Detection of patient-ventilator asynchrony from mechanical ventilation waveforms using a two-layer long short-term memory neural network

Patient-ventilator asynchrony classification in mechanically ventilated patients: Model-based or machine learning method?

BACKGROUND AND OBJECTIVE

Patient-ventilator asynchrony (PVA) is associated with poor clinical outcomes and remains under-monitored. Automated PVA detection would enable complete monitoring standard observational methods do not allow. While model-based and machine learning PVA approaches exist, they have variable performance and can miss specific PVA events. This study compares a model and rule-based algorithm with a machine learning PVA method by retrospectively validating both methods using an independent patient cohort.

METHODS

Hysteresis loop analysis (HLA) which is a rule-based method (RBM) and a tri-input convolutional neural network (TCNN) machine learning model are used to classify 7 different types of PVA, including: 1) flow asynchrony; 2) reverse triggering; 3) premature cycling; 4) double triggering; 5) delayed cycling; 6) ineffective efforts; and 7) auto triggering. Class activation mapping (CAM) heatmaps visualise sections of respiratory waveforms the TCNN model uses for decision making, improving result interpretability. Both PVA classification methods were used to classify incidence in an independent retrospective clinical cohort of 11 mechanically ventilated patients for validation and performance comparison.

RESULTS

Self-validation with the training dataset shows overall better HLA performance (accuracy, sensitivity, specificity: 97.5 %, 96.6 %, 98.1 %) compared to the TCNN model (accuracy, sensitivity, specificity: 89.5 %, 98.3 %, 83.9 %). In this study, the TCNN model demonstrates higher sensitivity in detecting PVA, but HLA was better at identifying non-PVA breathing cycles due to its rule-based nature. While the overall AI identified by both classification methods are very similar, the intra-patient distribution of each PVA type varies between HLA and TCNN.

CONCLUSION

The collective findings underscore the efficacy of both HLA and TCNN in PVA detection, indicating the potential for real-time continuous monitoring of PVA. While ML methods such as TCNN demonstrate good PVA identification performance, it is essential to ensure optimal model architecture and diversity in training data before widespread uptake as standard care. Moving forward, further validation and adoption of RBM methods, such as HLA, offers an effective approach to PVA detection while providing clear distinction into the underlying patterns of PVA, better aligning with clinical needs for transparency, explicability, adaptability and reliability of these emerging tools for clinical care.

Explainable artificial intelligence (XAI) for predicting the need for intubation in methanol-poisoned patients: a study comparing deep and machine learning models

The need for intubation in methanol-poisoned patients, if not predicted in time, can lead to irreparable complications and even death. Artificial intelligence (AI) techniques like machine learning (ML) and deep learning (DL) greatly aid in accurately predicting intubation needs for methanol-poisoned patients. So, our study aims to assess Explainable Artificial Intelligence (XAI) for predicting intubation necessity in methanol-poisoned patients, comparing deep learning and machine learning models. This study analyzed a dataset of 897 patient records from Loghman Hakim Hospital in Tehran, Iran, encompassing cases of methanol poisoning, including those requiring intubation (202 cases) and those not requiring it (695 cases). Eight established ML (SVM, XGB, DT, RF) and DL (DNN, FNN, LSTM, CNN) models were used. Techniques such as tenfold cross-validation and hyperparameter tuning were applied to prevent overfitting. The study also focused on interpretability through SHAP and LIME methods. Model performance was evaluated based on accuracy, specificity, sensitivity, F1-score, and ROC curve metrics. Among DL models, LSTM showed superior performance in accuracy (94.0%), sensitivity (99.0%), specificity (94.0%), and F1-score (97.0%). CNN led in ROC with 78.0%. For ML models, RF excelled in accuracy (97.0%) and specificity (100%), followed by XGB with sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%). Overall, RF and XGB outperformed other models, with accuracy (97.0%) and specificity (100%) for RF, and sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%) for XGB. ML models surpassed DL models across all metrics, with accuracies from 93.0% to 97.0% for DL and 93.0% to 99.0% for ML. Sensitivities ranged from 98.0% to 99.37% for DL and 93.0% to 99.0% for ML. DL models achieved specificities from 78.0% to 94.0%, while ML models ranged from 93.0% to 100%. F1-scores for DL were between 93.0% and 97.0%, and for ML between 96.0% and 98.27%. DL models scored ROC between 68.0% and 78.0%, while ML models ranged from 84.0% to 96.08%. Key features for predicting intubation necessity include GCS at admission, ICU admission, age, longer folic acid therapy duration, elevated BUN and AST levels, VBG_HCO3 at initial record, and hemodialysis presence. This study as the showcases XAI's effectiveness in predicting intubation necessity in methanol-poisoned patients. ML models, particularly RF and XGB, outperform DL counterparts, underscoring their potential for clinical decision-making.

