-thalamo-cortical (CTC) pathway, these results were compared with the ones already obtained in a previous study focused on CTC. Both pathways resulted to be interested in connecting cognitive areas. A difference in terms of cerebral areas reached by the two pathways suggests an asymmetry of the cerebro-cerebellar loop regarding cognitive tasks. Whereas findings related to motor tasks suggest a symmetrical communication between the cerebrum and the cerebellum. Even though results obtained from post processing of diffusion data revealed important information, DTI suffers from resolution issues due to the trade off between voxel size and SNR. Therefore, diffusion analysis would benefit from higher resolution data. Here, in order to partially resolve the resolution problem, a sequence able to acquire high-resolution images using non-coplanar radio frequency pulses, was developed. Data acquired with this sequence showed good SNR values and an anatomically reliable reconstruction of color-coded maps of fractional anisotropy. All these projects gave interesting information about WM structure, along with new advanced investigation tools. Further experiments where all the techniques, here applied separately, can be applied together, would be useful in order to improve results quality and perform new hypothesis about WM structure and properties.