MACHINE LEARNING TO IDENTIFY PSYCHOMOTOR BEHAVIORS OF DELIRIUM FOR PATIENTS IN LONG-TERM CARE FACILITY
ABSTARCT :
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, ncluding 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 cross-validation.
EXISTING SYSTEM :
? Our findings are consistent with those existing in the literature except those related to prevalence. It should be noted that other studies have mainly focused on incidence rather than prevalence
? Moreover, our results, following existing literature, shows that to predict correctly delirium is more useful to use predictive models that consider many factors collected routinely
? These methods may also help in small case-series even if they are thought to be useful for large data sets. MLT can identify complex relations between variables and are helpful in structuring predictive models to personalize healthcare assistance.
DISADVANTAGE :
? Patients with a psychiatric illness already diagnosed at admission, with communication problems (such as aphasia, coma status), or with a terminal disease, were excluded from the study.
? An experienced nurse explained the purpose of the study and obtained informed consent from the patient or his next of kin when the participant was unable to give his consent. Study participants were evaluated with the 4AT instrument by three trained nursing students.
? The project was conducted within the framework of thesis preparation of the nursing student and it was approved by the internal offices at the University Hospital of Padova.
PROPOSED SYSTEM :
? Previous studies on predicting delirium proposed the Neelon and Champagne confusion scale (NEECHAM)
? This study aimed to develop accurate and explainable machine learning models for three psychomotor behaviors of delirium for hospitalized adult patients
? In particular, the onset of delirium in elderly patients is associated with increased medical expenses, a high risk of falls and bedsores [10, 11], and a higher mortality rate within one year after discharge than in no-delirium inpatients
ADVANTAGE :
? A random forest (RF) algorithm was used to evaluate the association between the subject’s characteristics and the 4AT score screening tool for delirium.
? Both CAM and Nu-DESC scales were used to compare the ability of random forest (RF) models in predicting the risk of delirium episodes
? Bellelli for the first time, used the 4AT score to assess patients at risk for delirium.
? Machine learning techniques (MLTs) have been widely used in data-driven prediction models, for example in dementia or delirium as well with performances comparable with traditional logistic regression and included into clinical workflow
|