A new method for QRS detection in ECG signals using QRS-preserving filtering techniques

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiovascular Technology

Background:

• Accurate QRS complex detection is essential for electrocardiogram (ECG) analysis, including heart rate determination and classification.
• Noise in ECG signals, particularly noise spectrally overlapping with QRS complexes, significantly complicates automated detection.
• Existing methods often struggle with noise suppression while preserving vital QRS complex morphology.

Purpose of the Study:

• To develop and evaluate a novel preprocessing technique for robust QRS complex detection in noisy ECG signals.
• To introduce least-squares-optimisation-based smoothing and a nonlinear transformation to enhance QRS detection accuracy.
• To enable high-accuracy, low-computational-load QRS detection suitable for both real-time wearable devices and fast offline analysis.

Main Methods:

• Application of least-squares-optimisation-based smoothing to suppress noise while preserving QRS complexes.
• Utilisation of a novel nonlinear transformation to equalize QRS amplitudes without amplifying suppressed noise.
• Evaluation of both offline and real-time implementations using the standard MIT-BIH database.

Main Results:

• The proposed technique achieves high accuracy in R-peak detection after preprocessing.
• The offline implementation outperforms current state-of-the-art methods, including wavelet transforms and empirical mode decomposition.
• The real-time implementation demonstrates superior performance compared to existing real-time QRS detection techniques.

Conclusions:

• The proposed method offers an effective solution for accurate QRS detection in noisy ECG signals.
• Its low computational load makes it suitable for real-time applications in wearable devices like Holter monitors.
• The technique provides a significant advancement for both real-time and offline ECG analysis.