ECG Noise Cancellation Based on Grey Spectral Noise Estimation

Area of Science:

• Biomedical Engineering
• Signal Processing
• Wearable Health Technology

Background:

• Wearable devices are increasingly used in healthcare, often measuring electrocardiogram (ECG) signals.
• ECG signals acquired during body motion are susceptible to noise, including power line noise (PLn) and electromyogram (EMG).
• Effective noise cancellation is crucial for accurate ECG interpretation in wearable health monitoring.

Purpose of the Study:

• To develop and evaluate a novel grey spectral noise cancellation (GSNC) scheme for removing PLn and EMG noise from ECG signals.
• To enhance the accuracy and reliability of ECG data acquired from wearable devices.
• To compare the performance of the proposed GSNC scheme against existing noise reduction methods.

Main Methods:

• The proposed GSNC scheme utilizes a two-stage discrimination process.
• Empirical Mode Decomposition (EMD) is used in the first stage to decompose the ECG signal into intrinsic mode functions (IMFs).
• Grey Spectral Noise Estimation (GSNE) identifies noisy IMFs, which are then further processed and discriminated using Ensemble Empirical Mode Decomposition (EEMD) in the second stage.

Main Results:

• The GSNC scheme effectively identified and removed PLn and EMG noise from ECG signals.
• Performance evaluation on 43 datasets from the MIT-BIH cardiac arrhythmia database demonstrated the scheme's efficacy.
• The proposed GSNC method showed superior performance compared to traditional EMD and EEMD-based noise cancellation techniques.

Conclusions:

• The developed GSNC scheme provides an effective solution for noise reduction in wearable ECG signals.
• This method significantly improves the quality of ECG data, aiding in more accurate cardiac arrhythmia detection.
• The proposed approach offers a promising advancement for noise cancellation in biomedical signal processing.