RESUMO
Mechanical ventilation is a lifesaving tool and provides organ support for patients with respiratory failure. However, injurious ventilation due to inappropriate delivery of high tidal volume can initiate or potentiate lung injury. This could lead to acute respiratory distress syndrome, longer duration of mechanical ventilation, ventilator associated conditions and finally increased mortality. In this study, we explore the viability and compare machine learning methods to generate personalized predictive alerts indicating violation of the safe tidal volume per ideal body weight (IBW) threshold that is accepted as the upper limit for lung protective ventilation (LPV), prior to application to patients. We process streams of patient respiratory data recorded per minute from ventilators in an intensive care unit and apply several state-of-the-art time series prediction methods to forecast the behavior of the tidal volume metric per patient, 1 hour ahead. Our results show that boosted regression delivers better predictive accuracy than other methods that we investigated and requires relatively short execution times. Long short-term memory neural networks can deliver similar levels of accuracy but only after much longer periods of data acquisition, further extended by several hours computing time to train the algorithm. Utilizing Artificial Intelligence, we have developed a personalized clinical decision support tool that can predict tidal volume behavior within 10% accuracy and compare alerts recorded from a real world system to highlight that our models would have predicted violations 1 hour ahead and can therefore conclude that the algorithms can provide clinical decision support.
Assuntos
Inteligência Artificial , Respiração Artificial , Humanos , Unidades de Terapia Intensiva , Pulmão , Redes Neurais de Computação , Volume de Ventilação PulmonarRESUMO
A new methodology is proposed to compare database performance for streams of patient respiratory data from patients in an intensive care unit. New metrics are proposed through which databases may be compared both for this and similar streaming applications in the domain of the Internet of Things. Studies are reported using simulated patient data for four freely available databases. The statistical technique of non-parametric bootstrapping is used to minimise the total running time of the tests. We report mean values and bias corrected and accelerated confidence intervals for each metric and use these to compare the databases. We find that, among the four databases tested, ScaleDB is an optimum database technology when handling between 200 and 800 patients in this application, while PostgreSQL performs best outside of this range. Comparing the non-parametric bootstrapping method to a complete set of tests shows that the two approaches give results differing by a few percent.