Detection and quantitative analysis of patient-ventilator interactions in ventilated infants by deep learning networks

BACKGROUND

The study of patient-ventilator interactions (PVI) in mechanically ventilated neonates is limited by the lack of unified PVI definitions and tools to perform large scale analyses.

METHODS

An observational study was conducted in 23 babies randomly selected from 170 neonates who were ventilated with SIPPV-VG, SIMV-VG or PSV-VG mode for at least 12 h. 500 breaths were randomly selected and manually annotated from each recording to train convolutional neural network (CNN) models for PVI classification.

RESULTS

The average asynchrony index (AI) over all recordings was 52.5%. The most frequently occurring PVIs included expiratory work (median: 28.4%, interquartile range: 23.2-40.2%), late cycling (7.6%, 2.8-10.2%), failed triggering (4.6%, 1.2-6.2%) and late triggering (4.4%, 2.8-7.4%). Approximately 25% of breaths with a PVI had two or more PVIs occurring simultaneously. Binary CNN classifiers were developed for PVIs affecting ≥1% of all breaths (n = 7) and they achieved F1 scores of >0.9 on the test set except for early triggering where it was 0.809.

CONCLUSIONS

PVIs occur frequently in neonates undergoing conventional mechanical ventilation with a significant proportion of breaths containing multiple PVIs. We have developed computational models for seven different PVIs to facilitate automated detection and further evaluation of their clinical significance in neonates.

IMPACT

The study of patient-ventilator interactions (PVI) in mechanically ventilated neonates is limited by the lack of unified PVI definitions and tools to perform large scale analyses. By adapting a recent taxonomy of PVI definitions in adults, we have manually annotated neonatal ventilator waveforms to determine prevalence and co-occurrence of neonatal PVIs. We have also developed binary deep learning classifiers for common PVIs to facilitate their automatic detection and quantification.

Quantifiable identification of flow-limited ventilator dyssynchrony with the deformed lung ventilator model

BACKGROUND

Ventilator dyssynchrony (VD) can worsen lung injury and is challenging to detect and quantify due to the complex variability in the dyssynchronous breaths. While machine learning (ML) approaches are useful for automating VD detection from the ventilator waveform data, scalable severity quantification and its association with pathogenesis and ventilator mechanics remain challenging.

OBJECTIVE

We develop a systematic framework to quantify pathophysiological features observed in ventilator waveform signals such that they can be used to create feature-based severity stratification of VD breaths.

METHODS

A mathematical model was developed to represent the pressure and volume waveforms of individual breaths in a feature-based parametric form. Model estimates of respiratory effort strength were used to assess the severity of flow-limited (FL)-VD breaths compared to normal breaths. A total of 93,007 breath waveforms from 13 patients were analyzed.

RESULTS

A novel model-defined continuous severity marker was developed and used to estimate breath phenotypes of FL-VD breaths. The phenotypes had a predictive accuracy of over 97% with respect to the previously developed ML-VD identification algorithm. To understand the incidence of FL-VD breaths and their association with the patient state, these phenotypes were further successfully correlated with ventilator-measured parameters and electronic health records.

CONCLUSION

This work provides a computational pipeline to identify and quantify the severity of FL-VD breaths and paves the way for a large-scale study of VD causes and effects. This approach has direct application to clinical practice and in meaningful knowledge extraction from the ventilator waveform data.

Flow starvation during square-flow assisted ventilation detected by supervised deep learning techniques

BACKGROUND

Flow starvation is a type of patient-ventilator asynchrony that occurs when gas delivery does not fully meet the patients' ventilatory demand due to an insufficient airflow and/or a high inspiratory effort, and it is usually identified by visual inspection of airway pressure waveform. Clinical diagnosis is cumbersome and prone to underdiagnosis, being an opportunity for artificial intelligence. Our objective is to develop a supervised artificial intelligence algorithm for identifying airway pressure deformation during square-flow assisted ventilation and patient-triggered breaths.

METHODS

Multicenter, observational study. Adult critically ill patients under mechanical ventilation > 24 h on square-flow assisted ventilation were included. As the reference, 5 intensive care experts classified airway pressure deformation severity. Convolutional neural network and recurrent neural network models were trained and evaluated using accuracy, precision, recall and F1 score. In a subgroup of patients with esophageal pressure measurement (ΔPes), we analyzed the association between the intensity of the inspiratory effort and the airway pressure deformation.

