Remote Rehabilitation System Based on the Fusion of Noninvasive Wearable Device and Motion-sensing for Pulmonary Patients

Abstract

Chronic obstructive pulmonary disease is a type of lung disease caused by chronically poor airflow that makes breathing difficult. As a chronic illness it typically worsens over time. Therefore, pulmonary rehabilitation exercise and patient management for extensive period of time are required. However, due to a fast growing number of chronical disease patients, and shortage of primary care practitioners. There is a rising demand for automated rehabilitation exercise detection and monitoring system that capable to deploy in a supervised in-clinic or non-supervised in-home environment. This thesis proposed a method for in home multimodal sensors-based application for patients who have chronic breathing difficulties. The process involved fusion of sensory data—obtained using depth sensor camera, photoplethysmogram signals—that are input variable of a detection and evaluation framework. In addition, we incorporated a set of rehabilitation exercises specific for pulmonary patients into the system by fusing sensory data. Simultaneously, the system also features medical functions that accommodate the needs of medical professionals and those which ease the use of the application for patients, including exercises for tracking progress, patient performance, exercise assignments, and exercise guidance. Finally, the results indicate the accurate determination of pulmonary exercises from the fusion of sensory data. This remote rehabilitation system provides a comfortable and cost-effective option in the health-care rehabilitation system.