Wavelet Based Method for Congestive Heart Failure Recognition by Three Confirmation Functions

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Electrocardiogram (ECG) analysis is crucial for diagnosing cardiac conditions.
• Characterizing arrhythmias and congestive heart failure (CHF) requires robust signal processing techniques.
• Existing methods for ECG analysis may have limitations in accuracy and noise resilience.

Purpose of the Study:

• To propose a new wavelet-based feature extraction and classification method for ECG signals.
• To effectively discriminate between normal and abnormal ECG patterns, specifically identifying CHF.
• To evaluate the performance of the proposed method against existing techniques and in noisy environments.

Main Methods:

• Utilized wavelet packet transform (WPT) for feature extraction, focusing on percentage energy (PE) of sub-signals.
• Introduced the wavelet-based average framing percentage energy (WAFE) technique.
• Developed a classification system using three confirmation functions: percentage root mean square difference error (PRD), logarithmic difference signal ratio (LDSR), and correlation coefficient (CC).
• Tested the system on the MIT-BIH arrhythmia dataset and other ECG databases.

Main Results:

• The proposed WAFE method demonstrated superior performance compared to several known methods.
• Achieved a high recognition accuracy of 92.60% for classifying arrhythmias and CHF.
• The Receiver Operating Characteristic (ROC) curve confirmed the reliability and diagnostic accuracy of the proposed system.
• In an additive white Gaussian noise (AWGN) environment, the system maintained a recognition rate of 81.48% at 5 dB.

Conclusions:

• The WAFE technique is a potentially effective tool for the characterization of arrhythmias and CHF from ECG signals.
• The proposed classification method offers reliable discrimination capabilities.
• The system exhibits robustness in the presence of noise, suggesting its clinical applicability.