RESULTS

6428 breaths from 28 patients were analyzed, 42% were classified as having normal-mild, 23% moderate, and 34% severe airway pressure deformation. The accuracy of recurrent neural network algorithm and convolutional neural network were 87.9% [87.6-88.3], and 86.8% [86.6-87.4], respectively. Double triggering appeared in 8.8% of breaths, always in the presence of severe airway pressure deformation. The subgroup analysis demonstrated that 74.4% of breaths classified as severe airway pressure deformation had a ΔPes > 10 cmH2O and 37.2% a ΔPes > 15 cmH2O.

CONCLUSIONS

Recurrent neural network model appears excellent to identify airway pressure deformation due to flow starvation. It could be used as a real-time, 24-h bedside monitoring tool to minimize unrecognized periods of inappropriate patient-ventilator interaction.

The Development, Optimization, and Validation of Four Different Machine Learning Algorithms to Identify Ventilator Dyssynchrony

Invasive mechanical ventilation can worsen lung injury. Ventilator dyssynchrony (VD) may propagate ventilator-induced lung injury (VILI) and is challenging to detect and systematically monitor because each patient takes approximately 25,000 breaths a day yet some types of VD are rare, accounting for less than 1% of all breaths. Therefore, we sought to develop and validate accurate machine learning (ML) algorithms to detect multiple types of VD by leveraging esophageal pressure waveform data to quantify patient effort with airway pressure, flow, and volume data generated during mechanical ventilation, building a computational pipeline to facilitate the study of VD.

MATERIALS AND METHODS

We collected ventilator waveform and esophageal pressure data from 30 patients admitted to the ICU. Esophageal pressure allows the measurement of transpulmonary pressure and patient effort. Waveform data were cleaned, features considered essential to VD detection were calculated, and a set of 10,000 breaths were manually labeled. Four ML algorithms were trained to classify each type of VD: logistic regression, support vector classification, random forest, and XGBoost.

RESULTS

We trained ML models to detect different families and seven types of VD with high sensitivity (>90% and >80%, respectively). Three types of VD remained difficult for ML to classify because of their rarity and lack of sample size. XGBoost classified breaths with increased specificity compared to other ML algorithms.

DISCUSSION

We developed ML models to detect multiple types of VD accurately. The ability to accurately detect multiple VD types addresses one of the significant limitations in understanding the role of VD in affecting patient outcomes.

CONCLUSION

ML models identify multiple types of VD by utilizing esophageal pressure data and airway pressure, flow, and volume waveforms. The development of such computational pipelines will facilitate the identification of VD in a scalable fashion, allowing for the systematic study of VD and its impact on patient outcomes.

Development of a hybrid LSTM with chimp optimization algorithm for the pressure ventilator prediction

The utilization of mechanical ventilation is of utmost importance in the management of individuals afflicted with severe pulmonary conditions. During periods of a pandemic, it becomes imperative to build ventilators that possess the capability to autonomously adapt parameters over the course of treatment. In order to fulfil this requirement, a research investigation was undertaken with the aim of forecasting the magnitude of pressure applied on the patient by the ventilator. The aforementioned forecast was derived from a comprehensive analysis of many variables, including the ventilator's characteristics and the patient's medical state. This analysis was conducted utilizing a sophisticated computational model referred to as Long Short-Term Memory (LSTM). To enhance the predictive accuracy of the LSTM model, the researchers utilized the Chimp Optimization method (ChoA) method. The integration of LSTM and ChoA led to the development of the LSTM-ChoA model, which successfully tackled the issue of hyperparameter selection for the LSTM model. The experimental results revealed that the LSTM-ChoA model exhibited superior performance compared to alternative optimization algorithms, namely whale grey wolf optimizer (GWO), optimization algorithm (WOA), and particle swarm optimization (PSO). Additionally, the LSTM-ChoA model outperformed regression models, including K-nearest neighbor (KNN) Regressor, Random and Forest (RF) Regressor, and Support Vector Machine (SVM) Regressor, in accurately predicting ventilator pressure. The findings indicate that the suggested predictive model, LSTM-ChoA, demonstrates a reduced mean square error (MSE) value. Specifically, when comparing ChoA with GWO, the MSE fell by around 14.8%. Furthermore, when comparing ChoA with PSO and WOA, the MSE decreased by approximately 60%. Additionally, the analysis of variance (ANOVA) findings revealed that the p-value for the LSTM-ChoA model was 0.000, which is less than the predetermined significance level of 0.05. This indicates that the results of the LSTM-ChoA model are statistically significant.

