Heartbeat classification using morphological and dynamic features of ECG signals.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Accurate heartbeat classification is crucial for diagnosing cardiac arrhythmias.
• Existing methods often rely on single feature types, potentially limiting classification performance.

Purpose of the Study:

• To develop and validate a novel, robust heartbeat classification system.
• To improve diagnostic accuracy by combining morphological and dynamic electrocardiogram (ECG) features.

Main Methods:

• Extraction of morphological features using Wavelet Transform and Independent Component Analysis (ICA).
• Computation of dynamic features from RR intervals.
• Classification using a Support Vector Machine (SVM) classifier on concatenated features.
• Fusion of decisions from two ECG leads for enhanced classification.

Main Results:

• Achieved 99.3% overall accuracy in class-oriented evaluation (99.7% with rejection).
• Obtained 86.4% accuracy in subject-oriented evaluation.
• Results are comparable to current state-of-the-art in automatic heartbeat classification.

Conclusions:

• The proposed combined feature approach significantly enhances heartbeat classification accuracy.
• The method demonstrates effectiveness and robustness across different evaluation metrics.
• This technique offers a promising advancement for automated cardiac arrhythmia detection.