QRS amplitude and shape variability in magnetocardiograms.

Area of Science:

• Biophysics
• Biomedical Engineering
• Cardiology

Background:

• Magnetocardiography (MCG) commonly uses QRS complex averaging to enhance signal-to-noise ratio.
• Standard averaging methods can suppress signal variations due to physiological factors like respiration, leading to inaccuracies.
• These inaccuracies can compromise subsequent analyses, such as inverse source reconstructions.

Purpose of the Study:

• To develop a method for separating and selectively averaging QRS complexes with varying shapes and amplitudes.
• To overcome limitations of standard averaging techniques in magnetocardiography.
• To improve the accuracy of signal analysis and source localization in MCG.

Main Methods:

• A novel method employing spline interpolation of a standard averaged QRS complex.
• Fitting the spline function to individual QRS complexes using nonlinear regression (Levenberg-Marquardt algorithm).
• Utilizing five regression parameters: amplitude scaling, baseline drift (2 parameters), time scaling, and time shift.

Main Results:

• Respiration significantly influences both the amplitude and shape of the QRS complex.
• Respiration has a less pronounced effect on the baseline signal.
• Regression parameters from neighboring measurement channels exhibit linear correlation, enabling simultaneous selective averaging across multiple sensors.

Conclusions:

• The developed method effectively separates and selectively averages QRS complexes, accounting for respiration-induced variations.
• This approach enhances signal fidelity in magnetocardiography, reducing errors in signal portions and inverse source reconstructions.
• The correlated nature of regression parameters facilitates efficient, multi-channel selective averaging, advancing MCG data analysis.