Quantitative Magnetic Resonance Imagingadmin2015-07-25T09:27:38+00:00
Example of map of arterial delay obtained from resting state functional magnetic images through cross-correlation analysis with the superior sagittal sinus signal in a healthy subject.
Functional and effective connectivity
The main project of this research line is the investigation of the functional and effective connectivity of resting state networks in normal subjects and multiple sclerosis patients. The relatively simple experimental setup of the resting state fMRI makes it ideally suitable for large-scale investigation of how multiple sclerosis affects functional brain connectivity and how the connectivity impairment correlates with clinical disability scales. More specifically, we aim to investigate both functional and effective connectivity with motor disability.
Assessment of cerebral hemodynamic impairment
Alterations in cerebral hemodynamic have been reported in multiple sclerosis patients both in white matter (WM) and grey matter (GM). In particular cerebral arterial arrival time seems to be prolonged in normal appearing WM and deep GM, and associated with disability. As suggested in previous works resting state functional magnetic resonance imaging can be used to map the vascular delay since blood oxygenation level dependent fluctuation largely depends, even not only, on changes in cerebral blood flow. The aim of this study is to investigate the cerebral hemodynamic impairment in multiple sclerosis patients by using noncontrast mapping of arterial delay in rs-fMRI.
Combination of information from EEG and fMRI allows one to overcome the low spatial resolution of the former and the low temporal resolution of the latter. Even though both techniques are widely used, their integration is not equally widespread. This is mainly due to several technical matters that arise when combining the two methods as well as to an incomplete understanding of the relationship between the hemodynamic response measured by BOLD-fMRI and the underlying neuronal activity detected by EEG. The main aim of this research line is to investigate how changes in EEG frequency bands affect functional and effective connectivity of resting state networks (RSNs) identified with fMRI.
Example of quantitative high resolution (2x2x2 mm) ASL at 3T
Maximum Intensity Projection of in vivo ASL angiography
Quantitative magnetic resonance perfusion research area is mainly divided in two projects
Development of novel methods for quantification of perfusion for Dynamic Susceptibility Contrast MRI
Dynamic Susceptibility Contrast MRI permits the in vivo quantification of brain perfusion using an exogenous contrast agent. We aims to develop new deconvolution approaches to overcome the limitations of existing methods. In this field we developed Nonlinear Stochastic Regularization and we are now working on the application of Stable Spline to DSC-MRI. We have also developed a method for automatic extraction of the Arterial Input function directly from DSC-MRI images.
Development of novel methods for quantification of perfusion for Arterial Spin Labelling
Arterial Spin Labelling permits the in vivo absolute quantification of brain perfusion without needing any contrast agent injection. In this field we have developed a novel deconvolution algorithm, based on Stable Spline kernel, to quantify perfusion and arterial arrival time from QUASAR ASL data. ASL pre-processing is a fundamental step in the analysis of this kind of data, therefore we are developing novel methods for motion correction and outlier detection in order to enhance the robustness of ASL. Current developing involves novel encoding scheme combined with optimized readout for both angiography or perfusion quantification.
Processing pipeline of the two-shell diffusion tensor images analysed with NODDI
Models for improved analysis of multi-shell diffusion MRI data
Multi-shell diffusion imaging allows the study of the microstructural complexity of dendrites and axons in vivo with a higher specificity than that provided by the analysis of conventional diffusion tensor images. We are working on the development of a model-based method for the analysis of multi-shell acquired signals able to return a reliable description of the exchange of water between the intra-cellular and extra-cellular compartments. This method is expected to allow the quantification of specific markers of brain tissue microstructure, not obtainable by the conventional analysis of the one-shell diffusion tensor images. In addition, this method is expected to improve the reconstruction of fiber bundles by tractography.
Evaluation of Brain Damage in Multiple Sclerosis
We are performing analyses of one-shell and two-shell diffusion tensor images, acquired with 1.5T and/or 3T MR systems, to investigate the microstructural alterations of white and grey matter in multiple sclerosis. More specifically, we are interested in the use of diffusion MR imaging to detect abnormalities in the cortical or subcortical normal appearing gray matter of patients with relapsing-remitting multiple sclerosis (RRMS) patients compared with controls and patients with RRMS and epilepsy.
Comparison of methods to quantify susceptibility in vivo with magnetic resonance imaging
Development of novel methods for quantification of magnetic susceptibility
Quantitative susceptibility mapping is a novel technique that aims to absolutely quantify the magnetic susceptibility in vivo and noninvasively using magnetic resonance. This research area involves the development of novel strategies to solve the issues related to the quantification of susceptibility. Moreover, we are working on Susceptibility Tensor Imaging, the natural extension of QSM, that relaxes the hypothesis of isotropy of the susceptibility opening the possibility to better describe tissue’s microstructure. Further development will let to detect in vivo changing of iron content in pathologies, such as Parkinson disease or Multiple Sclerosis.
The relevance of cortical lesions has gained increasing attention, despite the higher difficulty in detecting them with conventional imaging techniques and their heterogeneity. Their importance is due to the correlation between their number and cognitive impairment in pathologies such as multiple sclerosis and their association with worse outcome. Up to now, cortical lesions identification and lesion load assessment are based on visual detection, a process that is time consuming and operator dependent.
The aim of the research is to develop a technique to automatically and effectively detect cortical lesions on either on single sequence (double inversion recovery, DIR) data, or from multispectral data.
3D rendering of ventricular surfaces automatically extracted from cardiac MRI data
Right and left cardiac volume
The quantitative estimation of clinical parameters from cardiac-MRI data in order to correctly assess the cardiac performance, and thus promptly identify eventual diseases, strongly relies on an accurate delineation of the ventricular contours. Nonetheless, this is a time-draining and operator-dependent task for the clinical routines, due to the massive amount of data produced. Many recent research efforts have focused on providing time-effective and reliable automated techniques, and even though some semi-automatic solutions are available in commercial products, this specific problem is still open.
The aim of the research is to develop methods and tools for the identification of left and right ventricular contours on cardiac MR images.