Prediction and simulation of PEEP setting effects with machine learning models / Predicción y simulación de los efectos de la configuración de la PEEP con modelos de aprendizaje automático

Objective To establish a new machine learning-based method to adjust positive end-expiratory pressure (PEEP) using only already routinely measured data. Design Retrospective observational study. Setting Intensive care unit (ICU). Patients or participants 51811 mechanically ventilated patients in multiple ICUs in the USA (data from MIMIC-III and eICU databases). Interventions No interventions. Main variables of interest Success parameters of ventilation (arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance). Results The multi-tasking neural network model performed significantly best for all target tasks in the primary test set. The model predicts arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance about 45 min into the future with mean absolute percentage errors of about 21.7%, 10.0% and 15.8%, respectively. The proposed use of the model was demonstrated in case scenarios, where we simulated possible effects of PEEP adjustments for individual cases. Conclusions Our study implies that machine learning approach to PEEP titration is a promising new method which comes with no extra cost once the infrastructure is in place. Availability of databases with most recent ICU patient data is crucial for the refinement of prediction performance. (AU)

Objetivo Establecer un nuevo método basado en el aprendizaje automático para ajustar la presión positiva al final de la espiración (PEEP según sus siglas en inglés) utilizando únicamente datos ya obtenidos de forma rutinaria. Diseño Estudio retrospectivo de observación. Ámbito Unidad de cuidados intesivos (UCI) Pacientes o participantes 51811 pacientes ventilados mecánicamente en múltiples UCIs de EE.UU. (tomados de las bases de datos MIMIC-III y eICU). Intervenciones Sin intervenciones. Variables de interés principales Parametros de éxito de la ventilación (presiones parciales arteriales de oxígeno y dióxido de carbono y distensibilidad del sistema respiratorio). Resultados El modelo de red neuronal multitarea obtuvo los mejores resultados en todos los objetivos del conjunto de pruebas primario. El modelo predice las presiones parciales arteriales de oxígeno y dióxido de carbono así como la distensibilidad del sistema respiratorio con aproximadamente 45 minutos de anticipación, mostrando errores porcentuales absolutos medios de aproximadamente 21.7%, 10.0% y 15.8%, respectivamente. El uso propuesto del modelo se demostró en situaciones hipotéticas en las que se simularon los posibles efectos de los ajustes de PEEP para casos individuales. Conclusiones Nuestro estudio implica que el enfoque de aprendizaje automático para el ajuste de la PEEP es un método nuevo y prometedor que no supone ningún coste adicional una vez que se dispone de la infraestructura necesaria. La disponibilidad de bases de datos con información de pacientes de UCI más recientes es crucial para perfeccionar el rendimiento de la predicción. (AU)

Pulmonary function testing in preoperative high-risk patients.

BACKGROUND: Postoperative respiratory failure is the most frequent complication in postsurgical patients. The purpose of this study is to assess whether pulmonary function testing in high-risk patients during preoperative assessment detects previously unknown respiratory impairments which may influence patient outcomes. METHODS: A targeted patient screening by spirometry and the measurement of the diffusing capacity of the lung for carbon monoxide (DLCO) was implemented in the anesthesia department of a tertiary university hospital. Patients of all surgical disciplines who were at least 75 years old or exhibited reduced exercise tolerance with the metabolic equivalent of task less than four (MET < 4) were examined. Clinical characteristics, history of lung diseases, and smoking status were also recorded. The statistical analysis entailed t-tests, one-way ANOVA, and multiple linear regression with backward elimination for group comparisons. RESULTS: Among 256 included patients, 230 fulfilled the test quality criteria. Eighty-one (35.2%) patients presented obstructive ventilatory disorders, out of which 65 were previously unknown. 38 of the newly diagnosed obstructive disorders were mild, 18 moderate, and 9 severe. One hundred forty-five DLCO measurements revealed 40 (27.6%) previously unknown gas exchange impairments; 21 were mild, 17 moderate, and 2 severe. The pulmonary function parameters of forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), and DLCO were significantly lower than the international reference values of a healthy population. Patients with a lower ASA class and no history of smoking exhibited higher FVC, FEV1, and DLCO values. Reduced exercise tolerance with MET < 4 was strongly associated with lower spirometry values. CONCLUSIONS: Our screening program detected a relevant number of patients with previously unknown obstructive ventilatory disorders and impaired pulmonary gas exchange. This newly discovered sickness is associated with low metabolic equivalents and may influence perioperative outcomes. Whether optimized management of patients with previously unknown impaired lung function leads to a better outcome should be evaluated in multicenter studies. TRIAL REGISTRATION: German Registry of Clinical Studies (DRKS00029337), registered on: June 22nd, 2022.

