Machine learning for predicting successful extubation in patients receiving mechanical ventilation

Development and validation of a clinical prediction model for early ventilator weaning in post-cardiac surgery

Background

Weaning patients from mechanical ventilation is a critical clinical challenge post cardiac surgery. The effective liberation of patients from the ventilator significantly improves their recovery and survival rates. This study aimed to develop and validate a clinical prediction model to evaluate the likelihood of successful extubation in post-cardiac surgery patients.

Method

A predictive nomogram was constructed for extubation success in individual patients, and receiver operating characteristic (ROC) and calibration curves were generated to assess its predictive capability. The superior performance of the model was confirmed using Delong's test in the ROC analysis. A decision curve analysis (DCA) was conducted to evaluate the clinical utility of the nomogram.

Results

Among 270 adults included in our study, 107 (28.84%) experienced delayed extubation. A predictive nomogram system was derived based on five identified risk factors, including the proportion of male patients, EuroSCORE II, operation time, pump time, bleeding during operation, and brain natriuretic peptide (BNP) level. Based on the predictive system, five independent predictors were used to construct a full nomogram. The area under the curve values of the nomogram were 0.880 and 0.753 for the training and validation cohorts, respectively. The DCA and clinical impact curves showed good clinical utility of this model.

Conclusion

Delayed extubation and weaning failure, common and potentially hazardous complications following cardiac surgery, vary in timing based on factors such as sex, EuroSCORE II, pump duration, bleeding, and postoperative BNP reduction. The nomogram developed and validated in this study can accurately predict when extubation should occur in these patients. This tool is vital for assessing risks on an individual basis and making well-informed clinical decisions.

Accuracy of respiratory muscle assessments to predict weaning outcomes: a systematic review and comparative meta-analysis

BACKGROUND

Several bedside assessments are used to evaluate respiratory muscle function and to predict weaning from mechanical ventilation in patients on the intensive care unit. It remains unclear which assessments perform best in predicting weaning success. The primary aim of this systematic review and meta-analysis was to summarize and compare the accuracy of the following assessments to predict weaning success: maximal inspiratory (PImax) and expiratory pressures, diaphragm thickening fraction and excursion (DTF and DE), end-expiratory (Tdiee) and end-inspiratory (Tdiei) diaphragm thickness, airway occlusion pressure (P0.1), electrical activity of respiratory muscles, and volitional and non-volitional assessments of transdiaphragmatic and airway opening pressures.

METHODS

Medline (via Pubmed), EMBASE, Web of Science, Cochrane Library and CINAHL were comprehensively searched from inception to 04/05/2023. Studies including adult mechanically ventilated patients reporting data on predictive accuracy were included. Hierarchical summary receiver operating characteristic (HSROC) models were used to estimate the SROC curves of each assessment method. Meta-regression was used to compare SROC curves. Sensitivity analyses were conducted by excluding studies with high risk of bias, as assessed with QUADAS-2. Direct comparisons were performed using studies comparing each pair of assessments within the same sample of patients.

RESULTS

Ninety-four studies were identified of which 88 studies (n = 6296) reporting on either PImax, DTF, DE, Tdiee, Tdiei and P0.1 were included in the meta-analyses. The sensitivity to predict weaning success was 63% (95% CI 47-77%) for PImax, 75% (95% CI 67-82%) for DE, 77% (95% CI 61-87%) for DTF, 74% (95% CI 40-93%) for P0.1, 69% (95% CI 13-97%) for Tdiei, 37% (95% CI 13-70%) for Tdiee, at fixed 80% specificity. Accuracy of DE and DTF to predict weaning success was significantly higher when compared to PImax (p = 0.04 and p < 0.01, respectively). Sensitivity and direct comparisons analyses showed that the accuracy of DTF to predict weaning success was significantly higher when compared to DE (p < 0.01).

CONCLUSIONS

DTF and DE are superior to PImax and DTF seems to have the highest accuracy among all included respiratory muscle assessments for predicting weaning success. Further studies aiming at identifying the optimal threshold of DTF to predict weaning success are warranted.

TRIAL REGISTRATION

PROSPERO CRD42020209295, October 15, 2020.

Predicting Hypoxia Using Machine Learning: Systematic Review

Background

Hypoxia is an important risk factor and indicator for the declining health of inpatients. Predicting future hypoxic events using machine learning is a prospective area of study to facilitate time-critical interventions to counter patient health deterioration.

Objective

This systematic review aims to summarize and compare previous efforts to predict hypoxic events in the hospital setting using machine learning with respect to their methodology, predictive performance, and assessed population.

Methods

A systematic literature search was performed using Web of Science, Ovid with Embase and MEDLINE, and Google Scholar. Studies that investigated hypoxia or hypoxemia of hospitalized patients using machine learning models were considered. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool.

Results

After screening, a total of 12 papers were eligible for analysis, from which 32 models were extracted. The included studies showed a variety of population, methodology, and outcome definition. Comparability was further limited due to unclear or high risk of bias for most studies (10/12, 83%). The overall predictive performance ranged from moderate to high. Based on classification metrics, deep learning models performed similar to or outperformed conventional machine learning models within the same studies. Models using only prior peripheral oxygen saturation as a clinical variable showed better performance than models based on multiple variables, with most of these studies (2/3, 67%) using a long short-term memory algorithm.

Conclusions

Machine learning models provide the potential to accurately predict the occurrence of hypoxic events based on retrospective data. The heterogeneity of the studies and limited generalizability of their results highlight the need for further validation studies to assess their predictive performance.

