Development and validation of a machine learning-based prediction model for near-term in-hospital mortality among patients with COVID-19

Abstract

ObjectivesTo develop and validate a model for prediction of near-term in-hospital mortality among patients with COVID-19 by application of a machine learning (ML) algorithm on time-series inpatient data from electronic health records.MethodsA cohort comprised of 567 patients with COVID-19 at a large acute care healthcare system between 10 February 2020 and 7 April 2020 observed until either death or discharge. Random forest (RF) model was developed on randomly drawn 70% of the cohort (training set) and its performance was evaluated on the rest of 30% (the test set). The outcome variable was in-hospital mortality within 20–84 hours from the time of prediction. Input features included patients’ vital signs, laboratory data and ECG results.ResultsPatients had a median age of 60.2 years (IQR 26.2 years); 54.1% were men. In-hospital mortality rate was 17.0% and overall median time to death was 6.5 days (range 1.3–23.0 days). In the test set, the RF classifier yielded a sensitivity of 87.8% (95% CI: 78.2% to 94.3%), specificity of 60.6% (95% CI: 55.2% to 65.8%), accuracy of 65.5% (95% CI: 60.7% to 70.0%), area under the receiver operating characteristic curve of 85.5% (95% CI: 80.8% to 90.2%) and area under the precision recall curve of 64.4% (95% CI: 53.5% to 75.3%).ConclusionsOur ML-based approach can be used to analyse electronic health record data and reliably predict near-term mortality prediction. Using such a model in hospitals could help improve care, thereby better aligning clinical decisions with prognosis in critically ill patients with COVID-19.

Deep Forest Model for Diagnosing COVID-19 From Routine Blood Tests

“…Data of 181 patients was collected from a major hospital in Wuhan, China to train and test the model. Prathamesh et al 42 developed a RF model for the prediction of near-term (20-48 hours) mortality based on time-series inpatient data from electronic health records. The dataset of 567 patients was collected from hospital in New York.…”

“…Bezzan et al 53 selected XGBoost model using Bayesian Optimization among several machine learning models to predict Linear Discriminant Analysis (LDA) [50], [53] AdaBoost [32], [36] Light Gradient Boosting Machine (LGBM) [31], [34], [46] Gradient Boosting Machine (GBM) [31], [44], [45] Naïve Bayes (NB) [9], [10], [27], [42], [50], [53] Decision Trees (DTs) [12], [27], [31], [42], [50], [53] Deep Neural Networks (DNNs) [ANN, CNN, LSTM, RNN]…”

“…[11], [13], [31], [36], [40], [44], [49], [53] K-Nearest Neighbour (KNN) [9], [10], [27], [31], [32], [45], [49], [50], [53] Support Vector Machine (SVM) [9], [10], [27], [31], [32], [36], [44], [45], [49], [50], [53] Extreme Gradient Boosting (XGBoost) [12], [31], [33], [35], [36], [38], [39], [46], [48], [50], [52] Logistic Regression (LR) [9], [10], [12], [27], [32], [33], [34], [36], [38], [42], [43], [44],…”

Predicting Hospitals Hygiene Rate during COVID-19 Pandemic

“…The rate of Hand hygiene compliance almost stands on the method employed for measurement and place of assessment, with differentiation between clinics, intensive care units (ICU) and other places in the medical organisations. Several studies emphasis on that covering medical include its different unites by hand hygiene dispensers is not the ideal way to gain a high level of hygiene compliance [6] [7]. That indicates the importance of using technologies to support decision-makers to allocate hand hygiene dispensers and get the vital data to monitor and manage hygiene systems in an optimal way [8].…”

Implementing automated prognostic models to inform palliative care: more than just the algorithm

“…6 A growing number of automated predictive models leverage vast data in the electronic medical record (EMR) to accurately predict short-term mortality risk in real time and can be paired with systems to prompt clinicians to refer to palliative care. [7][8][9][10][11][12] These models hold great promise to overcome the many clinician-level and system-level barriers to improving access to timely palliative care. First, mortality risk prediction algorithms have been shown to outperform clinician prognostic assessment, and clinicianmachine collaboration may even outperform both.…”