Pulmonary Response Prediction Comparison through Three Personalized Basis Functions in a Virtual Patient Model

Abstract

Recruitment manoeuvres with subsequent positive-end-expiratory-pressure (PEEP) have proved effective in recruiting lung volume and preventing alveoli collapse. However, determining a safe, effective, and patient-specific positive end-expiratory pressure (PEEP) is not standardized. This study examines three physiologically relevant basis function sets in conjunction with Hysteresis Loop Modelling (HLM) to enable better elastance evolution predictions for a virtual patient model, adding novel elements to model and predict distension. Prediction accuracy and robustness are validated against 18 volume controlled ventilated (VCV) patients at 7 different baseline PEEP levels (0 to 12cmH2O) yielding 623 cases. Predictions were made, up to 12cmH2O covering 6 × 2cmH2O PEEP steps. The 3 basis function sets yield an absolute median peak inspiratory pressure prediction error of 1.63cmH2O for VCV patients. Overall results demonstrate recruitment mechanics are best captured by an exponential basis function, matching physiological expectations, and accurately capture, for the first time, distension mechanics enabling the risk of lung injury to be predicted before changing ventilator settings. The overall outcomes significantly extend and more fully validate this virtual mechanical ventilation patient model.