Mestha Lalit K, Kyal Survi, Xu Beilei, Lewis Leslie Edward, Kumar Vijay
Annu Int Conf IEEE Eng Med Biol Soc. 2014;2014:3817-20. doi: 10.1109/EMBC.2014.6944455.
We describe a novel method to monitor pulse rate (PR) on a continuous basis of patients in a neonatal intensive care unit (NICU) using videos taken from a high definition (HD) webcam. We describe algorithms that determine PR from videoplethysmographic (VPG) signals extracted from multiple regions of interest (ROI) simultaneously available within the field of view of the camera where cardiac signal is registered. We detect motion from video images and compensate for motion artifacts from each ROI. Preliminary clinical results are presented on 8 neonates each with 30 minutes of uninterrupted video. Comparisons to hospital equipment indicate that the proposed technology can meet medical industry standards and give improved patient comfort and ease of use for practitioners when instrumented with proper hardware.
我们描述了一种使用高清(HD)网络摄像头拍摄的视频,对新生儿重症监护病房(NICU)中的患者进行连续脉搏率(PR)监测的新方法。我们描述了一些算法,这些算法可从视频体积描记法(VPG)信号中确定PR,这些信号是从摄像头视野内同时可用的多个感兴趣区域(ROI)中提取的,其中记录了心脏信号。我们从视频图像中检测运动,并补偿每个ROI的运动伪影。给出了对8名新生儿的初步临床结果,每名新生儿有30分钟不间断视频。与医院设备的比较表明,所提出的技术可以满足医疗行业标准,并且在配备适当硬件时,能为从业者提供更高的患者舒适度和易用性。