Automated evaluation of typical patient-ventilator asynchronies based on lung hysteretic responses

BACKGROUND

Patient-ventilator asynchrony is common during mechanical ventilation (MV) in intensive care unit (ICU), leading to worse MV care outcome. Identification of asynchrony is critical for optimizing MV settings to reduce or eliminate asynchrony, whilst current clinical visual inspection of all typical types of asynchronous breaths is difficult and inefficient. Patient asynchronies create a unique pattern of distortions in hysteresis respiratory behaviours presented in pressure-volume (PV) loop.

METHODS

Identification method based on hysteretic lung mechanics and hysteresis loop analysis is proposed to delineate the resulted changes of lung mechanics in PV loop during asynchronous breathing, offering detection of both its incidence and 7 major types. Performance is tested against clinical patient data with comparison to visual inspection conducted by clinical doctors.

RESULTS

The identification sensitivity and specificity of 11 patients with 500 breaths for each patient are above 89.5% and 96.8% for all 7 types, respectively. The average sensitivity and specificity across all cases are 94.6% and 99.3%, indicating a very good accuracy. The comparison of statistical analysis between identification and human inspection yields the essential same clinical judgement on patient asynchrony status for each patient, potentially leading to the same clinical decision for setting adjustment.

CONCLUSIONS

The overall results validate the accuracy and robustness of the identification method for a bedside monitoring, as well as its ability to provide a quantified metric for clinical decision of ventilator setting. Hence, the method shows its potential to assist a more consistent and objective assessment of asynchrony without undermining the efficacy of the current clinical practice.

A Novel Application of Spectrograms with Machine Learning Can Detect Patient Ventilator Dyssynchrony

Patients in intensive care units are frequently supported by mechanical ventilation. There is increasing awareness of patient-ventilator dyssynchrony (PVD), a mismatch between patient respiratory effort and assistance provided by the ventilator, as a risk factor for infection, narcotic exposure, lung injury, and adverse neurocognitive effects. One of the most injurious consequences of PVD are double cycled (DC) breaths when two breaths are delivered by the ventilator instead of one. Prior efforts to identify PVD have limited efficacy. An automated method to identify PVD, independent of clinician expertise, acumen, or time, would potentially permit early, targeted treatment to avoid further harm. We performed secondary analyses of data from a clinical trial of children with acute respiratory distress syndrome. Waveforms of ventilator flow, airway pressure and esophageal manometry were annotated to identify DC breaths and underlying PVD subtypes. Spectrograms were generated from those waveforms to train Convolutional Neural Network (CNN) models in detecting DC and underlying PVD subtypes: Reverse Trigger (RT) and Inadequate Support (IS). The DC breath detection model yielded AUROC of 0.980, while the multi-target detection model for underlying dyssynchrony yielded AUROC of 0.980 (RT) and 0.976 (IS). When operating at 75% sensitivity, DC breath detection had a number needed to alert (NNA) 1.3 (99% specificity), while underlying PVD had a NNA 1.6 (98.5% specificity) for RT and NNA 4.0 (98.2% specificity) for IS. CNNs using spectrograms of ventilator waveforms can identify DC breaths and detect the underlying PVD for targeted clinical interventions.

A Comparative Study on Predication of Appropriate Mechanical Ventilation Mode through Machine Learning Approach

Ventilation mode is one of the most crucial ventilator settings, selected and set by knowledgeable critical care therapists in a critical care unit. The application of a particular ventilation mode must be patient-specific and patient-interactive. The main aim of this study is to provide a detailed outline regarding ventilation mode settings and determine the best machine learning method to create a deployable model for the appropriate selection of ventilation mode on a per breath basis. Per-breath patient data is utilized, preprocessed and finally a data frame is created consisting of five feature columns (inspiratory and expiratory tidal volume, minimum pressure, positive end-expiratory pressure, and previous positive end-expiratory pressure) and one output column (output column consisted of modes to be predicted). The data frame has been split into training and testing datasets with a test size of 30%. Six machine learning algorithms were trained and compared for performance, based on the accuracy, F1 score, sensitivity, and precision. The output shows that the Random-Forest Algorithm was the most precise and accurate in predicting all ventilation modes correctly, out of the all the machine learning algorithms trained. Thus, the Random-Forest machine learning technique can be utilized for predicting optimal ventilation mode setting, if it is properly trained with the help of the most relevant data. Aside from ventilation mode, control parameter settings, alarm settings and other settings may also be adjusted for the mechanical ventilation process utilizing appropriate machine learning, particularly deep learning approaches.

