Reconstructing asynchrony for mechanical ventilation using a hysteresis loop virtual patient model.

Area of Science:

• Biomedical Engineering
• Respiratory Physiology
• Critical Care Medicine

Background:

• Mechanical ventilation (MV) requires accurate patient-specific lung mechanics identification.
• Spontaneous breathing (SB) effort can cause asynchrony, reducing the accuracy of lung mechanics.
• Existing methods struggle with waveform alterations due to breathing asynchrony.

Purpose of the Study:

• To develop and validate a method for detecting and quantifying breathing asynchrony during MV.
• To improve the accuracy of patient-specific lung mechanics identification in the presence of asynchrony.
• To reconstruct accurate ventilated waveforms unaffected by asynchronous breaths.

Main Methods:

• Hysteresis loop analysis (HLA) was used to detect asynchrony and its patterns.
• A nonlinear mechanics hysteresis loop model (HLM) reconstructed waveforms.
• An energy-dissipation metric (E_asyn) quantified asynchrony magnitude.

Main Results:

• Reconstructed pressure-volume (PV) loops showed high accuracy (5% RMSE for test-lung, 10% for clinical data).
• E_asyn accurately matched known asynchrony magnitude in experimental data (RMSE < 4.1%).
• Clinical data demonstrated varying levels of asynchrony detection and quantification across patients.

Conclusions:

• The method accurately reconstructs lung mechanics and detects asynchrony in controlled and clinical settings.
• Clinical validation confirmed the method's robustness and potential as a real-time asynchrony monitoring tool.
• The approach enhances the reliability of patient-specific lung mechanics during mechanical ventilation.