A model of mechanical dyssynchrony based on ECG features

Abstract

Cardiac dyssynchrony, characterized by uncoordinated electrical impulses in the heart, can impair its pumping function. Previous research has predominantly focused on either the mechanical or the electrical aspects of dyssynchrony. In this study, we propose an integrated approach to explain left ventricular mechanical dispersion, a measure of dyssynchrony, using electrocardiographic parameters in a linear regression model. A total of 42 echocardiographic studies were analyzed, and 20 physiological features were extracted, including age, sex, angles, areas, arguments, widths, and maximum variance. The Akaike method was used for feature selection, identifying seven relevant characteristics for the regression model. Statistical analyses, including scatter plots, residual analysis, and hypothesis testing, were performed to evaluate the model’s assumptions and performance. The regression model demonstrated an overall explanatory capacity of 62.2% (R 2 ). This sugests that electrical parameters could be utilized to explain a mechanical variable associated with cardiac dyssynchrony and encourages further investigation.