Blood pressure measurement based on photoelectric volume pulse wave
WU Yu dong1, ZHONG Shun cong1,2*, FU Xin bin3
(1.Laboratory of Optics, Terahertz and Non destructive Testing & Evaluation, School of Mechanical Engineering 
and Automation, Fuzhou University, Fuzhou 350108, China;2.Fujian Key Laboratory of Medical 
Instrument and Pharmaceutical Technology, Fuzhou 350000, China;3.Xiamen Special 
Equipment Inspection Institute, Xiamen 361000, China)

Abstract: Aiming at the problem of non invasive continuous measurement of blood pressure by electronic sphygmomanometer, the noninvasive blood pressure measurement method based on photoplethysmography and Stationary wavelet transform (SWT) was proposed. The signal acquisition system was built to acquire the pulse wave signal from the human fingertip. Then the pulse wave signal was decomposed by SWT to reconstruct fifth layer of high frequency signal. Ten characteristic parameters were extracted from the reconstructed signal as the input of ANN and the corresponding blood pressure as the output of ANN to train the model. A total of 10 700 pulse wave signals in different populations were analyzed. 10 000 signals were used to build the model, and the rest 700 signals were used to test the model. The test error was compared with the standards set by the American Association for the Advancement of Medical Devices (AAMI). The results indicate that the method can achieve noninvasive continuous measurement of blood pressure and provide some reference value for the real  time monitoring of blood pressure in intelligent monitoring equipment.

Key words: blood pressure; photoplethysmography signal; stationary wavelet transform; the artificial neural networks