Detecting Myocardial Infraction Using VCG Leads

Standard electrocardiogram (ECG) lead system and Frank vectorcardiogram (VCG) lead system are two most popular and basic lead systems. Most of the existing analyses on ECG signals are based on the standard ECG leads. In practice, Frank VCG leads that are orthogonal between them are more correlated with anatomy than standard ECG leads. The VCG feature extraction from Frank leads has been studied in this research. Myocardial infraction (MI) VCG signals including health control (HC), acute MI (AMI), sub-acute MI (SAMI) that were taken from PTB diagnostic ECG database were employed for the analysis in this study. The multivariate autoregressive (AR) coefficients were utilized as VCG features for the classification. The results show that Frank VCG leads can classify better compared to standard ECG leads.