Machine Learning Models to Predict 30-Day Mortality in Mechanically Ventilated Patients.
Ontology highlight
ABSTRACT:
SUBMITTER: Kim JH
PROVIDER: S-EPMC8157228 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundMechanical ventilation (MV) is vital for critically ill ICU patients but carries significant mortality risks. This study aims to develop a predictive model to estimate hospital mortality among MV patients, utilizing comprehensive health data to assist ICU physicians with early-stage alerts.MethodsWe developed a Machine Learning (ML) framework to predict hospital mortality in ICU patients receiving MV. Using the MIMIC-III database, we identified 25,202 eligible patients through ICD-9 codes. We employed backward elimination and the Lasso method, selecting 32 features based on clinical insights and literature. Data preprocessing included eliminating columns with over 90% missing data and using mean imputation for the remaining missing values. To address class imbalance, we used the Synthetic Minority Over-sampling Technique (SMOTE). We evaluated several ML models, including CatBoost, XGBoost, Decision Tree, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Logistic Regression, using a 70/30 train-test split. The CatBoost model was chosen for its superior performance in terms of accuracy, precision, recall, F1-score, AUROC metrics, and calibration plots.ResultsThe study involved a cohort of 25,202 patients on MV. The CatBoost model attained an AUROC of 0.862, an increase from an initial AUROC of 0.821, which was the best reported in the literature. It also demonstrated an accuracy of 0.789, an F1-score of 0.747, and better calibration, outperforming other models. These improvements are due to systematic feature selection and the robust gradient boosting architecture of CatBoost.ConclusionThe preprocessing methodology significantly reduced the number of relevant features, simplifying computational processes, and identified critical features previously overlooked. Integrating these features and tuning the parameters, our model demonstrated strong generalization to unseen data. This highlights the potential of ML as a crucial tool in ICUs, enhancing resource allocation and providing more personalized interventions for MV patients.
| S-EPMC11373795 | biostudies-literature
Project description:ObjectivesThe aim of this work was to train machine learning models to identify patients at end of life with clinically meaningful diagnostic accuracy, using 30-day mortality in patients discharged from the emergency department (ED) as a proxy.DesignRetrospective, population-based registry study.SettingSwedish health services.Primary and secondary outcome measuresAll cause 30-day mortality.MethodsElectronic health records (EHRs) and administrative data were used to train six supervised machine learning models to predict all-cause mortality within 30 days in patients discharged from EDs in southern Sweden, Europe.ParticipantsThe models were trained using 65 776 ED visits and validated on 55 164 visits from a separate ED to which the models were not exposed during training.ResultsThe outcome occurred in 136 visits (0.21%) in the development set and in 83 visits (0.15%) in the validation set. The model with highest discrimination attained ROC-AUC 0.95 (95% CI 0.93 to 0.96), with sensitivity 0.87 (95% CI 0.80 to 0.93) and specificity 0.86 (0.86 to 0.86) on the validation set.ConclusionsMultiple models displayed excellent discrimination on the validation set and outperformed available indexes for short-term mortality prediction interms of ROC-AUC (by indirect comparison). The practical utility of the models increases as the data they were trained on did not require costly de novo collection but were real-world data generated as a by-product of routine care delivery.
| S-EPMC6701621 | biostudies-literature
Project description:Background: Mechanically ventilated patients in the intensive care unit (ICU) have high mortality rates. There are multiple prediction scores, such as the Simplified Acute Physiology Score II (SAPS II), Oxford Acute Severity of Illness Score (OASIS), and Sequential Organ Failure Assessment (SOFA), widely used in the general ICU population. We aimed to establish prediction scores on mechanically ventilated patients with the combination of these disease severity scores and other features available on the first day of admission. Methods: A retrospective administrative database study from the Medical Information Mart for Intensive Care (MIMIC-III) database was conducted. The exposures of interest consisted of the demographics, pre-ICU comorbidity, ICU diagnosis, disease severity scores, vital signs, and laboratory test results on the first day of ICU admission. Hospital mortality was used as the outcome. We used the machine learning methods of k-nearest neighbors (KNN), logistic regression, bagging, decision tree, random forest, Extreme Gradient Boosting (XGBoost), and neural network for model establishment. A sample of 70% of the cohort was used for the training set; the remaining 30% was applied for testing. Areas under the receiver operating characteristic curves (AUCs) and calibration plots would be constructed for the evaluation and comparison of the models' performance. The significance of the risk factors was identified through models and the top factors were reported. Results: A total of 28,530 subjects were enrolled through the screening of the MIMIC-III database. After data preprocessing, 25,659 adult patients with 66 predictors were included in the model analyses. With the training set, the models of KNN, logistic regression, decision tree, random forest, neural network, bagging, and XGBoost were established and the testing set obtained AUCs of 0.806, 0.818, 0.743, 0.819, 0.780, 0.803, and 0.821, respectively. The calibration curves of all the models, except for the neural network, performed well. The XGBoost model performed best among the seven models. The top five predictors were age, respiratory dysfunction, SAPS II score, maximum hemoglobin, and minimum lactate. Conclusion: The current study indicates that models with the risk of factors on the first day could be successfully established for predicting mortality in ventilated patients. The XGBoost model performs best among the seven machine learning models.
