Recognition of Congestive Heart Failure Based on a Complex Correlation Measure of the Heart Rate Signal

The article examines the indicator of a complex correlation measure applied to the task of recognizing congestive heart failure by a heart rate signal. It is shown that the features of the Poincaré plot for the sequence of cardiointervals can be evaluated both by statistical descriptors and by special functions reflecting the temporal organization of the analyzed time series. Statistical parameters describe only the spatial distribution of points in the scattering region. The multiple delay correlation function, which includes the complex correlation measure, allows identifying the features of the dynamics of the heart rhythm in pathology, including the presence of nonlinear components. It is the analysis of the structure of the time series that can provide additional information, thereby providing a more complete description of the differences between the classes of norm and pathology. A comparative theoretical and experimental analysis of the studied descriptors was carried out. The experiments were carried out on samples of experimental data of the classes of normal heart rate and heart rate obtained from the databases of the PhysioNet portal. Statistical indicators of the analyzed functions are investigated, a statistical T-test of the Student for the presence of significant differences in these classes is carried out. A method is proposed for recognizing the congestive heart failure in a multidimensional space of consecutive complex correlation measure values obtained for different time delays. It is shown that for such a description a reliable classification of pathology can be achieved against the background of a normal heart rhythm.