Non-Contact Respiratory Measurement Using a Depth Camera for Elderly People

Author:

Imano Wakana,Kameyama Kenichi,Hollingdal Malene,Refsgaard Jens,Larsen Knud,Topp Cecilie,Kronborg Sissel Højsted,Gade Josefine Dam,Dinesen BirtheORCID

Abstract

Measuring respiration at home for cardiac patients, a simple method that can detect the patient’s natural respiration, is needed. The purpose of this study was to develop an algorithm for estimating the tidal volume (TV) and respiratory rate (RR) from the depth value of the chest and/or abdomen, which were captured using a depth camera. The data of two different breathing patterns (normal and deep) were acquired from both the depth camera and the spirometer. The experiment was performed under two different clothing conditions (undressed and wearing a T-shirt). Thirty-nine elderly volunteers (male = 14) were enrolled in the experiment. The TV estimation algorithm for each condition was determined by regression analysis using the volume data from the spirometer as the objective variable and the depth motion data from the depth camera as the explanatory variable. The RR estimation was calculated from the peak interval. The mean absolute relative errors of the estimated TV for males were 14.0% under undressed conditions and 10.7% under T-shirt-wearing conditions; meanwhile, the relative errors for females were 14.7% and 15.5%, respectively. The estimation error for the RR was zero out of a total of 206 breaths under undressed conditions and two out of a total of 218 breaths under T-shirt-wearing conditions for males. Concerning females, the error was three out of a total of 329 breaths under undressed conditions and five out of a total of 344 breaths under T-shirt-wearing conditions. The developed algorithm for RR estimation was accurate enough, but the estimated occasionally TV had large errors, especially in deep breathing. The cause of such errors in TV estimation is presumed to be a result of the whole-body motion and inadequate setting of the measurement area.

Publisher

MDPI AG

Subject

Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3