Evaluating Mathematical Models for Morphological Classification of the QRS Complex

Abstract

The present work aims to present a comparative study of the performance of different kernels for mathematical modeling and morphological classification of the QRS complex of the ECG signal. Initially, we use a simulator to generate synthetic signals from dynamic models containing variations within a set of physiological parameters. From the generation of twenty different types of QRS morphology, computing tests for mathematical modeling of the beat waveform (Q, R and S waves) were performed. For that, the following mathematical functions are employed: Gaussian function, Mexican Hat function and Rayleigh probability density function. Subsequently, 10 real signal records from the MIT-BIH (Massachucetts Institute Technology - Beth Israel Hospital) Arrhythmia database have their QRS complex morphologies also modeled by the proposed mathematical functions. The preliminary results demonstrate the proposed mathematical functions with adjustable parameters can be applied together for modeling and automatic classification of some QRS morphologies commonly present in real signals, with efficiency and precision. The computing of normalized RMS error allows the identification of the model which is more appropriate to a given morphology, which can change over the same patient record.