Predicting the ineffective efforts in ICU assisted ventilation via multiparametric models

Abstract

In the context of assisted ventilation in ICU, it is of vital importance to keep a high synchronization between the patient's attempt to breath and the assisted ventilation event. A series of relevant bioparameters continuously monitored, are meant to guide the medical professionals in appropriately adapting the operation and treatment, in this direction. Still, the dynamics and complex causal relations among bioparameters and the ventilation synchronization are not well studied. The purpose of this work is to elaborate on a methodology towards predicting level of ventilation synchronization based on the previous known evolution of monitored bioparameters. A nonlinear model is thus proposed, based on support vector machines regression, and the best input memory is investigated. The proposed model shows good correlation (over 0.7) with the actual outputs, and low error, constituting an encouraging step towards understanding of ICU ventilation dynamic phenomena.