Modelling of Arrhythmics ECG Signals with McSharrys Model

The automatic analysis of cardiac arrhythmias is important for diagnosis of cardiac abnormalities. This work is focused on modelling of arrhythmias with McSharry´s model. The proposed method uses preprocessing of signals with linear approximation distance thresholding method and line segment clustering method for establishing of initial parameters of McSharry´s model. The ECG data is taken from standard MIT/BIH arrhythmia database. The modelling was tested on the whole MIT arrhythmia database on the lead MLII and all signals can be modelled with 10 Gaussians functions without significant distortion.