Enhanced cardiac arrhythmia classification through integration of ensemble empirical mode decomposition and heart rate variability analysis

Cardiac arrhythmias are critical conditions requiring accurate classification for effective diagnosis as well as treatment. In this investigation, we provide a novel approach for cardiac arrhythmia classification that integrates two advanced techniques for feature extraction from ECG signals: “Ensemble Empirical Mode Decomposition” (EEMD) and “Heart Rate Variability” (HRV) analysis. The proposed approach employs EEMD to decompose ECG signals into intrinsic mode functions, capturing signal features, while HRV analysis provides additional physiological insights into heart rate fluctuations. Combining two strategies, our approach leverages a comprehensive set of features to improve the accuracy and resilience of arrhythmia classification. The system's effectiveness is explained via simulated tests utilizing the MIT-BIH arrhythmia database, with performance evaluated based on recall, accuracy, and precision metrics. Our results indicate that integrating EEMD and HRV features provides a more reliable and detailed classification of cardiac arrhythmias, offering a holistic perspective on heart rhythm dynamics.