Neural Network-Enabled Identification of Weak Inspiratory Efforts during Pressure Support Ventilation Using Ventilator Waveforms

During pressure support ventilation (PSV), excessive assist results in weak inspiratory efforts and promotes diaphragm atrophy and delayed weaning. The aim of this study was to develop a classifier using a neural network to identify weak inspiratory efforts during PSV, based on the ventilator waveforms. Recordings of flow, airway, esophageal and gastric pressures from critically ill patients were used to create an annotated dataset, using data from 37 patients at 2-5 different levels of support, computing the inspiratory time and effort for every breath. The complete dataset was randomly split, and data from 22 patients (45,650 breaths) were used to develop the model. Using a One-Dimensional Convolutional Neural Network, a predictive model was developed to characterize the inspiratory effort of each breath as weak or not, using a threshold of 50 cmH₂O*s/min. The following results were produced by implementing the model on data from 15 different patients (31,343 breaths). The model predicted weak inspiratory efforts with a sensitivity of 88%, specificity of 72%, positive predictive value of 40%, and negative predictive value of 96%. These results provide a 'proof-of-concept' for the ability of such a neural-network based predictive model to facilitate the implementation of personalized assisted ventilation.

A model-based approach to generating annotated pressure support waveforms

Large numbers of asynchronies during pressure support ventilation cause discomfort and higher work of breathing in the patient, and are associated with an increased mortality. There is a need for real-time decision support to detect asynchronies and assist the clinician towards lung-protective ventilation. Machine learning techniques have been proposed to detect asynchronies, but they require large datasets with sufficient data diversity, sample size, and quality for training purposes. In this work, we propose a method for generating a large, realistic and labeled, synthetic dataset for training and validating machine learning algorithms to detect a wide variety of asynchrony types. We take a model-based approach in which we adapt a non-linear lung-airway model for use in a diverse patient group and add a first-order ventilator model to generate labeled pressure, flow, and volume waveforms of pressure support ventilation. The model was able to reproduce basic measured lung mechanics parameters. Experienced clinicians were not able to differentiate between the simulated waveforms and clinical data (P = 0.44 by Fisher's exact test). The detection performance of the machine learning trained on clinical data gave an overall comparable true positive rate on clinical data and on simulated data (an overall true positive rate of 94.3% and positive predictive value of 93.5% on simulated data and a true positive rate of 98% and positive predictive value of 98% on clinical data). Our findings demonstrate that it is possible to generate labeled pressure and flow waveforms with different types of asynchronies.

Reconstructing asynchrony for mechanical ventilation using a hysteresis loop virtual patient model

BACKGROUND

Patient-specific lung mechanics during mechanical ventilation (MV) can be identified from measured waveforms of fully ventilated, sedated patients. However, asynchrony due to spontaneous breathing (SB) effort can be common, altering these waveforms and reducing the accuracy of identified, model-based, and patient-specific lung mechanics.

METHODS

Changes in patient-specific lung elastance over a pressure-volume (PV) loop, identified using hysteresis loop analysis (HLA), are used to detect the occurrence of asynchrony and identify its type and pattern. The identified HLA parameters are then combined with a nonlinear mechanics hysteresis loop model (HLM) to extract and reconstruct ventilated waveforms unaffected by asynchronous breaths. Asynchrony magnitude can then be quantified using an energy-dissipation metric, Easyn, comparing PV loop area between model-reconstructed and original, altered asynchronous breathing cycles. Performance is evaluated using both test-lung experimental data with a known ground truth and clinical data from four patients with varying levels of asynchrony.

RESULTS

Root mean square errors for reconstructed PV loops are within 5% for test-lung experimental data, and 10% for over 90% of clinical data. Easyn clearly matches known asynchrony magnitude for experimental data with RMS errors < 4.1%. Clinical data performance shows 57% breaths having Easyn > 50% for Patient 1 and 13% for Patient 2. Patient 3 only presents 20% breaths with Easyn > 10%. Patient 4 has Easyn = 0 for 96% breaths showing accuracy in a case without asynchrony.

CONCLUSIONS

Experimental test-lung validation demonstrates the method's reconstruction accuracy and generality in controlled scenarios. Clinical validation matches direct observations of asynchrony in incidence and quantifies magnitude, including cases without asynchrony, validating its robustness and potential efficacy as a clinical real-time asynchrony monitoring tool.

Quantification of respiratory effort magnitude in spontaneous breathing patients using Convolutional Autoencoders

BACKGROUND

Spontaneous breathing (SB) effort during mechanical ventilation (MV) is an important metric of respiratory drive. However, SB effort varies due to a variety of factors, including evolving pathology and sedation levels. Therefore, assessment of SB efforts needs to be continuous and non-invasive. This is important to prevent both over- and under-assistance with MV. In this study, a machine learning model, Convolutional Autoencoder (CAE) is developed to quantify the magnitude of SB effort using only bedside MV airway pressure and flow waveform.