Regional ventilation distribution before and after laparoscopic lung parenchymal resection.

Objective.The aim of the present study was to evaluate the influence of one-sided pulmonary nodule and tumour on ventilation distribution pre- and post- partial lung resection.Approach.A total of 40 consecutive patients scheduled for laparoscopic lung parenchymal resection were included. Ventilation distribution was measured with electrical impedance tomography (EIT) in supine and surgery lateral positions 72 h before surgery (T1) and 48 h after extubation (T2). Left lung to global ventilation ratio (Fl), the global inhomogeneity index (GI), standard deviation of regional ventilation delay (RVDSD) and pendelluft amplitude (Apendelluft) were calculated to assess the spatial and temporal ventilation distribution.Main results.After surgery (T2), ventilation at the operated chest sides generally deteriorated compared to T1 as expected. For right-side resection, the differences were significant at both supine and left lateral positions (p< 0.001). The change of RVDSDwas in general more heterogeneous. For left-side resection, RVDSDwas worse at T2 compared to T1 at left lateral position (p= 0.002). The other EIT-based parameters showed no significant differences between the two time points. No significant differences were observed between supine and lateral positions for the same time points respectively.Significance.In the present study, we found that the surgery side influenced the ventilation distribution. When the resection was performed on the right lung, the postoperative ipsilateral ventilation was reduced and the right lung ratio fell significantly. When the resection was on the left lung, the ventilation delay was significantly increased.

Pressure- and time-dependent alveolar recruitment/derecruitment in a spatially resolved patient-specific computational model for injured human lungs.

We present a novel computational model for the dynamics of alveolar recruitment/derecruitment (RD), which reproduces the underlying characteristics typically observed in injured lungs. The basic idea is a pressure- and time-dependent variation of the stress-free reference volume in reduced dimensional viscoelastic elements representing the acinar tissue. We choose a variable reference volume triggered by critical opening and closing pressures in a time-dependent manner from a straightforward mechanical point of view. In the case of (partially and progressively) collapsing alveolar structures, the volume available for expansion during breathing reduces and vice versa, eventually enabling consideration of alveolar collapse and reopening in our model. We further introduce a method for patient-specific determination of the underlying critical parameters of the new alveolar RD dynamics when integrated into the tissue elements, referred to as terminal units, of a spatially resolved physics-based lung model that simulates the human respiratory system in an anatomically correct manner. Relevant patient-specific parameters of the terminal units are herein determined based on medical image data and the macromechanical behavior of the lung during artificial ventilation. We test the whole modeling approach for a real-life scenario by applying it to the clinical data of a mechanically ventilated patient. The generated lung model is capable of reproducing clinical measurements such as tidal volume and pleural pressure during various ventilation maneuvers. We conclude that this new model is an important step toward personalized treatment of ARDS patients by considering potentially harmful mechanisms-such as cyclic RD and overdistension-and might help in the development of relevant protective ventilation strategies to reduce ventilator-induced lung injury (VILI).

Setting positive end-expiratory pressure by using electrical impedance tomography.

PURPOSE OF REVIEW: This review presents the principles and possibilities of setting positive end-expiratory pressure (PEEP) using electrical impedance tomography (EIT). It summarizes the major findings of recent studies where EIT was applied to monitor the effects of PEEP on regional lung function and to guide the selection of individualized PEEP setting. RECENT FINDINGS: The most frequent approach of utilizing EIT for the assessment of PEEP effects and the PEEP setting during the time period from January 2022 till June 2023 was based on the analysis of pixel tidal impedance variation, typically acquired during stepwise incremental and/or decremental PEEP variation. The most common EIT parameters were the fraction of ventilation in various regions of interest, global inhomogeneity index, center of ventilation, silent spaces, and regional compliance of the respiratory system. The studies focused mainly on the spatial and less on the temporal distribution of ventilation. Contrast-enhanced EIT was applied in a few studies for the estimation of ventilation/perfusion matching. SUMMARY: The availability of commercial EIT devices resulted in an increase in clinical studies using this bedside imaging technology in neonatal, pediatric and adult critically ill patients. The clinical interest in EIT became evident but the potential of this method in clinical decision-making still needs to be fully exploited.

