Intelligent model-based advisory system for the management of ventilated intensive care patients. Part II: Advisory system design and evaluation

Simulation to minimise patient self-inflicted lung injury: are we almost there?

Computational modelling has been used to enlighten pathophysiological issues in patients with acute respiratory distress syndrome (ARDS) using a sophisticated, integrated cardiopulmonary model. COVID-19 ARDS is a pathophysiologically distinct entity characterised by dissociation between impairment in gas exchange and respiratory system mechanics, especially in the early stages of ARDS. Weaver and colleagues used computational modelling to elucidate factors contributing to generation of patient self-inflicted lung injury, and evaluated the effects of various spontaneous respiratory efforts with different oxygenation and ventilatory support modes. Their findings indicate that mechanical forces generated in the lung parenchyma are only counterbalanced when the respiratory support mode reduces the intensity of respiratory efforts.

Development of an Interactive AI System for the Optimal Timing Prediction of Successful Weaning from Mechanical Ventilation for Patients in Respiratory Care Centers

Successful weaning from prolonged mechanical ventilation (MV) is an important issue in respiratory care centers (RCCs). Delayed or premature extubation increases both the risk of adverse outcomes and healthcare costs. However, the accurate evaluation of the timing of successful weaning from MV is very challenging in RCCs. This study aims to utilize artificial intelligence algorithms to build predictive models for the successful timing of the weaning of patients from MV in RCCs and to implement a dashboard with the best model in RCC settings. A total of 670 intubated patients in the RCC in Chi Mei Medical Center were included in the study. Twenty-six feature variables were selected to build the predictive models with artificial intelligence (AI)/machine-learning (ML) algorithms. An interactive dashboard with the best model was developed and deployed. A preliminary impact analysis was then conducted. Our results showed that all seven predictive models had a high area under the receiver operating characteristic curve (AUC), which ranged from 0.792 to 0.868. The preliminary impact analysis revealed that the mean number of ventilator days required for the successful weaning of the patients was reduced by 0.5 after AI intervention. The development of an AI prediction dashboard is a promising method to assist in the prediction of the optimal timing of weaning from MV in RCC settings. However, a systematic prospective study of AI intervention is still needed.

A physiology-based mathematical model for the selection of appropriate ventilator controls for lung and diaphragm protection

Mechanical ventilation is used to sustain respiratory function in patients with acute respiratory failure. To aid clinicians in consistently selecting lung- and diaphragm-protective ventilation settings, a physiology-based decision support system is needed. To form the foundation of such a system, a comprehensive physiological model which captures the dynamics of ventilation has been developed. The Lung and Diaphragm Protective Ventilation (LDPV) model centers around respiratory drive and incorporates respiratory system mechanics, ventilator mechanics, and blood acid–base balance. The model uses patient-specific parameters as inputs and outputs predictions of a patient’s transpulmonary and esophageal driving pressures (outputs most clinically relevant to lung and diaphragm safety), as well as their blood pH, under various ventilator and sedation conditions. Model simulations and global optimization techniques were used to evaluate and characterize the model. The LDPV model is demonstrated to describe a CO₂ respiratory response that is comparable to what is found in literature. Sensitivity analysis of the model indicate that the ventilator and sedation settings incorporated in the model have a significant impact on the target output parameters. Finally, the model is seen to be able to provide robust predictions of esophageal pressure, transpulmonary pressure and blood pH for patient parameters with realistic variability. The LDPV model is a robust physiological model which produces outputs which directly target and reflect the risk of ventilator-induced lung and diaphragm injury. Ventilation and sedation parameters are seen to modulate the model outputs in accordance with what is currently known in literature.

Clinical decision support recommending ventilator settings during noninvasive ventilation