METHOD

The CAE model was trained using 12,170,655 simulated SB flow and normal flow data (NB). The paired SB and NB flow data were simulated using a Gaussian Effort Model (GEM) with 5 basis functions. When the CAE model is given a SB flow input, it is capable of predicting a corresponding NB flow for the SB flow input. The magnitude of SB effort (SBE_Mag) is then quantified as the difference between the SB and NB flows. The CAE model was used to evaluate the SBE_Mag of 9 pressure control/ support datasets. Results were validated using a mean squared error (MSE) fitting between clinical and training SB flows.

RESULTS

The CAE model was able to produce NB flows from the clinical SB flows with the median SBE_Mag of the 9 datasets being 25.39% [IQR: 21.87-25.57%]. The absolute error in SBE_Mag using MSE validation yields a median of 4.77% [IQR: 3.77-8.56%] amongst the cohort. This shows the ability of the GEM to capture the intrinsic details present in SB flow waveforms. Analysis also shows both intra-patient and inter-patient variability in SBE_Mag.

CONCLUSION

A Convolutional Autoencoder model was developed with simulated SB and NB flow data and is capable of quantifying the magnitude of patient spontaneous breathing effort. This provides potential application for real-time monitoring of patient respiratory drive for better management of patient-ventilator interaction.

Assessing the Asynchrony Event Based on the Ventilation Mode for Mechanically Ventilated Patients in ICU

Respiratory system modelling can assist clinicians in making clinical decisions during mechanical ventilation (MV) management in intensive care. However, there are some cases where the MV patients produce asynchronous breathing (asynchrony events) due to the spontaneous breathing (SB) effort even though they are fully sedated. Currently, most of the developed models are only suitable for fully sedated patients, which means they cannot be implemented for patients who produce asynchrony in their breathing. This leads to an incorrect measurement of the actual underlying mechanics in these patients. As a result, there is a need to develop a model that can detect asynchrony in real-time and at the bedside throughout the ventilated days. This paper demonstrates the asynchronous event detection of MV patients in the ICU of a hospital by applying a developed extended time-varying elastance model. Data from 10 mechanically ventilated respiratory failure patients admitted at the International Islamic University Malaysia (IIUM) Hospital were collected. The results showed that the model-based technique precisely detected asynchrony events (AEs) throughout the ventilation days. The patients showed an increase in AEs during the ventilation period within the same ventilation mode. SIMV mode produced much higher asynchrony compared to SPONT mode (p < 0.05). The link between AEs and the lung elastance (AUC Edrs) was also investigated. It was found that when the AEs increased, the AUC Edrs decreased and vice versa based on the results obtained in this research. The information of AEs and AUC Edrs provides the true underlying lung mechanics of the MV patients. Hence, this model-based method is capable of detecting the AEs in fully sedated MV patients and providing information that can potentially guide clinicians in selecting the optimal ventilation mode of MV, allowing for precise monitoring of respiratory mechanics in MV patients.

A Model-based Approach to Generating Annotated Pressure Support Waveforms

During pressure support ventilation, every breath is triggered by the patient. Mismatches between the patient and the ventilator are called asynchronies. It has been reported that large numbers of asynchronies may be harmful and may lead to increased mortality. Automatic asynchrony detection and classification, with subsequent feedback to clinicians, will improve lung ventilation and, possibly, patient outcome. Machine learning techniques have been used to detect asynchronies. However, large, diverse and high-quality training and verification data sets are needed. In this work, we propose a model for generating a large, realistic, labeled, synthetic dataset for training and testing machine learning algorithms to detect a wide variety of asynchrony types. Next to a morphological evaluation of the obtained waveforms, validation of the proposed model includes a test with a machine learning algorithm trained on clinical data.

Identifying Patient-Ventilator Asynchrony on a Small Dataset Using Image-Based Transfer Learning

Mechanical ventilation is an essential life-support treatment for patients who cannot breathe independently. Patient-ventilator asynchrony (PVA) occurs when ventilatory support does not match the needs of the patient and is associated with a series of adverse clinical outcomes. Deep learning methods have shown a strong discriminative ability for PVA detection, but they require a large number of annotated data for model training, which hampers their application to this task. We developed a transfer learning architecture based on pretrained convolutional neural networks (CNN) and used it for PVA recognition based on small datasets. The one-dimensional signal was converted to a two-dimensional image, and features were extracted by the CNN using pretrained weights for classification. A partial dropping cross-validation technique was developed to evaluate model performance on small datasets. When using large datasets, the performance of the proposed method was similar to that of non-transfer learning methods. However, when the amount of data was reduced to 1%, the accuracy of transfer learning was approximately 90%, whereas the accuracy of the non-transfer learning was less than 80%. The findings suggest that the proposed transfer learning method can obtain satisfactory accuracies for PVA detection when using small datasets. Such a method can promote the application of deep learning to detect more types of PVA under various ventilation modes.

