Developing a Smart IoT Solution to Monitor on-Bed Movement Patterns

Abstract

IoT devices and systems become a part of modern living. They are mostly used to monitor daily activities, especially related to personal health and fitness. In fact, it is getting more crucial during the COVID-19 pandemic. In this study, a smart monitoring and alarming IoT system called 'NEF' was modified to recognize on-bed movement patterns including prone position applying different machine learning techniques. On-bed movement patterns were collected from 7 subjects. Considering only prone and supine positions, the models obtained from multilayer perceptron was the best. However, random forest yielded the highest overall correctly classified percentage. Further investigation is likely to include beddings such as pillows and blankets.