Classification of Rehabilitation Participation in Elderly In-patients with Mild Cognitive Impairments Utilizing Physiological Responses.

Abstract

This study investigated the possibility of utilising physiological responses and machine learning techniques to determine the degree of participation of in-patients with mild cognitive impairment at rehabilitation institutions. Physiological signals related to autonomic functions, cardio-activity, sweat gland activation, and skin surface temperature were obtained, and machine learning classifiers were used to classify rehabilitation participation levels as higher or lower participation when participants were required to perform a VR-based rehabilitative task. Classifiers such as a decision tree or support vector machine can effectively determine two different levels of participation suggesting the proposed approach can help therapists assess an important aspect of client satisfaction.