| S-EPMC8280779 | biostudies-literature
Project description:BackgroundCOVID-19 ARDS shares features with non-COVID ARDS but also demonstrates distinct physiological differences. Despite a lack of strong evidence, prone positioning has been advocated as a key therapy for COVID-19 ARDS. The effects of prone position in critically ill patients with COVID-19 are not fully understood, nor is the optimal time of initiation defined. In this nationwide cohort study, we aimed to investigate the association between early initiation of prone position and mortality in mechanically ventilated COVID-19 patients with low oxygenation on ICU admission.MethodsUsing the Swedish Intensive Care Registry (SIR), all Swedish ICU patients ≥ 18 years of age with COVID-19 admitted between March 2020, and April 2021 were identified. A study-population of patients with PaO2/FiO2 ratio ≤ 20 kPa on ICU admission and receiving invasive mechanical ventilation within 24 h from ICU admission was generated. In this study-population, the association between early use of prone position (within 24 h from intubation) and 30-day mortality was estimated using univariate and multivariable logistic regression models.ResultsThe total study cohort included 6350 ICU patients with COVID-19, of whom 46.4% were treated with prone position ventilation. Overall, 30-day mortality was 24.3%. In the study-population of 1714 patients with lower admission oxygenation (PaO2/FiO2 ratio ≤ 20 kPa), the utilization of early prone increased from 8.5% in March 2020 to 48.1% in April 2021. The crude 30-day mortality was 27.2% compared to 30.2% in patients not receiving early prone positioning. We found no significant association between early use of prone positioning and survival.ConclusionsDuring the first three waves of the COVID-19 pandemic, almost half of the patients in Sweden were treated with prone position ventilation. We found no association between early use of prone positioning and survival in patients on mechanical ventilation with severe hypoxemia on ICU admission. To fully elucidate the effect and timing of prone position ventilation in critically ill patients with COVID-19 further studies are desirable.
| S-EPMC9441133 | biostudies-literature
Project description:BackgroundPrevious studies linked a high intensity of ventilation, measured as mechanical power, to mortality in patients suffering from "classic" ARDS. By contrast, mechanically ventilated patients with a diagnosis of COVID-19 may present with intact pulmonary mechanics while undergoing mechanical ventilation for longer periods of time. We investigated whether an association between higher mechanical power and mortality is modified by a diagnosis of COVID-19.MethodsThis retrospective study included critically ill, adult patients who were mechanically ventilated for at least 24 h between March 2020 and December 2021 at a tertiary healthcare facility in Boston, Massachusetts. The primary exposure was median mechanical power during the first 24 h of mechanical ventilation, calculated using a previously validated formula. The primary outcome was 30-day mortality. As co-primary analysis, we investigated whether a diagnosis of COVID-19 modified the primary association. We further investigated the association between mechanical power and days being alive and ventilator free and effect modification of this by a diagnosis of COVID-19. Multivariable logistic regression, effect modification and negative binomial regression analyses adjusted for baseline patient characteristics, severity of disease and in-hospital factors, were applied.Results1,737 mechanically ventilated patients were included, 411 (23.7%) suffered from COVID-19. 509 (29.3%) died within 30 days. The median mechanical power during the first 24 h of ventilation was 19.3 [14.6-24.0] J/min in patients with and 13.2 [10.2-18.0] J/min in patients without COVID-19. A higher mechanical power was associated with 30-day mortality (ORadj 1.26 per 1-SD, 7.1J/min increase; 95% CI 1.09-1.46; p = 0.002). Effect modification and interaction analysis did not support that this association was modified by a diagnosis of COVID-19 (95% CI, 0.81-1.38; p-for-interaction = 0.68). A higher mechanical power was associated with a lower number of days alive and ventilator free until day 28 (IRRadj 0.83 per 7.1 J/min increase; 95% CI 0.75-0.91; p < 0.001, adjusted risk difference - 2.7 days per 7.1J/min increase; 95% CI - 4.1 to - 1.3).ConclusionA higher mechanical power is associated with elevated 30-day mortality. While patients with COVID-19 received mechanical ventilation with higher mechanical power, this association was independent of a concomitant diagnosis of COVID-19.
