Separation of cardiogenic oscillations from airflow waveforms using singular spectrum analysis.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Anesthesiology

Background:

• Cardiac contractions cause airflow fluctuations, leading to inaccurate ventilator support.
• This cardiogenic artifact can compromise mechanical ventilation in critical care and anesthesia.
• Improving ventilator function requires accurate removal of these oscillations.

Purpose of the Study:

• To evaluate Singular Spectrum Analysis (SSA) for removing cardiogenic artifacts from ventilator airflow signals.
• To assess the feasibility of using SSA to enhance mechanical ventilator performance.
• To improve the accuracy of airflow data during mechanical ventilation.

Main Methods:

• Collected airflow, CO2, and ECG data at high sampling rates.
• Applied one-dimensional SSA to normalized airflow signals.
• Reconstructed the respiratory waveform using SSA components corresponding to the highest singular values.

Main Results:

• SSA successfully extracted the respiratory component from airflow signals in patients.
• High weighted correlations (>0.96) confirmed the quality of respiratory waveform extraction.
• Cardiogenic oscillations were effectively isolated as lower magnitude components.

Conclusions:

• SSA is a viable method for removing cardiogenic artifacts and noise from airflow signals.
• This signal processing technique can augment ventilator performance.
• Computational feasibility supports the clinical application of SSA in mechanical ventilation.