Respiratory Signal Extraction from ECG using a Phase-Amplitude Cross-Frequency Coupling Index.

Area of Science:

• Biomedical Engineering
• Clinical Neurophysiology
• Signal Processing

Background:

• Direct measurement of respiration is less common than ECG or EEG in clinical settings, particularly in epilepsy monitoring.
• Existing methods for extracting respiration from ECG are susceptible to artifacts.
• There is a need for reliable respiration monitoring during clinical studies, especially for epilepsy patients.

Purpose of the Study:

• To introduce a novel method for extracting respiratory signals from electrocardiogram (ECG) data.
• To evaluate the performance of this new method against established techniques.
• To assess the potential of the new method for use in challenging clinical scenarios, such as during seizures.

Main Methods:

• Utilized phase-amplitude cross-frequency coupling (PAC) to derive respiration from ECG.
• Compared the PAC method with the standard ECG envelope extraction technique.
• Employed wavelet phase coherence to analyze the accuracy of the extracted respiration signals against actual respiration data.

Main Results:

• The PAC method demonstrated comparable performance to the widely used ECG envelope extraction approach.
• No significant differences were found in wavelet phase coherence or predicted respiration rates between the two methods.
• The PAC method exhibited reduced sensitivity to high-amplitude noise compared to existing methods.

Conclusions:

• Phase-amplitude coupling (PAC) offers a viable and artifact-resilient alternative for extracting respiration from ECG.
• The PAC method's robustness to noise makes it a promising tool for respiration monitoring during high-amplitude events like seizures.
• This technique could enhance respiratory monitoring capabilities in epilepsy monitoring units and other clinical studies using only ECG.