Title: Analyzing functional data over complex multi-dimensional domains
Authors: Laura Sangalli - Politecnico di Milano (Italy) [presenting]
Eleonora Arnone - Politecnico di Milano (Italy)
Luca Negri - Politecnico di Milano (Italy)
Abstract: A novel class of models is presented for the analysis of functional data defined over complex multidimensional domains, including curved bi-dimensional domains and complex three-dimensional domains. This class of models includes smoothing methods, regression methods and principal component analysis methods. These are implemented using numerical techniques such as finite elements and they are based on the idea of differential regularizations. We will illustrate the methods via an application to the study of neuroimaging data. In this applicative domain, the proposed methods offer important advantages with respect to the best state-of-the-art techniques, allowing to correctly take into account to complex anatomy of the brain.