School of Electrical and Electronic Engineering, Nanyang Technological University, 639798 Singapore.
IEEE Trans Biomed Eng. 2013 Jun;60(6):1509-17. doi: 10.1109/TBME.2012.2237401. Epub 2013 Jan 9.
We design a impulse radio ultrawideband radar monitoring system to track the chest wall movement of a human subject during respiration. Multiple sensors are placed at different locations to ensure that the backscattered signal could be detected by at least one sensor no matter which direction the human subject faces. We design a hidden Markov model to infer the subject facing direction and his or her chest movement. We compare the performance of our proposed scheme on 15 human volunteers with the medical gold standard using respiratory inductive plethysmography (RIP) belts, and show that on average, our estimation is over 81% correlated with the measurements of a RIP belt system. Furthermore, in order to automatically differentiate between periods of normal and abnormal breathing patterns, we develop a change point detection algorithm based on perfect simulation techniques to detect changes in the subject's breathing. The feasibility of our proposed system is verified by both the simulation and experiment results.
我们设计了一个脉冲无线电超宽带雷达监测系统，以跟踪人体呼吸时的胸腔运动。多个传感器放置在不同的位置，以确保无论人体面向哪个方向，至少有一个传感器可以检测到背向散射信号。我们设计了一个隐马尔可夫模型来推断受试者的面向方向和胸部运动。我们将我们提出的方案在 15 名志愿者上的性能与使用呼吸感应体积描记法（RIP）带的医学金标准进行了比较，并表明平均而言，我们的估计与 RIP 带系统的测量值相关度超过 81%。此外，为了自动区分正常和异常呼吸模式的时间段，我们开发了一种基于完美仿真技术的变点检测算法来检测受试者呼吸的变化。我们提出的系统的可行性通过仿真和实验结果得到了验证。
