Noise cancellation signal processing method and computer system for improved real-time electrocardiogram artifact...

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Signal Processing

Background:

• Magnetic Resonance (MR) scanning generates artifacts in electrocardiogram (ECG) signals.
• Accurate ECG monitoring is crucial for patient safety and triggering MR data acquisition.
• Existing methods struggle with real-time artifact correction in the MR environment.

Purpose of the Study:

• To develop a real-time system for ECG analysis and artifact correction during MR scanning.
• To improve patient monitoring and the accuracy of MR data acquisition triggering.
• To enable reliable electrophysiological signal monitoring within the MR environment.

Main Methods:

• A noise cancellation (NC) method based on linear time-invariant process assumption was applied.
• Digital finite impulse response (FIR) matrix computed using training scan data (ECG and gradient waveforms).
• Real-time artifact prediction via convolution of gradient waveforms and subtraction from raw ECG signal.

Main Results:

• The system demonstrated real-time operation at 1 kHz with a minimal processing delay of 1 ms.
• Validation confirmed effectiveness both off-line and during actual MR examinations.
• Successful artifact correction was achieved, producing a clean ECG signal.

Conclusions:

• The developed system effectively corrects ECG artifacts in real-time during MR scanning.
• It enhances patient monitoring and MR data acquisition triggering capabilities.
• The system paves the way for automatic electrophysiological signal monitoring in MR settings.