An interpretable 1D convolutional neural network for detecting patient-ventilator asynchrony in mechanical ventilation

BACKGROUND AND OBJECTIVE

Patient-ventilator asynchrony (PVA) is the result of a mismatch between the need of patients and the assistance provided by the ventilator during mechanical ventilation. Because the poor interaction between the patient and the ventilator is associated with inferior clinical outcomes, effort should be made to identify and correct their occurrence. Deep learning has shown promising ability in PVA detection; however, lack of network interpretability hampers its application in clinic.

METHODS

We proposed an interpretable one-dimensional convolutional neural network (1DCNN) to detect four most manifestation types of PVA (double triggering, ineffective efforts during expiration, premature cycling and delayed cycling) under pressure control ventilation mode and pressure support ventilation mode. A global average pooling (GAP) layer was incorporated with the 1DCNN model to highlight the sections of the respiratory waveform the model focused on when making a classification. Dilation convolution and batch normalization were introduced to the 1DCNN model for compensating the reduction of performance caused by the GAP layer.

RESULTS

The proposed interpretable 1DCNN exhibited comparable performance with the state-of-the-art deep learning model in PVA detection. The F1 scores for the detection of four types of PVA under pressure control ventilation and pressure support ventilation modes were greater than 0.96. The critical sections of the waveform used to detect PVA were highlighted, and found to be well consistent with the understanding of the respective type of PVA by experts.

CONCLUSIONS

The findings suggest that the proposed 1DCNN can help detect PVA, and enhance the interpretability of the classification process to help clinicians better understand the results obtained from deep learning technology.

Patient–Ventilator Interaction Testing Using the Electromechanical Lung Simulator xPULM™ during V/A-C and PSV Ventilation Mode

During mechanical ventilation, a disparity between flow, pressure and volume demands of the patient and the assistance delivered by the mechanical ventilator often occurs. This paper introduces an alternative approach of simulating and evaluating patient–ventilator interactions with high fidelity using the electromechanical lung simulator xPULM™. The xPULM™ approximates respiratory activities of a patient during alternating phases of spontaneous breathing and apnea intervals while connected to a mechanical ventilator. Focusing on different triggering events, volume assist-control (V/A-C) and pressure support ventilation (PSV) modes were chosen to test patient–ventilator interactions. In V/A-C mode, a double-triggering was detected every third breathing cycle, leading to an asynchrony index of 16.67%, which is classified as severe. This asynchrony causes a significant increase of peak inspiratory pressure (7.96 ± 6.38 vs. 11.09 ± 0.49 cmH2O, p < 0.01)) and peak expiratory flow (−25.57 ± 8.93 vs. 32.90 ± 0.54 L/min, p < 0.01) when compared to synchronous phases of the breathing simulation. Additionally, events of premature cycling were observed during PSV mode. In this mode, the peak delivered volume during simulated spontaneous breathing phases increased significantly (917.09 ± 45.74 vs. 468.40 ± 31.79 mL, p < 0.01) compared to apnea phases. Various dynamic clinical situations can be approximated using this approach and thereby could help to identify undesired patient–ventilation interactions in the future. Rapidly manufactured ventilator systems could also be tested using this approach.

Intelligent decision support with machine learning for efficient management of mechanical ventilation in the intensive care unit - A critical overview

BACKGROUND

Effective management of Mechanical Ventilation (MV) is vital for reducing morbidity, mortality, and cost of healthcare.

OBJECTIVE

This study aims to synthesize evidence for effective MV management through Intelligent decision support (IDS) with Machine Learning (ML).

METHOD

Databases that include EBSCO, IEEEXplore, Google Scholar, SCOPUS, and the Web of Science were systematically searched to identify studies on IDS for effective MV management regarding Tidal Volume (TV), asynchrony, weaning, and other outcomes such as the risk of Prolonged Mechanical ventilation (PMV). The quality of the articles identified was assessed with a modified Joanna Briggs Institute (JBI) critical appraisal checklist for cross-sessional research.

RESULTS

A total of 26 articles were identified for the study that has IDS for TV (n = 2, 7.8 %), asynchrony (n = 9, 34.6 %), weaning (n = 12, 46.2 %), and others (n = 3, 11.5 %). It was affirmed that implementing IDS in MV management will enhance seamless ICU patient management following the utilization of various Machine Learning (ML) algorithms in decision support. The studies relied on (n = 14) ML algorithms to predict the TV, asynchrony, weaning, risk of PMV and Positive End-Expiratory Pressure (PEEP) changes of 11-20262 ICU patients records with model inputs ranging from (n = 1) for timeseries analysis of TV to (n = 47) for weaning prediction.