| S-EPMC10077655 | biostudies-literature
Project description:ObjectivesTo determine the association between mean airway pressure and 90-day mortality in patients with acute respiratory failure requiring mechanical ventilation and to compare the predictive ability of mean airway pressure compared with inspiratory plateau pressure and driving pressure.DesignProspective observational cohort.SettingFive ICUs in Lima, Peru.SubjectsAdults requiring invasive mechanical ventilation via endotracheal tube for acute respiratory failure.InterventionsNone.Measurements and main resultsOf potentially eligible participants (n = 1,500), 65 (4%) were missing baseline mean airway pressure, while 352 (23.5%) were missing baseline plateau pressure and driving pressure. Ultimately, 1,429 participants were included in the analysis with an average age of 59 ± 19 years, 45% female, and a mean PaO2/FIO2 ratio of 248 ± 147 mm Hg at baseline. Overall, 90-day mortality was 50.4%. Median baseline mean airway pressure was 13 cm H2O (interquartile range, 10-16 cm H2O) in participants who died compared to a median mean airway pressure of 12 cm H2O (interquartile range, 10-14 cm H2O) in participants who survived greater than 90 days (p < 0.001). Mean airway pressure was independently associated with 90-day mortality (odds ratio, 1.38 for difference comparing the 75th to the 25th percentile for mean airway pressure; 95% CI, 1.10-1.74) after adjusting for age, sex, baseline Acute Physiology and Chronic Health Evaluation III, baseline PaO2/FIO2 (modeled with restricted cubic spline), baseline positive end-expiratory pressure, baseline tidal volume, and hospital site. In predicting 90-day mortality, baseline mean airway pressure demonstrated similar discriminative ability (adjusted area under the curve = 0.69) and calibration characteristics as baseline plateau pressure and driving pressure.ConclusionsIn a multicenter prospective cohort, baseline mean airway pressure was independently associated with 90-day mortality in mechanically ventilated participants and predicts mortality similarly to plateau pressure and driving pressure. Because mean airway pressure is readily available on all mechanically ventilated patients and all ventilator modes, it is a potentially more useful predictor of mortality in acute respiratory failure.
| S-EPMC8273919 | biostudies-literature
Project description:ObjectivesTo evaluate the contribution of a preextubation chest X-ray (CXR) to identify the risk of extubation failure in mechanically ventilated patients.DesignRetrospective cohort study.SettingsICUs in a tertiary center (the Medical Information Mart for Intensive Care IV database).PatientsPatients greater than or equal to 18 years old who were mechanically ventilated and extubated after a spontaneous breathing trial.InterventionsNone.Measurements and main resultsAmong 1,066 mechanically ventilated patients, 132 patients (12%) experienced extubation failure, defined as reintubation or death within 48 hours of extubation. To predict extubation failure, we developed the following models based on deep learning (EfficientNet) and machine learning (LightGBM) with the training data: 1) model using only the rapid-shallow breathing index (RSBI), 2) model using RSBI and CXR, 3) model using all candidate clinical predictors (i.e., patient demographics, vital signs, laboratory values, and ventilator settings) other than CXR, and 4) model using all candidate clinical predictors with CXR. We compared the predictive abilities between models with the test data to investigate the predictive contribution of CXR. The predictive ability of the model using CXR as well as RSBI was not significantly higher than that of the model using only RSBI (c-statistics, 0.56 vs 0.56; p = 0.95). The predictive ability of the model using clinical predictors with CXR was not significantly higher than that of the model using all clinical predictors other than CXR (c-statistics, 0.71 vs 0.70; p = 0.12). Based on SHapley Additive exPlanations values to interpret the model using all clinical predictors with CXR, CXR was less likely to contribute to the predictive ability than other predictors (e.g., duration of mechanical ventilation, inability to follow commands, and heart rate).ConclusionsAdding CXR to a set of other clinical predictors in our prediction model did not significantly improve the predictive ability of extubation failure in mechanically ventilated patients.
