A MACHINE LEARNING MODEL FOR REAL-TIME ASYNCHRONOUS BREATHING MONITORING

Convolutional long short-term memory neural network integrated with classifier in classifying type of asynchrony breathing in mechanically ventilated patients

BACKGROUND AND OBJECTIVE

Asynchronous breathing (AB) occurs when a mechanically ventilated patient's breathing does not align with the mechanical ventilator (MV). Asynchrony can negatively impact recovery and outcome, and/or hinder MV management. A model-based method to accurately classify different AB types could automate detection and have a measurable clinical impact.

METHODS

This study presents an approach using a 1-dimensional (1D) of airway pressure data as an input to the convolutional long short-term memory neural network (CNN-LSTM) with a classifier method to classify AB types into three categories: 1) reverse Triggering (RT); 2) premature cycling (PC); and 3) normal breathing (NB), which cover normal breathing and 2 primary forms of AB. Three types of classifier are integrated with the CNN-LSTM model which are random forest (RF), support vector machine (SVM) and logistic regression (LR). Clinical data inputs include measured airway pressure from 7 MV patients in IIUM Hospital ICU under informed consent with a total of 4500 breaths. Model performance is first assessed in a k-fold cross-validation assessing accuracy in comparison to the proposed CNN-LSTM integrated with each type of classifier. Then, confusion matrices are used to summarize classification performance for the CNN without classifier, CNN-LSTM without classifier, and CNN-LSTM with each of the 3 classifiers (RF, SVM, LR).

RESULTS AND DISCUSSION

The 1D CNN-LSTM with classifier method achieves 100 % accuracy using 5-fold cross validation. The confusion matrix results showed that the combined CNN-LSTM model with classifier performed better, demostrating higher accuracy, sensitivity, specificity, and F1 score, all exceeding 83.5 % across all three breathing categories. The CNN model without classifier and CNN-LSTM model without classifier displayed comparatively lower performance, with average values of F1 score below 71.8 % for all three breathing categories.

CONCLUSION

The results validate the effectiveness of the CNN-LSTM neural network model with classifier in accurately detecting and classifying the different categories of AB and NB. Overall, this model-based approach has the potential to precisely classify the type of AB and differentiate normal breathing. With this developed model, a better MV management can be provided at the bedside, and these results justify prospective clinical testing.

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.

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.

Automated detection and classification of patient-ventilator asynchrony by means of machine learning and simulated data

BACKGROUND AND OBJECTIVE

Mechanical ventilation is a lifesaving treatment for critically ill patients in an Intensive Care Unit (ICU) or during surgery. However, one potential harm of mechanical ventilation is related to patient-ventilator asynchrony (PVA). PVA can cause discomfort to the patient, damage to the lungs, and an increase in the length of stay in the ICU and on the ventilator. Therefore, automated detection algorithms are being developed to detect and classify PVAs, with the goal of optimizing mechanical ventilation. However, the development of these algorithms often requires large labeled datasets; these are generally difficult to obtain, as their collection and labeling is a time-consuming and labor-intensive task, which needs to be performed by clinical experts.

METHODS

In this work, we aimed to develop a computer algorithm for the automatic detection and classification of PVA. The algorithm employs a neural network for the detection of the breath of the patient. The development of the algorithm was aided by simulations from a recently published model of the patient-ventilator interaction.

RESULTS

The proposed method was effective, providing an algorithm with reliable detection and classification results of over 90% accuracy. Besides presenting a detection and classification algorithm for a variety of PVAs, here we show that using simulated data in combination with clinical data increases the variability in the training dataset, leading to a gain in performance and generalizability.

CONCLUSIONS

In the future, these algorithms can be utilized to gain a better understanding of the clinical impact of PVAs and help clinicians to better monitor their ventilation strategies.

Predicting mechanical ventilation effects on six human tissue transcriptomes

BACKGROUND

Mechanical ventilation (MV) is a lifesaving therapy used for patients with respiratory failure. Nevertheless, MV is associated with numerous complications and increased mortality. The aim of this study is to define the effects of MV on gene expression of direct and peripheral human tissues.

METHODS

Classification models were applied to Genotype-Tissue Expression Project (GTEx) gene expression data of six representative tissues-liver, adipose, skin, nerve-tibial, muscle and lung, for performance comparison and feature analysis. We utilized 18 prediction models using the Random Forest (RF), XGBoost (eXtreme Gradient Boosting) decision tree and ANN (Artificial Neural Network) methods to classify ventilation and non-ventilation samples and to compare their prediction performance for the six tissues. In the model comparison, the AUC (area under receiver operating curve), accuracy, precision, recall, and F1 score were used to evaluate the predictive performance of each model. We then conducted feature analysis per each tissue to detect MV marker genes followed by pathway enrichment analysis for these genes.

