No Cuff Blood Pressure Measurement Method Based On Intelligent Mobile Phone

High blood pressure(HBP)is a common chronic cardiovascular disease which is the most risk factor of cardiovascular disease(CVD)as well.Therefore,taking a blood pressure(BP)regularly and efficiently is a necessary approach to discover and deal with it.And furthermore.The sufferings and other related complications of the patient could also be reduced accordingly.Hospitals and families mainly adopt Korotkoff-Sound Method or Oscillography to get BP measured,by which a relative reliable data of arterial BP could be received.Whereas the patient has to buy professional BP devices and will pay much more time in a regular diagnose check.With the wide spreading of smartphones,the research on mobile diagnose check,in which the sensing module of smartphones is associated with a strong calculating power,has wildly been concerned.The introduction of smartphones as a BP device,therefore,reduce the cost of medical devices.As a result,its penetration rate could also be increased.Meanwhile,the convenience of this mobile measurement means a lot to the timely awareness of the patient's healthy risk as well as the prevention works of HBP.By analyzing the sleeveless band measuring method,this paper introduces a promising method for arterial BP estimation,which is to measure the pulse transit time and derive pressure values from it.And this method could be apply to the measurement of BP on smartphones.On this basis,this paper proposes to use a smartphone camera and a microphone to collet the finger pulse wave and heart beat signal respectively.The peak point of the finger pulse wave signal and the extremum of the heart beat signals are used as feature points.The pulse wave propagation time(PTT)could be got By calculating the time difference of the feature points,,and the human blood pressure value is estimated by the linear relationship between the pulse wave propagation time and human BP.There's a difference exist in the study when using different filtering algorithms to process the feature points according to the finger-tip pulse wave signals,which affects the results of BP estimation.Therefore,wavelet filtering,median filtering,and empirical algorithms are used to process the finger-tip pulse wave signals respectively.Accordingly,the advantages and disadvantages of various filtering methods could be seen by comparing the results.At the same time,as the method of estimating the BP for the pulse wave transit time needs to calibrate the characteristic parameters of the human body,a simple individual characteristic value calibration scheme which could be applied to the smartphone is also proposed to improve the measurement accuracy.This paper designs repetitive experiments,exercise experiments and tracking experiments to test the method of using smart phones to estimate human BP.The experimental results show that,when the finger pulse wave signal is processed by method,the experimental error is relatively small,which is better than the other two method.However,this result does not apply with each attempt.The repeatable test and follow-up test results are in good agreement with the Comparison SphygmomanOmeter Measurements.The accuracy is consistent with the AAMI standard and can be used for daily BP estimation.