Method of Estimating Positions for Multiple People in Non-Contact Vital Signs Monitoring Systems

Vital signs monitoring systems have been studied ea-gerly for reducing labor burden in nursing facilities. A microwave sensor can detect vital signs, such as heart rate and respiratory rate, without any contact with the human body. However, the detection area is narrow, and it is difficult to monitor multiple people in the room. In this paper, we propose a new vital signs monitoring system for multiple people in the room of nursing facilities. Also, we present a method of estimating the positions with non-contact using thermal images for the proposed system.

I. Introduction

In recent years, the working population has decreased due to declining birthrate and aging population in Japan [1]. Especially, in a caregiving environment, the number of care recipients per caregiver has increased, and the heavy labor burden becomes a serious problem. It is important to detect sudden changes in the physical conditions of the recipients such as abnormal heart rate and respiratory rate. However, it is quite difficult for the caregivers to monitor the multiple recipients constantly. In general, a visible light camera or a miniaturized wearable terminal with an activity meter is used for constant monitoring. It is reported that the heart rate and respiratory rate are extracted from the camera image by using image processing with machine learning [2]. However, monitoring with a camera has a problem of infringing the recipient's privacy, and it is difficult to obtain the consent of the recipient. The way using wearable terminals also has a problem of giving physical and mental stress to the recipient due to contacting with the body all the time.