Modelling ventilation with spontaneous breaths: Improving accuracy with shape functions and slice method.

Area of Science:

• Critical Care Medicine
• Biomedical Engineering
• Respiratory Physiology

Background:

• Accurate detection of spontaneous breaths (SBs) and respiratory asynchronies during mechanical ventilation (MV) is crucial for patient care and preventing lung injury.
• Conventional models struggle to accurately capture these events, necessitating improved methods.

Purpose of the Study:

• To introduce novel equations incorporating custom shape functions and the Slice method for a more robust bedside model.
• To enable real-time detection of patient-ventilator asynchronies and improve mechanical ventilation management.

Main Methods:

• Developed three new equations with shape functions for pressure- and volume-dependent elastance changes.
• Created a fourth model integrating these shape functions with the Slice method (RDEA + Slice).
• Validated models using retrospective data from 8 ICU patients, assessing accuracy with R² and Mean Residual Error (MRE).

Main Results:

• The RDEA + Slice model demonstrated strong correlation with patient data (R² 0.90-0.97), significantly outperforming conventional models (R² 0.25-0.87).
• Achieved significantly lower MRE values (0.012-0.032), indicating superior accuracy in capturing dynamic ventilatory changes.
• Identifiability analysis confirmed reliable model parameter estimation, supporting clinical applicability.

Conclusions:

• The novel bedside models, particularly RDEA + Slice, show potential for improving mechanical ventilation.
• Accurate capture of SBs and asynchronies can refine ventilator settings, reduce lung injury, and support personalized ICU care.