[Forum SIS] DEC Seminar - March 20th
Raffaella Piccarreta
raffaella.piccarreta a unibocconi.it
Ven 14 Mar 2014 10:30:59 CET
Dear all,
we are glad to announce the following DEC seminar
Thursday, March 20th
Università Bocconi,
Room 3-E4-SR03
Via Rontgen 1 - 3rd floor
Time: 12:30pm
Raquel Prado
(University of California Santa Cruz, USA)
Bayesian modeling approaches for inferring activation and connectivity from brain signals
Abstract
We consider Bayesian state-space models for the analysis of large-dimensional brain signals including
functional magnetic resonance imaging (fMRI) signals and multi-channel electroencephalogram
(EEG) data. In the case of fMRI signals, we develop a modeling approach that allows us to
jointly infer local hemodynamic response functions and activation parameters, as well as global
effective and functional connectivity parameters across multiple regions of interest (ROIs) in the
brain. Sparsity priors are placed on the parameters that describe coupling relationships between
ROIs. We also consider dynamic factor models with structured latent factors for the analysis of
multi-channel EEG and fMRI data. We discuss several aspects of the proposed models, as well as
posterior inference through Markov chain Monte Carlo and sequential Monte Carlo methods. We
apply the statistical models and methods to the analysis of two real fMRI datasets and one EEG
dataset.
******************************************************************
The DEC seminar schedule is available at http://www.unibocconi.eu/statseminar
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20140314/4b69d4a4/attachment-0001.html>
-------------- parte successiva --------------
Un allegato non testuale è stato rimosso....
Nome: R. Prado 20.03.14.pdf
Tipo: application/pdf
Dimensione: 42937 bytes
Descrizione: non disponibile
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20140314/4b69d4a4/attachment-0001.pdf>
Maggiori informazioni sulla lista
Sis