| S-EPMC9191311 | biostudies-literature
Project description:Predicting 30-day procedure-related mortality risk and 30-day unplanned readmission in patients undergoing lower extremity endovascular interventions for peripheral artery disease (PAD) may assist in improving patient outcomes. Risk prediction of 30-day mortality can help clinicians identify treatment plans to reduce the risk of death, and prediction of 30-day unplanned readmission may improve outcomes by identifying patients who may benefit from readmission prevention strategies. The goal of this study is to develop machine learning models to stratify risk of 30-day procedure-related mortality and 30-day unplanned readmission in patients undergoing lower extremity infra-inguinal endovascular interventions. We used a cohort of 14,444 cases from the American College of Surgeons National Surgical Quality Improvement Program database. For each outcome, we developed and evaluated multiple machine learning models, including Support Vector Machines, Multilayer Perceptrons, and Gradient Boosting Machines, and selected a random forest as the best-performing model for both outcomes. Our 30-day procedure-related mortality model achieved an AUC of 0.75 (95% CI: 0.71–0.79) and our 30-day unplanned readmission model achieved an AUC of 0.68 (95% CI: 0.67–0.71). Stratification of the test set by race (white and non-white), sex (male and female), and age (≥65 years and <65 years) and subsequent evaluation of demographic parity by AUC shows that both models perform equally well across race, sex, and age groups. We interpret the model globally and locally using Gini impurity and SHapley Additive exPlanations (SHAP). Using the top five predictors for death and mortality, we demonstrate differences in survival for subgroups stratified by these predictors, which underscores the utility of our model.
| S-EPMC9678279 | biostudies-literature
Project description:Background and Purpose: Hospital readmissions impose a substantial burden on the healthcare system. Reducing readmissions after stroke could lead to improved quality of care especially since stroke is associated with a high rate of readmission. The goal of this study is to enhance our understanding of the predictors of 30-day readmission after ischemic stroke and develop models to identify high-risk individuals for targeted interventions. Methods: We used patient-level data from electronic health records (EHR), five machine learning algorithms (random forest, gradient boosting machine, extreme gradient boosting-XGBoost, support vector machine, and logistic regression-LR), data-driven feature selection strategy, and adaptive sampling to develop 15 models of 30-day readmission after ischemic stroke. We further identified important clinical variables. Results: We included 3,184 patients with ischemic stroke (mean age: 71 ± 13.90 years, men: 51.06%). Among the 61 clinical variables included in the model, the National Institutes of Health Stroke Scale score above 24, insert indwelling urinary catheter, hypercoagulable state, and percutaneous gastrostomy had the highest importance score. The Model's AUC (area under the curve) for predicting 30-day readmission was 0.74 (95%CI: 0.64-0.78) with PPV of 0.43 when the XGBoost algorithm was used with ROSE-sampling. The balance between specificity and sensitivity improved through the sampling strategy. The best sensitivity was achieved with LR when optimized with feature selection and ROSE-sampling (AUC: 0.64, sensitivity: 0.53, specificity: 0.69). Conclusions: Machine learning-based models can be designed to predict 30-day readmission after stroke using structured data from EHR. Among the algorithms analyzed, XGBoost with ROSE-sampling had the best performance in terms of AUC while LR with ROSE-sampling and feature selection had the best sensitivity. Clinical variables highly associated with 30-day readmission could be targeted for personalized interventions. Depending on healthcare systems' resources and criteria, models with optimized performance metrics can be implemented to improve outcomes.
| S-EPMC8044392 | biostudies-literature
Project description:ObjectiveOur study aimed to identify predictors as well as develop machine learning (ML) models to predict the risk of 30-day mortality in patients with sepsis-associated encephalopathy (SAE).Materials and methodsML models were developed and validated based on a public database named Medical Information Mart for Intensive Care (MIMIC)-IV. Models were compared by the area under the curve (AUC), accuracy, sensitivity, specificity, positive and negative predictive values, and Hosmer-Lemeshow good of fit test.ResultsOf 6994 patients in MIMIC-IV included in the final cohort, a total of 1232 (17.62%) patients died following SAE. Recursive feature elimination (RFE) selected 15 variables, including acute physiology score III (APSIII), Glasgow coma score (GCS), sepsis related organ failure assessment (SOFA), Charlson comorbidity index (CCI), red blood cell volume distribution width (RDW), blood urea nitrogen (BUN), age, respiratory rate, PaO2, temperature, lactate, creatinine (CRE), malignant cancer, metastatic solid tumor, and platelet (PLT). The validation cohort demonstrated all ML approaches had higher discriminative ability compared with the bagged trees (BT) model, although the difference was not statistically significant. Furthermore, in terms of the calibration performance, the artificial neural network (NNET), logistic regression (LR), and adapting boosting (Ada) models had a good calibration-namely, a high accuracy of prediction, with P-values of 0.831, 0.119, and 0.129, respectively.ConclusionsThe ML models, as demonstrated by our study, can be used to evaluate the prognosis of SAE patients in the intensive care unit (ICU). Online calculator could facilitate the sharing of predictive models.
| S-EPMC9252033 | biostudies-literature