Basis function identification of lung mechanics in mechanical ventilation for predicting outcomes of therapy changes: A first virtual patient

摘要

Mechanical ventilation is a primary therapy for patients with acute respiratory failure. However, incorrect ventilator settings can cause further lung damage with inter- and intra- patient heterogeneity an issue. Lung protective ventilator settings include titrating positive end-expiratory pressure (PEEP) to the point of minimum elastance. However, predicting minimum elastance can be difficult, and use of high PEEP can lead to ventilator induced lung injury. In this study, a basis function elastance and resistance model is developed to allow use of information available at a lower PEEP level to identify patient-specific mechanics and predict lung mechanics at a higher PEEP. Accurate prediction of pressure and PEEP to within 5% over 8 recruitment manoeuvres validates the functionality of this virtual patient modelling and system identification approach.