Delay Differential Equation Models of Normal and Diseased Electrocardiograms

摘要: Time series analysis with nonlinear delay differential equations (DDEs) is a powerful tool since it reveals spectral as well properties of the underlying dynamical system. Here global DDE models are used to analyze electrocardiography recordings (ECGs) in order capture distinguishing features for different heart conditions such normal beat, congestive failure, and atrial fibrillation. To data types number terms delays model nonlinearity have be selected. The structure selection done supervised way by selecting that best separates types. We analyzed 24 h from 15 young healthy subjects sinus rhythm (NSR) failure (CHF) patients suffering fibrillation (AF) selected Physionet database. For presented here we 5 min non-overlapping windows on raw without any artifact removal. classification performance Cohen Kappa coefficient computed directly confusion matrix. overall three groups was around 72–99 % approaches. 2 all above \(95\,\%\).