RESULTS

XGBoost outperformed the other methods and predicted samples had undergone MV with an average accuracy for the six tissues of 0.951 and average AUC of 0.945. The feature analysis detected a combination of MV marker genes per each tested tissue, some common across several tissues. MV marker genes were mainly related to inflammation and fibrosis as well as cell development and movement regulation. The MV marker genes were significantly enriched in inflammatory and viral pathways.

CONCLUSION

The XGBoost method demonstrated clear enhanced performance and feature analysis compared to the other models. XGBoost was helpful in detecting the tissue-specific marker genes for identifying transcriptomic changes related to MV. Our results show that MV is associated with reduced development and movement in the tissues and higher inflammation and injury not only in direct tissues such as the lungs but also in peripheral tissues and thus should be carefully considered before being implemented.

Multilayer perceptron neural network model development for mechanical ventilator parameters prediction by real time system learning

Background and objective

In pandemic situations like COVID 19, real time monitoring of patient condition and continuous delivery of inspired oxygen can be made possible only through artificial intelligence-based system modeling. Even now manual control of mechanical ventilator parameters is continuing despite the ever-increasing number of patients in critical epidemic conditions. Here a suggestive multi-layer perceptron neural network model is developed to predict the level of inspired oxygen delivered by the mechanical ventilator along with mode and positive end expiratory pressure (PEEP) changes for reducing the effort of health care professionals.

Methods

The artificial neural network model is developed by Python programming using real time data. Parameter identification for model inputs and outputs is done by in corporating consistent real time patient data including periodical arterial blood gas analysis, continuous pulse oximetry readings and mechanical ventilator settings using statistical pairwise analysis using R programming.

Results

Mean square error values and R values of the model are calculated and found to be an average of 0.093 and 0.81 respectively for various data sets. Accuracy loss will be in good fit with validation loss for a comparable number of epochs.

Conclusions

Comparison of the model output is undertaken with physician’s prediction using statistical analysis and shows an accuracy error of 4.11 percentages which is permissible for a good predictive system.

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.

A Wearable System with Embedded Conductive Textiles and an IMU for Unobtrusive Cardio-Respiratory Monitoring

The continuous and simultaneous monitoring of physiological parameters represents a key aspect in clinical environments, remote monitoring and occupational settings. In this regard, respiratory rate (RR) and heart rate (HR) are correlated with several physiological and pathological conditions of the patients/workers, and with environmental stressors. In this work, we present and validate a wearable device for the continuous monitoring of such parameters. The proposed system embeds four conductive sensors located on the user's chest which allow retrieving the breathing activity through their deformation induced during cyclic expansion and contraction of the rib cage. For monitoring HR we used an embedded IMU located on the left side of the chest wall. We compared the proposed device in terms of estimating HR and RR against a reference system in three scenarios: sitting, standing and supine. The proposed system reliably estimated both RR and HR, showing low error averaged along subjects in all scenarios. This is the first study focused on the feasibility assessment of a wearable system based on a multi-sensor configuration (i.e., conductive sensors and IMU) for RR and HR monitoring. The promising results encourage the application of this approach in clinical and occupational settings.

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

BACKGROUND AND OBJECTIVE

Mismatch between invasive mechanical ventilation and the requirements of patients results in patient-ventilator asynchrony (PVA), which is associated with a series of adverse clinical outcomes. Although the efficiency of the available approaches for automatically detecting various types of PVA from the ventilator waveforms is unsatisfactory, the feasibility of powerful deep learning approaches in addressing this problem has not been investigated.

METHODS

We propose a 2-layer long short-term memory (LSTM) network to detect two most frequently encountered types of PVA, namely, double triggering (DT) and ineffective inspiratory effort during expiration (IEE), on two datasets. The performance of the networks is evaluated first using cross-validation on the combined dataset, and then using a cross testing scheme, in which the LSTM networks are established on one dataset and tested on the other.

RESULTS

Compared with the reported rule-based algorithms and the machine learning models, the proposed 2-layer LSTM network exhibits the best overall performance, with the F1 scores of 0.983 and 0.979 for DT and IEE detection, respectively, on the combined dataset. Furthermore, it outperforms the other approaches in cross testing.

CONCLUSIONS

The findings suggest that LSTM is an excellent technique for accurate recognition of PVA in clinics. Such a technique can help detect and correct PVA for a better patient ventilator interaction.

Artificial Intelligence in the Intensive Care Unit

This article is one of ten reviews selected from the Annual Update in Intensive Care and Emergency Medicine 2020. Other selected articles can be found online at https://www.biomedcentral.com/collections/annualupdate2020. Further information about the Annual Update in Intensive Care and Emergency Medicine is available from http://www.springer.com/series/8901.