Fully Automated Subtraction of Heart Activity for Fetal Magnetoencephalography Data^.

Area of Science:

• Biomedical Engineering
• Neuroscience
• Medical Imaging

Background:

• Fetal magnetoencephalography (fMEG) records fetal brain activity but is challenged by low signal amplitude and interference from maternal/fetal heart activity.
• Current methods for signal interference removal can degrade or redistribute the fetal brain signal, limiting analysis accuracy.

Purpose of the Study:

• To develop and evaluate a novel, fully automated procedure for heart activity removal in fMEG data.
• To improve the extraction of fetal brain signals while preserving signal integrity and enhancing signal-to-noise ratio.

Main Methods:

• Developed FAUNA (Fully Automated removal of heart activity) utilizing Principal Component Analysis (PCA) and Ridge Regression.
• Compared FAUNA's performance against the established orthogonal projection (OP) algorithm using simulated fetal brain activity datasets.
• Analyzed signal extraction efficiency, signal-to-noise ratio improvement, and signal redistribution across sensors.

Main Results:

• FAUNA successfully extracted fetal brain signals with a significantly increased signal-to-noise ratio compared to OP.
• The new method demonstrated no redistribution of fetal brain activity across sensors, preserving spatial information.
• FAUNA achieved superior attenuation of interfering heart signals, enabling more reliable automated analysis.

Conclusions:

• FAUNA offers a significant advancement in processing fMEG data by effectively removing cardiac interference.
• This automated approach enhances the quality and reliability of fetal brain signal analysis, supporting clinical applications.
• FAUNA overcomes limitations of current methods, paving the way for more robust fetal neurodevelopmental research.