Exploiting Machine Learning Algorithms and Methods for the Prediction of Agitated Delirium After Cardiac Surgery: Models Development and Validation Study

Abstract

BackgroundDelirium is a temporary mental disorder that occasionally affects patients undergoing surgery, especially cardiac surgery. It is strongly associated with major adverse events, which in turn leads to increased cost and poor outcomes (eg, need for nursing home due to cognitive impairment, stroke, and death). The ability to foresee patients at risk of delirium will guide the timely initiation of multimodal preventive interventions, which will aid in reducing the burden and negative consequences associated with delirium. Several studies have focused on the prediction of delirium. However, the number of studies in cardiac surgical patients that have used machine learning methods is very limited.ObjectiveThis study aimed to explore the application of several machine learning predictive models that can pre-emptively predict delirium in patients undergoing cardiac surgery and compare their performance.MethodsWe investigated a number of machine learning methods to develop models that can predict delirium after cardiac surgery. A clinical dataset comprising over 5000 actual patients who underwent cardiac surgery in a single center was used to develop the models using logistic regression, artificial neural networks (ANN), support vector machines (SVM), Bayesian belief networks (BBN), naïve Bayesian, random forest, and decision trees.ResultsOnly 507 out of 5584 patients (11.4%) developed delirium. We addressed the underlying class imbalance, using random undersampling, in the training dataset. The final prediction performance was validated on a separate test dataset. Owing to the target class imbalance, several measures were used to evaluate algorithm’s performance for the delirium class on the test dataset. Out of the selected algorithms, the SVM algorithm had the best F1 score for positive cases, kappa, and positive predictive value (40.2%, 29.3%, and 29.7%, respectively) with a P=.01, .03, .02, respectively. The ANN had the best receiver-operator area-under the curve (78.2%; P=.03). The BBN had the best precision-recall area-under the curve for detecting positive cases (30.4%; P=.03).ConclusionsAlthough delirium is inherently complex, preventive measures to mitigate its negative effect can be applied proactively if patients at risk are prospectively identified. Our results highlight 2 important points: (1) addressing class imbalance on the training dataset will augment machine learning model’s performance in identifying patients likely to develop postoperative delirium, and (2) as the prediction of postoperative delirium is difficult because it is multifactorial and has complex pathophysiology, applying machine learning methods (complex or simple) may improve the prediction by revealing hidden patterns, which will lead to cost reduction by prevention of complications and will optimize patients’ outcomes.

Health informatics publication trends in Saudi Arabia: a bibliometric analysis over the last twenty-four years

“…The study location was mostly in Saudi Arabia (57.5%, n=140). Other study locations included the United States (1.9%, n=3) [22][23][24], Malaysia (1.9%, n=3) [25][26][27], and Canada (0.8%, n=2) [28,29]. The majority of publications (35.1%, n=85) used quantitative methods, and the minority of publications (9.1%, n=22) used mixed methods.…”

“…For publications in 2000-2009, data sources used were surveys or questionnaire data [30], patient or medical data [31][32][33], clinical or health care research datasets [34], and patient or disease registries [35]. For publications in 2010-2019, more data sources were used, including surveys or questionnaire data [12,24,26,, interviews or focus groups [25,27,28,37,69,94,, patient or medical data [23,, clinical or health care research datasets [23,[143][144][145][146][147][148][149][150][151][152][153], patient or disease registries [29,67,154,155], and social media (Facebook [156,157], Twitter [158,159], Quora [22], and WhatsApp [160]) and new social media datasets [161].…”

Dynamic Predictive Scores for Cardiac Surgery-Associated Agitated Delirium: a single-center retrospective observational study

“…Similar to previous investigations 2 , we found that postoperative delirium was associated with negative outcomes. Several authors agree that early detection could prevent or assist in the treatment of delirium 15 . Some of the proposed preventive interventions have been shown to have bene cial effects on delirium, such as early mobilization, early extubation and minimization of alarm use and noisiness.…”

A novel model to label delirium in an intensive care unit from clinician actions

“…We compared performance of five binary ML classifiers [ 16 , 17 , 19 , 22 ], including logistic regression (stats R-package), Classification and Regression Trees (CART; rpart R-package) [ 44 , 45 ], supervised random forests (randomForest) [ 46 , 47 ], naïve Bayes (e1071) [ 48 , 49 ], and support vector machines (SVM; e1071) [ 49 , 50 ]. (Additional file 1 : File A.1) The logistic regression model underwent refinement and feature selection by stepwise forwards and backwards selection, L1/LASSO (Least Absolute Shrinkage and Selection Operator) penalization [ 51 , 52 ], L2/Ridge penalization [ 53 ], and combined L1-L2 penalization (penalized).…”

“…The preferred measures in clinical practice for labeling delirium are nurse- or provider-administered, validated screening tools, including the CAM-ICU[ 13 , 23 ] and the Intensive Care Delirium Screening Checklist (ICDSC) [ 13 , 28 , 29 ]. CAM-ICU administered during treatment is a mainstay label of delirium in the ML research setting [ 14 – 19 ]. However, variations in institutional practice and physician buy-in can lead to inconsistent use of the CAM or ICDSC in the clinical setting [ 13 ].…”