NIV therapy is used to provide positive pressure ventilation for patients. There are protocols describing what ventilator settings to use to initialize NIV; however, the guidelines for titrating ventilator settings are less specific. We developed an advisory system to recommend NIV ventilator setting titration and recorded respiratory therapist agreement rates at the bedside. We developed an algorithm (NIV advisor) to recommend when to change the non-invasive ventilator settings of IPAP, EPAP, and FiO₂ based on patient respiratory parameters. The algorithm utilized a multi-target approach; oxygenation, ventilation, and patient effort. The NIV advisor recommended ventilator settings to move the patient's respiratory parameters in a preferred target range. We implemented a pilot study evaluating the usability of the NIV advisor on 10 patients receiving critical care with non-invasive ventilation (NIV). Respiratory therapists were asked their agreement on recommendations from the NIV advisor at the patient's bedside. Bedside respiratory therapists agreed with 91% of the ventilator setting recommendations from the NIV advisor. The POB and VT values were the respiratory parameters that were most often out of the preferred target range. The IPAP ventilator setting was the setting most often considered in need of changing by the NIV advisor. The respiratory therapists agreed with the majority of the recommendations from the NIV advisor. We consider the IPAP recommendations informative in providing the respiratory therapist assistance in targeting preferred POB and Vt values, as these values were frequently out of the target ranges. This pilot implementation was unable to produce the results required to determine the value of the EPAP recommendations. The FiO₂ recommendations from the NIV advisor were treated as ancillary information behind the IPAP recommendations.

Dynamic Characteristics of Mechanical Ventilation System of Double Lungs with Bi-Level Positive Airway Pressure Model

In recent studies on the dynamic characteristics of ventilation system, it was considered that human had only one lung, and the coupling effect of double lungs on the air flow can not be illustrated, which has been in regard to be vital to life support of patients. In this article, to illustrate coupling effect of double lungs on flow dynamics of mechanical ventilation system, a mathematical model of a mechanical ventilation system, which consists of double lungs and a bi-level positive airway pressure (BIPAP) controlled ventilator, was proposed. To verify the mathematical model, a prototype of BIPAP system with a double-lung simulators and a BIPAP ventilator was set up for experimental study. Lastly, the study on the influences of key parameters of BIPAP system on dynamic characteristics was carried out. The study can be referred to in the development of research on BIPAP ventilation treatment and real respiratory diagnostics.

A Comparative Data-Based Modeling Study on Respiratory CO2 Gas Exchange during Mechanical Ventilation

The goal of this study is to derive a minimally complex but credible model of respiratory CO2 gas exchange that may be used in systematic design and pilot testing of closed-loop end-tidal CO2 controllers in mechanical ventilation. We first derived a candidate model that captures the essential mechanisms involved in the respiratory CO2 gas exchange process. Then, we simplified the candidate model to derive two lower-order candidate models. We compared these candidate models for predictive capability and reliability using experimental data collected from 25 pediatric subjects undergoing dynamically varying mechanical ventilation during surgical procedures. A two-compartment model equipped with transport delay to account for CO2 delivery between the lungs and the tissues showed modest but statistically significant improvement in predictive capability over the same model without transport delay. Aggregating the lungs and the tissues into a single compartment further degraded the predictive fidelity of the model. In addition, the model equipped with transport delay demonstrated superior reliability to the one without transport delay. Further, the respiratory parameters derived from the model equipped with transport delay, but not the one without transport delay, were physiologically plausible. The results suggest that gas transport between the lungs and the tissues must be taken into account to accurately reproduce the respiratory CO2 gas exchange process under conditions of wide-ranging and dynamically varying mechanical ventilation conditions.

A simple gas exchange model predicting arterial oxygen content for various FiO2 levels

The application of mechanical ventilation is a life-saving routine therapy that allows the patient to overcome the physiological impact of surgeries, trauma or critical illness by ensuring vital oxygenation and carbon dioxide removal. Above a certain level of minute ventilation (usually set to ensure acceptable carbon dioxide removal and oxygenation) oxygenation is only marginally affected by a further increase in minute ventilation. Thus, oxygenation is predominantly influenced by inspiratory oxygen fraction (FiO2) Usually, finding the appropriate setting is a trial-and-error procedure, as the clinician is unaware of the exact value that needs to be set in order to reach the desired arterial oxygen partial pressures (PaO2) in the patient.

A mathematical model approach quantifying patients' response to changes in mechanical ventilation: evaluation in volume support

This paper presents a mathematical model-approach to describe and quantify patient-response to changes in ventilator support. The approach accounts for changes in metabolism (V̇O2, V̇CO2) and serial dead space (VD), and integrates six physiological models of: pulmonary gas-exchange; acid-base chemistry of blood, and cerebrospinal fluid; chemoreflex respiratory-drive; ventilation; and degree of patients' respiratory muscle-response. The approach was evaluated with data from 12 patients on volume support ventilation mode. The models were tuned to baseline measurements of respiratory gases, ventilation, arterial acid-base status, and metabolism. Clinical measurements and model simulated values were compared at five ventilator support levels. The models were shown to adequately describe data in all patients (χ(2), p > 0.2) accounting for changes in V̇CO2, VD and inadequate respiratory muscle-response. F-ratio tests showed that this approach provides a significantly better (p < 0.001) description of measured data than: (a) a similar model omitting the degree of respiratory muscle-response; and (b) a model of constant alveolar ventilation. The approach may help predict patients' response to changes in ventilator support at the bedside.

