Suppr超能文献

在体育活动期间使用带有运动伪影去除功能的摄像头增强非接触式心率监测

Enhanced Contactless Heart Rate Monitoring Using Camera with Motion Artifact Removal During Physical Activities.

作者信息

Vatanparvar Korosh, Li Jiyang, Gwak Migyeong, Zhu Li, Kuang Jilong, Gao Alex

出版信息

Annu Int Conf IEEE Eng Med Biol Soc. 2023 Jul;2023:1-5. doi: 10.1109/EMBC40787.2023.10340279.

Abstract

Contactless monitoring of heart rate (HR) can improve passive and continuous tracking of cardiovascular activities and overall people's health. Remote photoplethysmography (rPPG) using a camera eliminates the need for a wearable device. rPPG-based HR has shown promising results to be accurate and comparable to conventional methods such as contact PPG. Most experiments use stationary subjects while motion is known to affect the accuracy of remote PPG. In this paper, a novel methodology is introduced to enhance the accuracy and reliability of HR monitoring based on rPPG in the presence of physical activities like Yoga. This method quickly and accurately tracks HR and analyzes head motion to exclude unreliable data within short windows of rPPG signals. The method was tested with smartphone video data collected from 60 subjects when they are doing activities with varying levels of movement. Results show that our method without motion removal improves the accuracy of the HR readings by 0.7 bpm, reaching 3.57 bpm on average for a 30-sec-window. The accuracy is further improved by another 1.3 bpm after removing the motion artifacts, and reaches 2.29 bpm.Clinical relevance- The enhancement of HR readings from shorter rPPG signal with motion tolerance during physical activities can ultimately help with a more reliable HR tracking of people in uncontrolled settings like home which is a critical step towards remote health-care or wellness tracking.

摘要

心率(HR)的非接触式监测可以改善对心血管活动和整体人群健康的被动式持续跟踪。使用摄像头的远程光电容积脉搏波描记法(rPPG)消除了对可穿戴设备的需求。基于rPPG的心率已显示出有希望的结果,其准确性与诸如接触式PPG等传统方法相当。大多数实验使用静止的受试者,而众所周知运动是会影响远程PPG的准确性。本文介绍了一种新颖的方法,以提高在诸如瑜伽等身体活动情况下基于rPPG的心率监测的准确性和可靠性。该方法能够快速准确地跟踪心率,并分析头部运动以在rPPG信号的短窗口内排除不可靠数据。该方法使用从60名受试者在进行不同运动水平活动时收集的智能手机视频数据进行了测试。结果表明,我们的方法在不进行运动去除的情况下将心率读数的准确性提高了0.7次/分钟,在30秒窗口内平均达到3.57次/分钟。去除运动伪影后,准确性进一步提高了1.3次/分钟,达到2.29次/分钟。临床意义——在身体活动期间,通过对具有运动耐受性的较短rPPG信号增强心率读数,最终有助于在诸如家庭等不受控制的环境中对人们进行更可靠的心率跟踪,这是迈向远程医疗保健或健康跟踪的关键一步。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验