Assessment of artifacts reduction and denoising techniques in Electrocardiographic signals using Ensemble...

Area of Science:

• Biomedical Engineering
• Signal Processing
• Medical Monitoring

Background:

• Ambulatory vital signs monitoring is crucial for diagnosis and treatment.
• Noise and motion artifacts in Electrocardiographic (ECG) signals obscure vital information.
• Existing artifact reduction techniques lack comparative performance assessments and distortion measurements.

Purpose of the Study:

• To introduce a novel Ensemble Average method for comparing ECG denoising techniques.
• To evaluate the performance of various artifact reduction methods on ECG signals.
• To quantify both signal enhancement and waveform distortion introduced by denoising algorithms.

Main Methods:

• Utilized a dataset of synthetic and real-world noisy ECG signals with introduced motion artifacts.
• Applied several established ECG enhancement techniques to the dataset.
• Assessed algorithm performance using Signal-to-Noise Ratio (SNR) and a proposed Ensemble Average distortion measurement.

Main Results:

• All tested methods significantly improved the Signal-to-Noise Ratio (SNR).
• Wavelet-ICA demonstrated the best performance in reducing motion artifacts with the least waveform distortion.
• The Ensemble Average method quantified the percentage of signal distortion introduced by denoising techniques.

Conclusions:

• The Ensemble Average method provides a complementary metric for evaluating denoising techniques.
• Performance assessment should consider both SNR improvement and waveform distortion.
• The proposed method aids in selecting optimal ECG artifact reduction strategies.