In this seminar, Dr. Silvana Silva will present her research during the past years within the field of neuroscience, which has focused on data modeling and analysis using brain images obtained with EEG, MEG, and fMRI. First, a network-theoretical analysis of MEG, where sensors are identified with nodes and the interaction between channel time-series with the network’s connections. This is a challenging approach, as EEG/MEG time-series are mixtures of source activity, leading to interpretation problems. Her research addressed the question of to what extent the network topology can be reconstructed from sensor-level FC measures, performing functional network analysis in simulated data, where we used a diffusion MRI-constrained whole-brain computational model. Second, a mathematical formalism derived to characterize the attractor structure of the time-varying energy landscape, i.e. the stationary points and their connections, depicted by brain dynamics. This formalism allowed us to distinguish quantitatively between the different human brain states of wakefulness and different stages of sleep in fMRI data. Finally, the analysis of EEG data from a language-related experiment was designed to analyze the impact of native language on neural entrainment. Data from sensor space was source reconstructed and evaluated using modularity and complexity measures, revealing an underlying functional parcellation in the primary auditory cortex, and an ordered pattern of the data that parallels the hierarchic characteristics present in the auditory stimuli.
Silvana Silva studied computer science at the University of Oslo, Norway, where she obtained her Master of Science degree in mathematical modeling (2003). She joined the Signal Processing for Communications group of the Universitat Politècnica de Catalunya and obtained her PhD in 2012. After her first postdoc position at UPC, in 2015 she joined the Computational Neuroscience Group of Prof. Gustavo Deco at Universitat Pompeu Fabra, to work on the development of new whole-brain dynamical models. In June 2018 she joined the Speech Acquisition and Perception Group led by Prof. Nuria Sebastian-Galles, to work with EEG data obtained from experiments performed in the lab, and the development of models for classification using machine-learning techniques. Additionally, from July 2015 to May 2018, Silvana held teaching duties at the Department of Information and Communication Technologies of the UPF, and in 2017 she was awarded a four-month postdoctoral research grant to visit the Center for Theoretical Neuroscience at Columbia University, New York. She has worked as a collaborator for the CCCB project ALIA: la consciència en vinyetes. Since 2019 she has contributed with teaching for the Màster Interuniversitari of Neurociències of Universitat de Barcelona.