Development of algorithms for reliable estimation of breathing effort from pulse pleythsmograph signal

使用光传感器获得的光体积图(PPG)波形是心脏跳动的脉动特征。从PPG信号中可以提取出脉率、血氧饱和度(SpO2)等丰富的生理信息，近年来，从这些信号中提取呼吸信号成为研究的热点。尽管人们对从PPG信号中提取呼吸信号感兴趣并进行了研究，但似乎只有一个呼吸参数，即呼吸频率，被普遍估计。这是一个限制，因为其他参数，如呼吸努力，有望提供关于受试者生理的更深入的信息。此外，这一领域的研究还没有转化为可以在家庭睡眠监测等实际临床应用中使用的有形系统。 我们的工业合作伙伴Braebon Medical Corporation是一家总部位于渥太华的公司，专门生产和销售睡眠监测设备，正在寻求开发和测试创新的算法，可以从PPG信号中提取各种呼吸参数。我们渥太华大学的研究小组在开发这一领域的算法方面拥有丰富的经验。在项目的第一阶段，我们开发了一种从PPG信号估计呼吸频率的新算法。在基准数据库上测试时，我们的算法比另一种最先进的算法性能更好。在该项目的第二阶段，由Braebon支持，我们将专注于开发一种算法，以可靠地从PPG信号估计呼吸努力。这项技术将为Braebon提供相对于其他睡眠监测产品的竞争优势，并有望为Braebon在可靠睡眠领域开辟重要的市场机会 监控。

The photoplethysmogram (PPG) waveform obtained using a photo sensor is a pulsatile characterization of the beating of the heart. A rich variety of physiological information such as pulse rate, and oxygen saturation (SpO2) can be extracted from the PPG signal, and recently, researchers have focused on extracting respiratory signals from this signal. Despite interest in and research on the extraction of respiratory signals from PPG signals, it seems that only one respiratory parameter, namely, the respiratory rate, is commonly estimated. This is a limitation, since other parameters such as respiratory effort promise to provide more in-depth information about the physiology of the subject. Moreover, research in this area has not been transformed into tangible systems that can find use in real-word clinical applications like home sleep monitoring. Our industrial collaborator, Braebon Medical Corporation, an Ottawa-based company specializing in the production and sale of sleep monitoring devices, is looking to develop and test innovative algorithms that can extract various respiratory parameters from PPG signals. Our research group at the University of Ottawa has extensive experience in developing algorithms in this area. In the first phase of the project, we developed a novel algorithm for estimating respiratory rate from PPG signals. When tested on a benchmark database, our algorithm performed better than another state-of-the-art algorithm. In the second phase of the project, which is supported by Braebon, we will focus on developing an algorithm to reliably estimate respiratory effort from PPG signals. This technology will provide Braebon a competitive edge over other sleep monitoring products, and is expected to open up significant market opportunities for Braebon in the field of reliable sleep monitoring.