Electrical Impedance Tomography to Monitor Hypoxemic Respiratory Failure.

Hypoxemic respiratory failure is one of the leading causes of mortality in intensive care. Frequent assessment of individual physiological characteristics and delivery of personalized mechanical ventilation (MV) settings is a constant challenge for clinicians caring for these patients. Electrical impedance tomography (EIT) is a radiation-free bedside monitoring device that is able to assess regional lung ventilation and changes in aeration. With real-time tomographic functional images of the lungs obtained through a thoracic belt, clinicians can visualize and estimate the distribution of ventilation at different ventilation settings or following procedures such as prone positioning. Several studies have evaluated the performance of EIT to monitor the effects of different MV settings in patients with acute respiratory distress syndrome, allowing more personalized MV. For instance, EIT could help clinicians find the positive end-expiratory pressure that represents a compromise between recruitment and overdistension and assess the effect of prone positioning on ventilation distribution. The clinical impact of the personalization of MV remains to be explored. Despite inherent limitations such as limited spatial resolution, EIT also offers a unique noninvasive bedside assessment of regional ventilation changes in the ICU. This technology offers the possibility of a continuous, operator-free diagnosis and real-time detection of common problems during MV. This review provides an overview of the functioning of EIT, its main indices, and its performance in monitoring patients with acute respiratory failure. Future perspectives for use in intensive care are also addressed.

Prediction and simulation of PEEP setting effects with machine learning models.

OBJECTIVE: To establish a new machine learning-based method to adjust positive end-expiratory pressure (PEEP) using only already routinely measured data. DESIGN: Retrospective observational study. SETTING: Intensive care unit (ICU). PATIENTS OR PARTICIPANTS: 51811 mechanically ventilated patients in multiple ICUs in the USA (data from MIMIC-III and eICU databases). INTERVENTIONS: No interventions. MAIN VARIABLES OF INTEREST: Success parameters of ventilation (arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance) RESULTS: The multi-tasking neural network model performed significantly best for all target tasks in the primary test set. The model predicts arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance about 45 min into the future with mean absolute percentage errors of about 21.7%, 10.0% and 15.8%, respectively. The proposed use of the model was demonstrated in case scenarios, where we simulated possible effects of PEEP adjustments for individual cases. CONCLUSIONS: Our study implies that machine learning approach to PEEP titration is a promising new method which comes with no extra cost once the infrastructure is in place. Availability of databases with most recent ICU patient data is crucial for the refinement of prediction performance.

Effect of body position on ventilation distribution in healthy newborn infants: an observational study.

OBJECTIVES: Newborn infants have unique respiratory physiology compared with older children and adults due to their lungs' structural and functional immaturity and highly compliant chest wall. To date, ventilation distribution has seldom been studied in this age group. This study aims to assess the effect of body position on ventilation distribution in spontaneously breathing healthy neonates. DESIGN: Prospective observational study. SETTING: Maternity wards of Oulu University Hospital. PATIENTS: 20 healthy, spontaneously breathing, newborn infants. INTERVENTIONS: Electrical impedance tomography data were recorded with a 32-electrode belt (Sentec AG, Landquart, Switzerland) in six different body positions in random order. Ventilation distribution was retrospectively assessed 10 minutes after each position change. MAIN OUTCOME MEASURES: In each position, regional tidal impedance variation (ΔZ) and ventral-to-dorsal and right-to-left centre of ventilation were measured. RESULTS: The mean global ΔZ was the largest in supine position and it was smaller in prone and lateral positions. Yet, global ΔZ did not differ in supine positions, ventilation distribution was more directed towards the non-dependent lung region in supine tilted position (p<0.001). In prone, a reduction of global ΔZ was observed (p<0.05) corresponding to an amount of 10% of global tidal variation in supine position. In both lateral positions, tidal ventilation was distributed more to the corresponding non-dependent lung region. CONCLUSIONS: Prone or lateral body positioning in healthy spontaneously breathing newborns leads to a redistribution of ventilation to the non-dependent lung regions and at the same time global tidal volume is reduced as compared with supine.

Effect of routine suction on lung aeration in critically ill neonates and young infants measured with electrical impedance tomography.

