Prinable Joseph, Jones Peter, Thamrin Cindy, McEwan Alistair
Annu Int Conf IEEE Eng Med Biol Soc. 2017 Jul;2017:726-729. doi: 10.1109/EMBC.2017.8036927.
Asthma is a serious public health problem. Continuous monitoring of breathing may offer an alternative way to assess disease status. In this paper we present a novel hardware implementation for the capture and storage of a photoplethysmography (PPG) signal. The LED duty cycle was altered to determine the effect on respiratory rate accuracy. The oximeter was mounted to the left index finger of ten healthy volunteers. The breathing rate derived from the oximeter was validated against a nasal airflow sensor. The duty cycle of a pulse oximeter was changed between 5%, 10% and 25% at a sample rate of 500 Hz. A PPG signal and reference signal was captured for each duty cycle. The PPG signals were post processed in Matlab to derive a respiration rate using an existing Matlab toolbox. At a 25% duty cycle the RMSE was <;2 breaths per minute for the top performing algorithm. The RMSE increased to over 5 breaths per minute when the duty cycle was reduced to 5%. The power consumed by the hardware for a 5%, 10% and 25% duty cycle was 5.4 mW, 7.8 mW, and 15 mW respectively. For clinical assessment of respiratory rate, a RSME of <;2 breaths per minute is recommended. Further work is required to determine utility in asthma management. However for non-clinical applications such as fitness tracking, lower accuracy may be sufficient to allow a reduced duty cycle setting.
哮喘是一个严重的公共卫生问题。持续监测呼吸可能提供一种评估疾病状态的替代方法。在本文中,我们展示了一种用于采集和存储光电容积脉搏波描记法(PPG)信号的新型硬件实现方式。改变发光二极管(LED)的占空比以确定其对呼吸率准确性的影响。将脉搏血氧仪安装在十名健康志愿者的左手食指上。通过鼻气流传感器对源自脉搏血氧仪的呼吸率进行验证。在500Hz的采样率下,将脉搏血氧仪的占空比在5%、10%和25%之间进行切换。针对每个占空比采集一个PPG信号和参考信号。利用现有的Matlab工具箱在Matlab中对PPG信号进行后处理以得出呼吸率。对于表现最佳的算法,在25%占空比时,均方根误差(RMSE)<2次呼吸/分钟。当占空比降至5%时,RMSE增加到超过5次呼吸/分钟。硬件在5%、10%和25%占空比下消耗的功率分别为5.4毫瓦、7.8毫瓦和15毫瓦。对于呼吸率的临床评估,建议RMSE<2次呼吸/分钟。需要进一步开展工作以确定其在哮喘管理中的效用。然而,对于诸如健身追踪等非临床应用,较低的准确性可能足以允许采用较低的占空比设置。