Anisotropic median-diffusion for filtering noisy electrocardiogram signals

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Electrocardiogram (ECG) signals are vital for diagnosing heart conditions.
• Noise in ECG recordings can impede accurate diagnosis.
• Existing filtering methods may alter important signal characteristics.

Purpose of the Study:

• To introduce and evaluate anisotropic median-diffusion for filtering noisy ECG signals.
• To demonstrate a method for estimating the filter's parameters.
• To assess the filter's performance in preserving ECG signal integrity.

Main Methods:

• Application of anisotropic median-diffusion filtering to ECG signals.
• Parameter estimation for the proposed filter.
• Validation using MIT-BIH ECG databases, including signals with premature ventricular contractions.
• Comparison with other established filtering techniques.

Main Results:

• The anisotropic median-diffusion filter effectively reduces noise in ECG signals.
• The filtering process preserves the precise timing and amplitudes of cardiac events.
• Morphological features within the QRS complex are maintained.
• The method shows superior performance compared to other filtering approaches.

Conclusions:

• Anisotropic median-diffusion is a promising technique for noise reduction in ECG signals.
• This method offers an effective way to improve ECG signal quality for diagnostic purposes.
• The filter's ability to preserve signal morphology is critical for accurate heart disease diagnosis.