Analysis of ECG-based arrhythmia detection system using machine learning.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Machine Learning

Background:

• Electrocardiogram (ECG) signal analysis is crucial for diagnosing cardiac arrhythmias.
• Traditional methods often face challenges with noise and complex signal patterns.
• Developing robust automated classification systems is essential for clinical applications.

Purpose of the Study:

• To develop and evaluate a novel approach for ECG signal analysis and arrhythmia classification.
• To leverage the strengths of 3D Discrete Wavelet Transform (DWT) and Support Vector Machine (SVM) for enhanced performance.
• To accurately categorize diverse types of heartbeats using a comprehensive dataset.

Main Methods:

• ECG signal preprocessing using 3D DWT for de-noising and feature extraction.
• Feature extraction through wavelet coefficients derived from the 3D DWT.
• Classification of ECG signals into nine distinct heartbeat types using Support Vector Machine (SVM).

Main Results:

• The optimal classification performance was achieved using level 4 approximations with the Symlet-8 (Sym8) wavelet.
• The SVM classifier demonstrated a high average accuracy of 99.02% on the CPSC 2018 arrhythmia dataset.
• The proposed method significantly outperformed Complex Support Vector Machine (CSVM) and Weighted Support Vector Machine (WSVM) in accuracy.

Conclusions:

• The integrated 3D DWT and SVM approach provides a highly accurate and effective method for ECG signal classification.
• This technique offers superior performance compared to existing methods for diagnosing various arrhythmia diseases.
• The study highlights the potential of advanced signal processing and machine learning for improving cardiac diagnostics.