Simulations for mechanical ventilation in children: review and future prospects

Mechanical ventilation is a very effective therapy, but with many complications. Simulators are used in many fields, including medicine, to enhance safety issues. In the intensive care unit, they are used for teaching cardiorespiratory physiology and ventilation, for testing ventilator performance, for forecasting the effect of ventilatory support, and to determine optimal ventilatory management. They are also used in research and development of clinical decision support systems (CDSSs) and explicit computerized protocols in closed loop. For all those reasons, cardiorespiratory simulators are one of the tools that help to decrease mechanical ventilation duration and complications. This paper describes the different types of simulators described in the literature for physiologic simulation and modeling of the respiratory system, including a new simulator (SimulResp), and proposes a validation process for these simulators.

Optimization of mechanical ventilator settings for pulmonary disease states

The selection of mechanical ventilator settings that ensure adequate oxygenation and carbon dioxide clearance while minimizing the risk of ventilator-associated lung injury (VALI) is a significant challenge for intensive-care clinicians. Current guidelines are largely based on previous experience combined with recommendations from a limited number of in vivo studies whose data are typically more applicable to populations than to individuals suffering from particular diseases of the lung. By combining validated computational models of pulmonary pathophysiology with global optimization algorithms, we generate in silico experiments to examine current practice and uncover optimal combinations of ventilator settings for individual patient and disease states. Formulating the problem as a multiobjective, multivariable constrained optimization problem, we compute settings of tidal volume, ventilation rate, inspiratory/expiratory ratio, positive end-expiratory pressure and inspired fraction of oxygen that optimally manage the tradeoffs between ensuring adequate oxygenation and carbon dioxide clearance and minimizing the risk of VALI for different pulmonary disease scenarios.

Retrospective evaluation of a decision support system for controlled mechanical ventilation

Management of mechanical ventilation in intensive care patients is complicated by conflicting clinical goals. Decision support systems (DSS) may support clinicians in finding the correct balance. The objective of this study was to evaluate a computerized model-based DSS for its advice on inspired oxygen fraction, tidal volume and respiratory frequency. The DSS was retrospectively evaluated in 16 intensive care patient cases, with physiological models fitted to the retrospective data and then used to simulate patient response to changes in therapy. Sensitivity of the DSS's advice to variations in cardiac output (CO) was evaluated. Compared to the baseline ventilator settings set as part of routine clinical care, the system suggested lower tidal volumes and inspired oxygen fraction, but higher frequency, with all suggestions and the model simulated outcome comparing well with the respiratory goals of the Acute Respiratory Distress Syndrome Network from 2000. Changes in advice with CO variation of about 20% were negligible except in cases of high oxygen consumption. Results suggest that the DSS provides clinically relevant and rational advice on therapy in agreement with current 'best practice', and that the advice is robust to variation in CO.

Control of breathing during mechanical ventilation: who is the boss?

Over the past decade, concepts of control of breathing have increasingly moved from being theoretical concepts to "real world" applied science. The purpose of this review is to examine the basics of control of breathing, discuss the bidirectional relationship between control of breathing and mechanical ventilation, and critically assess the application of this knowledge at the patient's bedside. The principles of control of breathing remain under-represented in the training curriculum of respiratory therapists and pulmonologists, whereas the day-to-day bedside application of the principles of control of breathing continues to suffer from a lack of outcomes-based research in the intensive care unit. In contrast, the bedside application of the principles of control of breathing to ambulatory subjects with sleep-disordered breathing has out-stripped that in critically ill patients. The evolution of newer technologies, faster real-time computing abilities, and miniaturization of ventilator technology can bring the concepts of control of breathing to the bedside and benefit the critically ill patient. However, market forces, lack of scientific data, lack of research funding, and regulatory obstacles need to be surmounted.