Open Collections

Armstrong, Graham

University of British Columbia

Stroke affects 50,000 Canadians every year and millions of people worldwide. Stroke occurs when the brain tissue is damaged by being deprived of essential compounds from the blood stream. Patients with stroke frequently experience a variety of impairments including prolonged cognitive deficits. Neuropsychological assessment is the most effective way of measuring the nature and magnitude of cognitive impairments. One of the main functions of neuropsychological assessment in a rehabilitation setting is to inform clinical decision-making regarding patient treatment trajectories. Recommendations provided by a neuropsychologist depend on the utility of the neuropsychological battery for predicting the patient’s levels of daily functioning. Daily functioning assesses an individual’s ability to complete tasks of daily life. However, this research has lacked the necessary specificity to capture the comprehensive association between cognition and functional abilities. Further, domain specific score practices, regularly employed by clinicians, can lead to substantial levels of misclassification. The goal of the present study was to evaluate machine learning multiple linear regression relative to ordinary least squares (OLS) on a multivariate level as well as at the level of individual predictors, the clinical utility of a comprehensive neuropsychological battery, lastly the potential impact of an adjunctive clinical decision algorithm in a rehabilitation setting was examined. Data were taken from 167 neuropsychological assessments from patients with stroke who participated in rehabilitation at Kelowna General Hospital. The results indicate that a broad neuropsychological assessment accounts for a significant level of post stroke daily functioning scores. Machine learning is a more powerful tool for identifying individual cognitive predictors of stroke than traditional OLS methods. Machine learning did not provide incremental improvement over OLS methods at a multivariate level. The adjunctive clinical decision-making algorithm did not provide sufficient clinical decision input in this setting. Keywords: Stroke, Neuropsychological Assessment, Functional Outcome, Supervised Machine Learning

Text

2023-02-01

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/83775

Psychology

University of British Columbia

UBCO

http://creativecommons.org/licenses/by-nc-nd/4.0/