An Adaptive ECG Noise Removal Process Based on Empirical Mode Decomposition (EMD)

Area of Science:

• Biomedical Engineering
• Signal Processing

Background:

• Electrocardiogram (ECG) signals are crucial for diagnosing cardiac disorders.
• ECG signals are susceptible to noise during acquisition and processing, hindering accurate interpretation.
• Existing noise reduction methods often compromise signal integrity.

Purpose of the Study:

• To propose an adaptive Empirical Mode Decomposition (EMD) based technique for effective ECG noise removal.
• To minimize signal distortion while significantly reducing noise in ECG recordings.
• To enhance the diagnostic capabilities of automated medical systems through cleaner ECG data.

Main Methods:

• An adaptive noise removal technique based on Empirical Mode Decomposition (EMD) is developed.
• The method processes ECG signals to eliminate noise, focusing on preserving signal characteristics.
• Performance is evaluated using standard metrics like Mean Square Error and Signal-to-Noise Ratio (SNR) changes.

Main Results:

• The proposed EMD-based method demonstrates significant noise reduction in ECG signals.
• The technique achieves low signal distortion, outperforming traditional EMD approaches.
• Evaluations on MIT-BIH and BUT QDB databases confirm the method's effectiveness across various SNR levels.

Conclusions:

• The adaptive EMD technique offers an effective solution for ECG noise removal.
• This method enhances signal quality, leading to improved diagnostic functions in automated systems.
• The approach is compatible with existing ECG noise reduction techniques.