Averaging algorithm based on data statistics in magnetocardiography

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Magnetocardiogram (MCG) signals suffer from low signal-to-noise ratios, necessitating averaging for analysis.
• Traditional MCG averaging relies on manual determination of parameters like R-peak thresholds and epoch windows.
• Automated parameter selection is crucial for efficient and reproducible MCG data analysis.

Purpose of the Study:

• To develop and validate a fully automatic algorithm for averaging magnetocardiogram (MCG) data.
• To eliminate the need for manual parameter setting in MCG signal processing.
• To improve the reliability and efficiency of MCG analysis for clinical applications.

Main Methods:

• Utilized magnitude histograms of root-mean-square waveforms for automatic threshold determination (R-peaks and T-peaks).
• Employed peak detection to calculate R-R and R-T intervals for epoch window estimation.
• Incorporated a routine for handling double R-peaks to ensure complete automation.

Main Results:

• The algorithm successfully determined thresholds and epoch windows automatically.
• Average latencies of R-T and R-R intervals were accurately calculated.
• The automated averaging process was validated on recordings from 40 normal subjects and 15 patients with myocardial ischemia.

Conclusions:

• The developed algorithm reliably performs automatic MCG averaging.
• This automated approach enhances the efficiency and accuracy of MCG signal analysis.
• The algorithm shows promise for clinical use in diagnosing cardiac conditions like myocardial ischemia.