Multitask learning to predict successful weaning in critically ill ventilated patients: A retrospective analysis of the MIMIC-IV database

Objective

Weaning is an essential issue in critical care. This study explores the efficacy of multitask learning models in predicting successful weaning in critically ill ventilated patients using the Medical Information Mart for Intensive Care (MIMIC) IV database.

Methods

We employed a multitask learning framework with a shared bottom network to facilitate common knowledge extraction across all tasks. We used the Shapley additive explanations (SHAP) plot and partial dependence plot (PDP) for model explainability. Furthermore, we conducted an error analysis to assess the strength and limitation of the model. Area under receiver operating characteristic curve (AUROC), calibration plot and decision curve analysis were used to determine the performance of the model.

Results

A total of 7758 critically ill patients were included in the analyses, and 78.5% of them were successfully weaned. Multitask learning combined with spontaneous breath trial achieved a higher performance to predict successful weaning compared with multitask learning combined with shock and mortality (area under receiver operating characteristic curve, AUROC, 0.820 ± 0.002 vs 0.817 ± 0.001, p < 0.001). We assessed the performance of the model using calibration and decision curve analyses and further interpreted the model through SHAP and PDP plots. The error analysis identified a relatively high error rate among those with low disease severities, including low mean airway pressure and high enteral feeding.

Conclusion

We demonstrated that multitask machine learning increased predictive accuracy for successful weaning through combining tasks with a high inter-task relationship. The model explainability and error analysis should enhance trust in the model.

Machine Learning Algorithms Predict Successful Weaning From Mechanical Ventilation Before Intubation: Retrospective Analysis From the Medical Information Mart for Intensive Care IV Database

BACKGROUND

The prediction of successful weaning from mechanical ventilation (MV) in advance of intubation can facilitate discussions regarding end-of-life care before unnecessary intubation.

OBJECTIVE

We aimed to develop a machine learning-based model that predicts successful weaning from ventilator support based on routine clinical and laboratory data taken before or immediately after intubation.

METHODS

We used the Medical Information Mart for Intensive Care IV database, which is an open-access database covering 524,740 admissions of 382,278 patients in Beth Israel Deaconess Medical Center, United States, from 2008 to 2019. We selected adult patients who underwent MV in the intensive care unit (ICU). Clinical and laboratory variables that are considered relevant to the prognosis of the patient in the ICU were selected. Data collected before or within 24 hours of intubation were used to develop machine learning models that predict the probability of successful weaning within 14 days of ventilator support. Developed models were integrated into an ensemble model. Performance metrics were calculated by 5-fold cross-validation for each model, and a permutation feature importance and Shapley additive explanations analysis was conducted to better understand the impacts of individual variables on outcome prediction.

RESULTS

Of the 23,242 patients, 19,025 (81.9%) patients were successfully weaned from MV within 14 days. Using the preselected 46 clinical and laboratory variables, the area under the receiver operating characteristic curve of CatBoost classifier, random forest classifier, and regularized logistic regression classifier models were 0.860 (95% CI 0.852-0.868), 0.855 (95% CI 0.848-0.863), and 0.823 (95% CI 0.813-0.832), respectively. Using the ensemble voting classifier using the 3 models above, the final model revealed the area under the receiver operating characteristic curve of 0.861 (95% CI 0.853-0.869), which was significantly better than that of Simplified Acute Physiology Score II (0.749, 95% CI 0.742-0.756) and Sequential Organ Failure Assessment (0.588, 95% CI 0.566-0.609). The top features included lactate and anion gap. The model's performance achieved a plateau with approximately the top 21 variables.

CONCLUSIONS

We developed machine learning algorithms that can predict successful weaning from MV in advance to intubation in the ICU. Our models can aid the appropriate management for patients who hesitate to decide on ventilator support or meaningless end-of-life care.

Developing a machine-learning model for real-time prediction of successful extubation in mechanically ventilated patients using time-series ventilator-derived parameters

Background

Successful weaning from mechanical ventilation is important for patients admitted to intensive care units. However, models for predicting real-time weaning outcomes remain inadequate. Therefore, this study aimed to develop a machine-learning model for predicting successful extubation only using time-series ventilator-derived parameters with good accuracy.

Methods

Patients with mechanical ventilation admitted to the Yuanlin Christian Hospital in Taiwan between August 2015 and November 2020 were retrospectively included. A dataset with ventilator-derived parameters was obtained before extubation. Recursive feature elimination was applied to select the most important features. Machine-learning models of logistic regression, random forest (RF), and support vector machine were adopted to predict extubation outcomes. In addition, the synthetic minority oversampling technique (SMOTE) was employed to address the data imbalance problem. The area under the receiver operating characteristic (AUC), F1 score, and accuracy, along with the 10-fold cross-validation, were used to evaluate prediction performance.

Results

In this study, 233 patients were included, of whom 28 (12.0%) failed extubation. The six ventilatory variables per 180 s dataset had optimal feature importance. RF exhibited better performance than the others, with an AUC value of 0.976 (95% confidence interval [CI], 0.975-0.976), accuracy of 94.0% (95% CI, 93.8-94.3%), and an F1 score of 95.8% (95% CI, 95.7-96.0%). The difference in performance between the RF and the original and SMOTE datasets was small.

Conclusion

The RF model demonstrated a good performance in predicting successful extubation in mechanically ventilated patients. This algorithm made a precise real-time extubation outcome prediction for patients at different time points.