Machine Learning to Identify Psychomotor Behaviors of Delirium for Patients in Long-Term Care Facility

Abstract

This study aimed to develop accurate and explainable machine learning models for three psychomotor behaviors of delirium for hospitalized adult patients. A prospective pilot study was conducted with 33 participants admitted to a long-term care facility between August 10 and 25, 2020. During the pilot study, we collected 560 cases that included 33 clinical variables and the survey items from the short confusion assessment method (S-CAM), and developed a mobile-based application. Multiple machine learning algorithms, including four rule-mining algorithms (C4.5, CBA, MCAR, and LEM2) and four other statistical learning algorithms (LR, ANNs, SVMs with three kernel functions, and random forest), were validated by paired Wilcoxon signed-rank tests on both macro-averaged F1 and weighted average F1-measures during the 10-times stratified 2-fold crossvalidation. The LEM2 algorithm achieved the best prediction performance (macro-averaged F1-measure of 49.35%; weighted average F1-measure of 96.55%), correctly identifying adult patients at delirium risk. In the pairwise comparison between predictive powers observed from independent models, the LEM2 model showed a medium or large effect size between 0.4925 and 0.8766 when compared with LR, ANN, SVM with RBF, and MCAR models. We have confirmed that acute consciousness in S-CAM assessment is closely associated with different predictors for screening three psychomotor behaviors of delirium: 1) education level, dementia type or its level, sleep disorder, dehydration, and infection in mixed-type delirium; 2) gender, education level, dementia type, dehydration, bedsores, and foley catheter in hyperactive delirium; and 3) pain, sleep disorder, and haloperidol use in hypoactive delirium.

Navigating the machine learning pipeline: a scoping review of inpatient delirium prediction models

“…The search strategy identified a total of 921 records; after duplicate removal and title and abstract screening, 114 full-text studies were retrieved, of which 3916–54 met the selection criteria for inclusion in the final analysis (figure 2).…”

The development of a web-based app employing machine learning for delirium prevention in long-term care facilities in South Korea

“…Next, the type of delirium (hyperactive, hypoactive, or mixed) is assessed among patients testing positive for delirium in the S-CAM. Written permission was obtained from the developer to use S-CAM in the app [ 35 ]. The S-CAM consists of four steps.…”

“…Post analysis, the LEM2 algorithm showed the best prediction performance (macro-averaged F1-measure of 49.35%; weighted average F1-measure of 96.55%), sharply identifying patients at risk of delirium. Pairwise comparisons between predictive powers were observed in independent models, where the LEM2 model had moderate or large effect sizes, between 0.4925 and 0.8766, when compared to the LR, ANN, SVM with RBF, and MCAR models [ 35 ].…”

“…Iterative design was performed and cloud computing for machine learning to predict the high risk of delirium. The prediction algorithm was validated using the risk factor data of 33 LCF residents and subsequently implemented into the delirium prevention app [ 35 ]. The app was then pilot tested, and its usability and feasibility were analysed.…”

“…The validity of the delirium prediction algorithm was evaluated using explainable machine learning models utilizing data from 33 LCF participants between 10 and 25 August 2020. Machine learning algorithms, such as C4.5, CBA, MCAR, and LEM2, and statistical learning algorithms, such as LR, ANNs, SVMs with three kernel functions, and random forest, were validated by paired Wilcoxon signed-rank tests on both macro -averaged F1 and weighted average F1-measures during the 10-times stratified twofold cross-validation [ 35 ]. Furthermore, for a user-friendly design, five nurses working at an LCF participated in evaluating the appropriateness and given comments of the app.…”