Neurodegenerative diseases represent one of the most prevalent health problems worldwide. Their complex nature requires multidisciplinary investigation approaches. In this field, combining X-ray advanced micro-imaging techniques, functional Magnetic Resonance Imaging (fMRI) and new analysis algorithms could become a significant tool for pre-clinical research and potentially for clinical routine.
High-resolution imaging techniques can allow a direct quantitative estimation of important morphological and topological parameters characterizing the vascular and neuronal networks in the spinal cord, contributing to the comprehension of the BOLD contrast in this area of the central nervous system. In particular, accurate modelling of the biophysical origin of the BOLD contrast requires a degree of spatial information that X-ray phase-contrast tomography can easily reach.  
Potentially, the knowledge acquired on small rodents can be exploited to better understand the physiology of human spinal cord and the relevant pathologies. Similar approaches can be instrumental for the application in the clinical practice of non-invasive techniques, like fMRI, that have the potential of contributing to the development of new treatments for serious pathologies.
This projects is based on the long-standing collaboration with TOMALab of the CNR-Nanotec (Rome) that has considerable expertise on the X ray phase contrast tomography imaging technique and on the pertinent imaging algorithms.


PAMINA is a web-based hardware and software platform specifically conceived to gather, manage, retrieve, show and process data and metadata in a modern neuroimaging laboratory. The system is targeted for the field of clinical research on human brain functions (from acquisition to final processing), but it can be used as well in the clinical and diagnostic field. PAMINA is an operative and convincing example of scientific and technological knowledge transfer between clinical/fundamental research and firms. PAMINA enables high-level data processing and allows data visualization depending on user qualification and professional objectives, having as only goal the patient’s health preservation.
It has been designed in a highly proven operative context and it presents a lot of interesting technological innovation aspects. Among these, we must cite the multimodal analysis environment integrated with an intelligent repository that allows the storage of all information at any processing step, from raw data to final results.


  • can communicate with DICOM format (MRI and TC scanners);
  • allows the retrieval of raw data as soon as they are acquired, classifying them depending on their typology;
  • allows the retrieval of data in the Nifti exchange format, useful as input to the principal analysis software tools;
  • allows a high-performing and robust data processing, using a grid calculation approach;
  • guarantees a simple and safe user access, thanks to the web based interface which relies on standard authentication methods.

PAMINA implements the following software/tools:
AFNI, AIR, BrainSuite, DIRAC Diffusion Toolkit, fsl, FreeSurfer, GAMMA, HCP, ITK, LONI Pipeline and Workflows, MINC, MiND, SPM8-12, SVPASEG, and SHAPETools.

Functional Magnetic Resonance Imaging (fMRI) is the tool of election in the investigation of brain function in both physiological and pathological conditions. In addition to the investigation of task-evoked activity in several conditions, fMRI technique is largely used in the mapping of intrinsic Resting State (RS) BOLD fluctuations occurring in absence of task. The functional connectivity can be defined as the temporal correlation among two or more anatomically spatially remote regions and it has recently catalyzed attention in the exploration of human brain function. This technique could be implemented by several methods based on the calculation of the correlation between signal time course of one brain region and the time course of the rest, or part, of the brain. The basic assumption is that the temporal similarity between the signals in each region strongly suggests that they are in constant communication with one another constituting a functional network. Consistency of RS networks in healthy human adults is well established and changes of networks in pathological conditions is profusely studied. It has been demonstrated the existence of networks whose connection may be altered by aging or disease. The popularity of this technique belongs to different reasons. Among neuroimaging techniques, fMRI is the ideal in the investigation of functional connectivity since it has a spatial resolution superior to other methods, which is necessary to discriminate different networks. Moreover, the relative ease of data acquisition (patient are not required to perform a task) makes it a suitable approach also in a clinical environment, where, often, the most disabling diseases affect the execution of easier tasks. All these reasons, and many others of more technical aspect, made functional connectivity one of the most popular area of research in neuroscience.


