Automatic Nighttime Agitation and Sleep Disruption Detection using Wearable Ankle Device and Machine Learning

Abstract

Nighttime agitation behavior, such as wandering and restlessness during awake and sleep, in people with Alzheimer'sdisease (AD) is expensive to manage and adversely affects sleep. Nighttime agitation is mostly noted by subjective caregiver reports. An automated process for this assessment would improve clinical management. Here, we report on the RestEaZe system that uses an ankle band having a 3-axis accelerometer, a 3-axis gyroscope, and three textile capacitive sensors, along with the application of developed machine learning algorithm to automatically classify sleep status and nighttime agitation behaviours in older adults with AD. We created three binary classifiers- ?IsAwake?, ?IsWandering?, ?IsRestless?, and implemented our model in three phases pre-processing of data, creation of machine learning model and evaluation matrices. Finally, we evaluated our model over various train-test split with 5-fold CV.