Research Statement
Up until now, my research at Western has been focusing on resting state activity as measured by functional magnetic resonance imaging (fMRI) in patients with disorders of consciousness (DOC): coma, vegetative state, minimally conscious state and locked in syndrome. It has been clear from the beginning that studying the resting state blood oxygen level dependent (BOLD) signal will have important clinical implications in DOC patients. Extracting the spatial patterns of correlated, functionally connected regions from the resting state activity of severely damaged brains in DOC patients is quite challenging, even though different tools have been developed in recent years to analyze resting state activity of normal brains. Most of my research over the past ten years has been devoted to developing new tools, more suitable for severely damaged brains. Such tools allow better detection of severely disrupted spatial patterns and enable us to obtain information about plausible neuronal origins for the residual networks of connected regions, excluding other physiological sources, such as heart beat and breathing. Important results have been published in a number of high-ranking journals. In these studies I have had a leading role in developing or performing the analysis. One example is the work recently published in Lancet Neurology, describing neural correlates of consciousness in patients who have emerged from a minimally conscious state, where I am listed as co-last author. Another important development has been the publication of a completely new imaging technique to map functional connectivity relationships, as measured by resting state fMRI. One of my MSC students, first author of the paper, is now the CEO of a company, BraiNet – Brain Imaging Solutions INC., of which I am Co-Director and which seeks to develop diagnostic and neuronavigation tools from fMRI. GraphICA, the main software developed by BraiNet, is now being used by six different institutions all over the world for clinical research applications.
Although the main driving force of my research has been extracting biomarkers from fMRI spontaneous activity to diagnose the different states of DOC, recently I have dedicated more attention to understanding the prognostic value of these markers. In 2015, my collaborators and I received funding for an Italian project that sets out to study the clinical, structural, and functional markers of recovery of consciousness. The study, in collaboration with the Neurorehabilitation Center of Telese and the SDN of Naples, Italy, will involve a longitudinal study in DOC patients using simultaneously acquired PET and MRI. FDG-PET (to measure metabolic activity) and different MRI sequences are being acquired in a one-hour protocol.
I am applying the same protocol, in a similar scanner at the St Joseph Hospital here in London Ontario, for a project funded by my Discovery NSERC grant, which was awarded in 2014. In that case, the plan is to study the structure-function relationship of the human brain in healthy subjects and data from 25 participants were acquired in 2017. In recent years, some investigators have begun to investigate the possibility of inferring functional from structural connectivity by drawing structural connectivity matrices, using diffusion spectrum imaging tractography and extracting the couplings to be used in computational models of the large-scale dynamics of the cerebral cortex. From such dynamics, BOLD signals can be simulated and functional connectivity maps can be calculated, to be quantitatively compared with empirical fMRI observations. Since my appointment at Western, my lab has been focusing on a generalized version of the Ising model, implemented on the human connectome. A paper investigating the structure-function relationship in the human brain implementing the generalized Ising model has been recently accepted and one more manuscript is ready for submission. Three of my MSC students completed their masters on different applications of the generalized Ising model and each of my present three PhD students is involved in at least one project using the same model, investigating spontaneous activity, stimulated activity and/or causality. Implementations of metabolic measurements in the generalized Ising model are also under investigation in my lab. Future work will focus on characterizing the changes in brain structure and residual cerebral function in survivors of severe brain injuries with DOC. We plan to test whether the developed generalized Ising model is able to capture the main functional features of a severely injured brain, once the disrupted structural connectivity maps are used as an input for the model.
– Andrea Soddu