Coupling of EIT with computational lung modeling for predicting patient-specific ventilatory responses.

Area of Science:

• Biomedical Engineering
• Computational Physiology
• Respiratory Medicine

Background:

• Personalized mechanical ventilation for respiratory failure remains a clinical challenge.
• Current methods lack patient-specific predictive capabilities for optimizing ventilation strategies.

Purpose of the Study:

• To develop and validate a patient-specific computational lung model integrated with electrical impedance tomography (EIT) for personalized mechanical ventilation.
• To enable prediction of patient responses to different ventilation maneuvers before clinical application.

Main Methods:

• Integration of a patient-specific computational lung model (based on CT scans) with a novel 'virtual EIT' module.
• Simulation of EIT images from computational predictions for independent validation against clinical EIT data.
• Testing the coupled approach in an acute respiratory distress syndrome patient.

Main Results:

• The computational model accurately predicted global airflow quantities and local tissue aeration.
• Simulated EIT images showed good agreement with clinically measured EIT data.
• The approach demonstrated good correlation between predicted and measured airflow and EIT data.

Conclusions:

• The proposed framework enables patient-specific prediction of responses to mechanical ventilation.
• This computationally guided approach holds potential for optimizing mechanical ventilation strategies.
• Future applications include assisting in the definition of patient-specific optimal ventilation protocols.