Endotracheal suctioning is a widely used procedure to remove secretions from the airways of ventilated patients. Despite its prevalence, regional effects of this maneuver have seldom been studied. In this study, we explore its effects on regional lung aeration in neonates and young infants using electrical impedance tomography (EIT) as part of the large EU-funded multicenter observational study CRADL. 200 neonates and young infants in intensive care units were monitored with EIT for up to 72 h. EIT parameters were calculated to detect changes in ventilation distribution, ventilation inhomogeneity and ventilation quantity on a breath-by-breath level 5-10 min before and after suctioning. The intratidal change in aeration over time was investigated by means of regional expiratory time constants calculated from all respiratory cycles using an innovative procedure and visualized by 2D maps of the thoracic cross-section. 344 tracheal suctioning events from 51 patients could be analyzed. They showed no or very small changes of EIT parameters, with a dorsal shift of the center of ventilation by 0.5% of the chest diameter and a 7% decrease of tidal impedance variation after suctioning. Regional time constants did not change significantly. Routine suctioning led to EIT-detectable but merely small changes of the ventilation distribution in this study population. While still a measure requiring further study, the time constant maps may help clinicians interpret ventilation mechanics in specific cases.

BRACETS: Bimodal repository of auscultation coupled with electrical impedance thoracic signals.

BACKGROUND AND OBJECTIVE: Respiratory diseases are among the most significant causes of morbidity and mortality worldwide, causing substantial strain on society and health systems. Over the last few decades, there has been increasing interest in the automatic analysis of respiratory sounds and electrical impedance tomography (EIT). Nevertheless, no publicly available databases with both respiratory sound and EIT data are available. METHODS: In this work, we have assembled the first open-access bimodal database focusing on the differential diagnosis of respiratory diseases (BRACETS: Bimodal Repository of Auscultation Coupled with Electrical Impedance Thoracic Signals). It includes simultaneous recordings of single and multi-channel respiratory sounds and EIT. Furthermore, we have proposed several machine learning-based baseline systems for automatically classifying respiratory diseases in six distinct evaluation tasks using respiratory sound and EIT (A1, A2, A3, B1, B2, B3). These tasks included classifying respiratory diseases at sample and subject levels. The performance of the classification models was evaluated using a 5-fold cross-validation scheme (with subject isolation between folds). RESULTS: The resulting database consists of 1097 respiratory sounds and 795 EIT recordings acquired from 78 adult subjects in two countries (Portugal and Greece). In the task of automatically classifying respiratory diseases, the baseline classification models have achieved the following average balanced accuracy: Task A1 - 77.9±13.1%; Task A2 - 51.6±9.7%; Task A3 - 38.6±13.1%; Task B1 - 90.0±22.4%; Task B2 - 61.4±11.8%; Task B3 - 50.8±10.6%. CONCLUSION: The creation of this database and its public release will aid the research community in developing automated methodologies to assess and monitor respiratory function, and it might serve as a benchmark in the field of digital medicine for managing respiratory diseases. Moreover, it could pave the way for creating multi-modal robust approaches for that same purpose.

Fluctuations in Oxygen Saturation during Synchronized Nasal Intermittent Positive Pressure Ventilation and Nasal High-Frequency Oscillatory Ventilation in Very Low Birth Weight Infants: A Randomized Crossover Trial.

BACKGROUND: Very low birth weight (VLBW) infants on noninvasive ventilation (NIV) experience frequent fluctuations in oxygen saturation (SpO2) that are associated with an increased risk for mortality and severe morbidities. METHODS: In this randomized crossover trial, VLBW infants (n = 22) born 22+3 to 28+0 weeks on NIV with supplemental oxygen were allocated on two consecutive days in random order to synchronized nasal intermittent positive pressure ventilation (sNIPPV) and nasal high-frequency oscillatory ventilation (nHFOV) for 8 h. nHFOV and sNIPPV were set to equivalent mean airway pressure and transcutaneous pCO2. Primary outcome was the time spent within the SpO2 target (88-95%). RESULTS: During sNIPPV, VLBW infants spent significantly more time within the SpO2 target (59.9%) than during nHFOV (54.6%). The proportion of time spent in hypoxemia (22.3% vs. 27.1%) and the mean fraction of supplemental oxygen (FiO2) (29.4% vs. 32.8%) were significantly reduced during sNIPPV, while the respiratory rate (50.1 vs. 42.6) was significantly higher. Mean SpO2, SpO2 above the target, number of prolonged (>1 min) and severe (SpO2 <80%) hypoxemic episodes, parameters of cerebral tissue oxygenation using NIRS, number of FiO2 adjustments, heart rate, number of bradycardias, abdominal distension and transcutaneous pCO2 did not differ between both interventions. CONCLUSIONS: In VLBW infants with frequent fluctuations in SpO2, sNIPPV is more efficient than nHFOV to retain the SpO2 target and to reduce FiO2 exposure. These results demand more detailed investigations into cumulative oxygen toxicities during different modes of NIV over the weaning period, particularly with regard to consequences for long-term outcomes.

