Influence of Myographic Interference and Isoelectric Line Drift on Correlation Coefficient in Classification of Cardiocomplexes

摘要

Distribution and mean value of a sample correlation coefficient were found. Theoretical results are compared to computer simulation results. The simulation results are consistent with the theoretical predictions. Cardiovascular disease is now the primary cause of death worldwide, so it is an important task to accurately and timely diagnose it. Currently, a popular diagnosis method is Holter monitoring - continuous long-term (up to several days) recording of the electrocardiogram (ECG). For a doctor to analyze such a long record and make correct diagnosis within an acceptable time, cardiocomplexes of each type are sorted into relevant groups (norm/pathology), and only representative cardiocomplexes from each group are analyzed [1, 2]. As a measure of similarity of cardiocomplexes, the correlation coefficient can be used. If the value of the correlation coefficient between the reference and analyzed cardiocomplexes exceeds a certain threshold, the studied cardiocomplex is placed in the same group with the reference cardiocomplex.