A New Physiological Signal Acquisition Patch Using Compressed Sensing Method

As a result of the shortage of medical care resources in an aging society, the development of wearable medical devices is steadily progressing every day. Respiration rate is one of the basic indicators for long-term detection of physiological health. However, the current research and product reports still have some issues in inaccuracy of estimation, reliability of long-term use and so on. In this work, we introduce a physiological signal acquisition patch with 3axis (gyroscope and accelerometer) sensor. It uses compressive sensing and principal component analysis methods to improve estimation accuracy and to reduce more than 60% of the transmitted data for power consumption considerations. With Euler Angle calculation, the respiration rate can be estimated more accurately than by traditional accelerometers and other sensors. We have proven in experiments on 7 subjects that our patch can set the signal compression ratio at 63%$\sim$70% without affecting the final respiration rate estimation accuracy, which corroborates the feasibility and effectiveness of our idea.