An approach to study recruitment/derecruitment dynamics in a patient-specific computational model of an injured human lung.

We present a new approach for physics-based computational modeling of diseased human lungs. Our main object is the development of a model that takes the novel step of incorporating the dynamics of airway recruitment/derecruitment into an anatomically accurate, spatially resolved model of respiratory system mechanics, and the relation of these dynamics to airway dimensions and the biophysical properties of the lining fluid. The importance of our approach is that it potentially allows for more accurate predictions of where mechanical stress foci arise in the lungs, since it is at these locations that injury is thought to arise and propagate from. We match the model to data from a patient with acute respiratory distress syndrome (ARDS) to demonstrate the potential of the model for revealing the underlying derangements in ARDS in a patient-specific manner. To achieve this, the specific geometry of the lung and its heterogeneous pattern of injury are extracted from medical CT images. The mechanical behavior of the model is tailored to the patient's respiratory mechanics using measured ventilation data. In retrospective simulations of various clinically performed, pressure-driven ventilation profiles, the model adequately reproduces clinical quantities measured in the patient such as tidal volume and change in pleural pressure. The model also exhibits physiologically reasonable lung recruitment dynamics and has the spatial resolution to allow the study of local mechanical quantities such as alveolar strains. This modeling approach advances our ability to perform patient-specific studies in silico, opening the way to personalized therapies that will optimize patient outcomes.

New application of saline contrast-enhanced electrical impedance tomography method for right ventriculography besides lung perfusion: detection of right-to-left intracardiac shunt.

AIM: Saline contrast-enhanced electrical impedance tomography (EIT) has been used to identify the respiratory failure etiologies through assessment of regional lung perfusion at the bedside. In this study, we introduce a novel approach to detect right-to-left intracardiac shunt based on the center of heart (CoH) parameter determined from the early phase of impedance-time curve after saline bolus injection. METHODS AND RESULT: The timepoints when the saline bolus enters the heart (T0) and the lung regions (T1) are identified at first. A moving time window from T0 to T1 is then generated with steps of 0.5 s and the slope of the impedance-time curve in each pixel within the window calculated. CoH is calculated as the geometric center of pixel slope values in the right-to-left image direction. To illustrate how this method works in practice, we calculated the CoH values at T0 to T1 in 10 control hypoxic patients with no right-to-left shunt. In addition, we examined two critically ill patients with right-to-left intracardiac shunt. one was postcardiac surgery patient who had a residual atrial septal defect by color doppler of transesophageal echocardiograph. The other patient had a congenital heart disease of ventricular septal defect by color doppler of trans-thoracic echocardiography. A large difference in CoH between T0 to T1 was observed in the two patients with intracardiac shunt than in the control patients (11.06 ± 3.17% vs. 1.99 ± 1.43%, P = 0.030). CONCLUSION: saline bolus EIT for lung perfusion might be used as ventriculography to identify the right-to-left intracardiac shunt at the bedside.

The influence of gravity on electrical impedance tomography measurements during upper body position change.

Objective: The aim of the study was to examine the influence of gravity on regional ventilation measured by electrical impedance tomography (EIT) with the standard electrode belt position at the 5th intercostal space during tilting from supine to sitting positions. Methods: A total of 30 healthy volunteers were examined prospectively in supine position during quiet tidal breathing. Subsequently, the bed was tilted so that the upper body of the subjects achieved 30, 60 and 90° every 3 min. Regional ventilation distribution and end-expiratory lung impedance (EELI) were monitored with EIT throughout the whole experiment. Absolute tidal volumes were measured with spirometry and the volume-impedance ratio was calculated for each position. Results: The volume-impedance ratio did not differ statistically between the studied body positions but 11 subjects exhibited a large change in ratio at one of the positions (outside 99.3% coverage). In general, ventilation distribution became more heterogeneous and moved towards dorsal regions as the upper body was tilted to 90-degree position. EELI increased and tidal volume decreased. The lung regions identified at various positions differed significantly. Conclusion: Gravity has non-negligible influence on EIT data, as the upper body tilted from supine to sitting positions. The standard electrode belt position might be reconsidered if ventilation distribution is to be compared between supine and sitting positions.