CONCLUSIONS

The small data size, poor study design, and result reporting, with the heterogeneity of techniques used in the various studies, hampered the development of a unified approach for managing MV efficiency in TV monitoring, asynchrony, and weaning predictions. Notwithstanding, the ensemble model was able to predict TV, asynchrony, and weaning to a higher accuracy than the other algorithms.

Risk Factors for Patient-Ventilator Asynchrony and Its Impact on Clinical Outcomes: Analytics Based on Deep Learning Algorithm

Background and objectives: Patient-ventilator asynchronies (PVAs) are common in mechanically ventilated patients. However, the epidemiology of PVAs and its impact on clinical outcome remains controversial. The current study aims to evaluate the epidemiology and risk factors of PVAs and their impact on clinical outcomes using big data analytics. Methods: The study was conducted in a tertiary care hospital; all patients with mechanical ventilation from June to December 2019 were included for analysis. Negative binomial regression and distributed lag non-linear models (DLNM) were used to explore risk factors for PVAs. PVAs were included as a time-varying covariate into Cox regression models to investigate its influence on the hazard of mortality and ventilator-associated events (VAEs). Results: A total of 146 patients involving 50,124 h and 51,451,138 respiratory cycles were analyzed. The overall mortality rate was 15.6%. Double triggering was less likely to occur during day hours (RR: 0.88; 95% CI: 0.85-0.90; p < 0.001) and occurred most frequently in pressure control ventilation (PCV) mode (median: 3; IQR: 1-9 per hour). Ineffective effort was more likely to occur during day time (RR: 1.09; 95% CI: 1.05-1.13; p < 0.001), and occurred most frequently in PSV mode (median: 8; IQR: 2-29 per hour). The effect of sedatives and analgesics showed temporal patterns in DLNM. PVAs were not associated mortality and VAE in Cox regression models with time-varying covariates. Conclusions: Our study showed that counts of PVAs were significantly influenced by time of the day, ventilation mode, ventilation settings (e.g., tidal volume and plateau pressure), and sedatives and analgesics. However, PVAs were not associated with the hazard of VAE or mortality after adjusting for protective ventilation strategies such as tidal volume, plateau pressure, and positive end expiratory pressure (PEEP).

Lung Mechanics of Mechanically Ventilated Patients With COVID-19: Analytics With High-Granularity Ventilator Waveform Data

Background: Lung mechanics during invasive mechanical ventilation (IMV) for both prognostic and therapeutic implications; however, the full trajectory lung mechanics has never been described for novel coronavirus disease 2019 (COVID-19) patients requiring IMV. The study aimed to describe the full trajectory of lung mechanics of mechanically ventilated COVID-19 patients. The clinical and ventilator setting that can influence patient-ventilator asynchrony (PVA) and compliance were explored. Post-extubation spirometry test was performed to assess the pulmonary function after COVID-19 induced ARDS. Methods: This was a retrospective study conducted in a tertiary care hospital. All patients with IMV due to COVID-19 induced ARDS were included. High-granularity ventilator waveforms were analyzed with deep learning algorithm to obtain PVAs. Asynchrony index (AI) was calculated as the number of asynchronous events divided by the number of ventilator cycles and wasted efforts. Mortality was recorded as the vital status on hospital discharge. Results: A total of 3,923,450 respiratory cycles in 2,778 h were analyzed (average: 24 cycles/min) for seven patients. Higher plateau pressure (Coefficient: -0.90; 95% CI: -1.02 to -0.78) and neuromuscular blockades (Coefficient: -6.54; 95% CI: -9.92 to -3.16) were associated with lower AI. Survivors showed increasing compliance over time, whereas non-survivors showed persistently low compliance. Recruitment maneuver was not able to improve lung compliance. Patients were on supine position in 1,422 h (51%), followed by prone positioning (499 h, 18%), left positioning (453 h, 16%), and right positioning (404 h, 15%). As compared with supine positioning, prone positioning was associated with 2.31 ml/cmH₂O (95% CI: 1.75 to 2.86; p < 0.001) increase in lung compliance. Spirometry tests showed that pulmonary functions were reduced to one third of the predicted values after extubation. Conclusions: The study for the first time described full trajectory of lung mechanics of patients with COVID-19. The result showed that prone positioning was associated with improved compliance; higher plateau pressure and use of neuromuscular blockades were associated with lower risk of AI.