Outcome Prediction for Heart Failure Telemonitoring Via Generalized Linear Models with Functional Covariates


Stefano Baraldo, Politecnico di Milano, Dipartimento di Matematica ‘F. Brioschi’, piazza Leonardo Da Vinci 32, 20133 Milano, Italy.
E-mail: stefano1.baraldo@mail.polimi.it


An effective methodology for dealing with data extracted from clinical surveys on heart failure linked to the Public Health Database is proposed. A model for recurrent events is used for modelling the occurrence of hospital readmissions in time, thus deriving a suitable way to compute individual cumulative hazard functions. Estimated cumulative hazard trajectories are then treated as functional data, and they are used as covariates along with clinical survey data within the framework of generalized linear models with functional covariates.