Functional magnetic resonance imaging (fMRI) of the blood oxygen level dependent (BOLD) response has commonly been used to investigate the neuropathology by examining the positive phase of the BOLD response, assuming a fixed shape for the hemodynamic response function (HRF). However, the individual phases of the HRF may be characterized by different trend as a function of the brain areas activated. In this project, we want realise a statistical brain map of these characteristics. The results will bring on a better interpretation of hemodynamic response estimates across a wide variety of psychological and neuroscientific studies, because better it will the reconstruction of the BOLD if will use a HRF. We plan to leverage on the HCP dataset (freely available) to investigate the spatial variability of HRF corresponding to the various available stimulations.

The goal of the project is obtain a statistical brain map of the hemodynamic response functions as a function of the brain areas active. We focus on estimation of response amplitude/height, time-to-peak, and full-width at half-max in the HRF as potential measures of response magnitude. Actually, the choice of the HRF range fails from the use of a single canonical HRF obtained with a basis set of canonical function, well away from a real behaviour.


Despite its recent discovery, fMRI is the most popular neuroimaging technique for mapping brain activation. The human brain is a heterogeneous and complex organ composed by functional components, at different spatial scales, finer than a millimeter, which works at different temporal scale. The goal of an imaging technique is to be able to fully resolve these complexities and this ability also depends on its spatial and temporal resolution. In order to improve efficiency, variations of the conventional fMRI techniques were developed.
Among advanced fMRI methods, Multi Band (MB) MRI and Multi Echo (ME) MRI, as well as their combination (MB-ME MRI), hold a paramount role, especially due to the advantage of being usable on the wide platform of installed clinical scanners.


Multi Band

fMRI experiments last several minutes in some cases, when several data have to be acquired. These acquisition sessions are uncomfortable for patients and are sensitive to several artifacts especially those due to motion.
In order to reduce examination time and to increase temporal and spatial resolution, an important goal in in vivo MRI is to optimize the speed of image acquisition. MB imaging provides an attractive and alternative solution to these challenges.
Differently from other approaches, such as the use of ultrahigh magnetic field, (above 7 T) MB sequences have the advantage of being usable on the wide platform of installed clinical scanners
The MB technique is based on the simultaneous excitation of multiple brain slices using a single NMR RF (Radio Frequency) pulse, tailored at multiple frequencies. Each receiving coil measures a signal that is a linear combination of signals from the excited slices modulated by the coil sensitivity. The real innovation of MB technique, compared with its ancestors, is that it allows to reduce the value of an important acquisition parameter, the TR (Repetition Time, the period between two consecutive RF pulses), with the resulting several improvement that this reduction involves. Today, MB approach has the potential to become an essential data acquisition strategy in the study of both functional and structural feature of human brain.


Multi Echo

Standard fMRI data are usually acquired in a single optimal TE (Echo Time, the time after the RF pulse at which the response signal is measured) in order to achieve a maximum functional contrast. This optimal value is set equal to the average T2* parameter (the transverse relaxation time, the time required for the response signal from a given tissue to decay) across brain tissue, at a certain field strength, in order to maximize the response of interest. Nevertheless, even if the optimal TE was chosen, thus a value which approximates the average T2*, this parameter varies considerably across brain so that the sensitivity of MRI technique with a fixed TE cannot be optimal across the whole brain, but could elicit great variability in the contrast of functional images. Moreover, functional MRI data are characterized by fluctuations of different origins. In addition to fluctuation of interest, which are the ones of neuronal origin and are TE-dependent, the MRI signal contains nuisance signal fluctuations. These latter are due to subject motion, respiration, cardiac function, hardware instabilities etc. and they are often TE-independent. Taking advantage of this, the combination of data from ME MRI sequence enable the separation of confounds from the signal of interest
Definitively, the combination of data from ME MRI sequences improves sensitivity of fMRI technique and it enables to achieve the maximum feasible functional contrast at a certain field strength in the whole brain.
Moreover, ME MRI acquisition is implemented by a small modification of single echo acquisition, and it could have a large everyday-life impact, since the improvement on repeatability and quality of data is especially true for clinical environment where patients exhibit high amount of movements or in clinical single subject studies.