1、A Non-invasive Pulse Wave Detection ResearchAbstract: The pulse wave contains many of hemodynamic information, so obtain cardiovascular blood flow parameters by detecting the pulse wave is an important role in examination, treatment and guardianship process of cardiovascular disease. Recently a non-

2、invasive pulse wave detection and analysis system has been developed including software and the hardware. Bi-channel signals can be acquired, stored and shown on the screen dynamically at the same time. Pulse wave can be shown and printed after pulse wave analysis and pulse wave velocity analysis. T

3、his system embraces a computer which is designed for fast data saving, analysis and processing, and a portable data sampling machine which is base on a SCM. Key words: non-invasive pulse wave pulse analysis cardiovascularI. INTRODUCTION The cardiovascular diseases are the most terrible factor of thr

4、eatening the human health. The detecting methods of cardiovascular blood parameters have important clinical significance, and can provide the reliable basis for diagnosing disease, monitoring the curative effect and judging prognosis. How to get the exact and reliable cardiovascular blood flow param

5、eters by using non-invasive and economic detecting methods is the incessantly exploring problem in the biomedical engineering field 1.A mass of clinical test results confirmed that there are close relations between pulse wave characteristics and cardiovascular disease. The wave shape, velocity and r

6、hythm synthetically reflect the physiological and pathological features of human cardiovascular system 2II. PULSE WAVE ANALYSIS The pulse wave contains some physiological and pathological information about the cardiovascular system of the main parts as shown in Figure 1. The practically measured ear

7、 and toe pulse waves from a 20 years old man. Some characteristic point representing the ejected and reflected waves can be utilized to identify pulse transmit time in SBP state.Figure1. Ear and toe pulse wave Some characteristic points are summarized in Table 1. Table1.summary of characteristic poi

8、ntsCharacteristic PointsPhysiological SignificancesPoint AStarting point of ear pulse wavePoint BAortic valves start to closePoint CClosure of aortic valves is just completePoint DThe reflected peak on ear pulse wavePoint aStarting point of toe pulse wavePoint bThe turning point of the rising edge o

9、f the toe pulse wavePoint cThe turning point of the falling edge of the toe pulse wave Clinical studies have shown that, the characteristics of the pulse wave would have a series of regular changes along with the changed of vascular peripheral resistance, vascular wall elasticity and blood viscosity

10、 3-5. The upslope and descending branch become more steep and form high and sharp main wave A when a health young people has a lower vascular peripheral resistance and a better vascular wall elasticity. Tidal wave B is not obvious because the refection wave has a lower velocity, and the blood reflux

11、 has large velocity, and the blood reflux has large intensity to impact the aortic valves, so that dichotic wave crest C and dichotic wave trough is obvious. With the increased peripheral resistance and the deteriorated vascular wall elastically, the wave is gradually increase, the tidal B become mo

12、re obvious, it is relative location is gradually increase and more and more approach even exceed the main wave A. At the same time, the dichotic wave crest C and dichotic wave trough D are gradually mix and difficult to distinguish them, bring about the whole pulse waveform shown as steamed bread. T

13、he position of tidal wave id gradually drive up and near the main wave when the peripheral resistance increases and the arterial elasticity reduces, it is height is an important objective indicator to reflect vascular resistance and vascular wall elasticity in clinical, and it has been fully confirm

14、ed by animal experiments and clinical detections.III. PRINCIPLE OF NONINVASIVE PULSE WAVE DETECTION On the clinical, pulse wave of some surface artery which is near some Cody bone usually is gathered, such as radial pulse wave, and carotid arteries, ankle artery etc. This is because these parts are

15、exposed easily and the structure of the organization is good, therefore it is relatively easy to obtain the precise pulse signal 6. Assume that artery canal is the thin-walled elastic cylindrical tube and the wall thickness can be negligible, compare to the diameter. Friction of sensors, skin, blood

16、, vessels and bone is not existed 7. Pulse pressure sensor exert to blood vessels, which lead to parts of vessel wall to be flat but not close. Beside we suppose that thickness between sensor and vessel wall epidermis can be neglected compared to the vessel diameter, as shown in Figure 2.Figure2. se

17、nsor and artery function schemesThe function relationship of sensor and arterial wall can be expressed by the ideal model in Figure 3. p1 represents the blood pressure of artery, F represents the pressure on the sensor, and the vessel is the approximation for the ideal film. From the analysis of pre

18、ssure on vessels, we can conclude that in this state, the direction of tension on the ideal film run parallel to the sensor surface8-10, therefore it doesnt generate any binding moment. From the figure, we can see that the direction of stress vector F is perpendicular to the direction of the tension

19、 T, so F and T are mutually independent, and the size of F only depend on artery blood pressure PI and the area of flat vessel A:If F is constant and A as a constancy, we can obtain that F is proportional to the p1. In a cardiac cycle, the waveform change with time of p1 is similar to that of F. So

20、we can consider the wave change with time of F as the approximation for p1 11.Figure3. the ideal model of sensor and artery functionIV. INSTRUMENT INSTITUTIONS AND FUNCTION REALIZATIONThis system adopts physical pressure sensor, the hardware parts accomplish amplification, filtering, signal processi

21、ng. The pulse wave signal is transmitted to the computer though the USB after the digital-analog conversion 12. This systems hardware circuit also includes a sensor drive circuit, automatic gain adjustment circuit. The whole system hardware block diagram shown in Figure 4.Figure4. the block diagram

22、of the systems hardware (1) primary differential amplification module The signal that output from sensor is sent to the input of primary amplifier circuit. After the differential circuit, we can get relatively stable output signal13. (2) amplifier and filter moduleBecause the signal is still relativ

23、ely small after the differential amplification circuit, it need further magnify though the follow-up circuit. At the same time, in order to prevent the power line interference and other electromagnetic interferences, the pulse wave signal is also require filtering treatment. Because the signals most

24、 frequency components distribute between 0.115HZ, we use a low-pass filter to eliminate the high frequency noise interferences 14. In order to get better filtering results, we designed a second-order Butterworth low-pass filter that had 40Hz cut-off frequency.Compared with passive RC low-pass filter

25、, Butterworth low-pass filter can effectively filtering noise and has a smaller attenuation.(3) automatic gain adjustment moduleOur experiments confirmed that the finger pulse wave was impacted by the finger temperature, the waveform amplitude increased with the temperature increment. In order to en

26、hance the degree of this equipments automation and improve the measurement accuracy, we designed an automatic gain adjustment circuit. In this paper, we introduced pulse-width modulation technology for automatic gain adjustment15, the illustrative diagram as shown in Figure 5. There, f(t) express th

27、e pulse wave signal, y(t) is the modulation signal, q(t) is chopper signal and is that demodulated pulse wave signal. Single-chop controls the duty cycle of the signal q(t) to play in the analog switch that modulates pulse wave signal f(t) to y(t), then demodulates into by a second-order Butterworth

28、 low-pass filter, and that :f '(t)=d × f(t)Figure5. pulse-width modulation illustrative diagramFirstly, modulate the changes of the pulse waves amplitude to a high-frequency pulse duty cycle using the pulse-width modulation, and then modulate the signal through the second-order low-pass fil

29、ter to filter out the high frequency pulse-width modulation signal. Because the signal is also small, we should magnify it by a fixed gain amplifier. After that, the output signal of the automatic gain adjustment circuit is put into the analog-digital conversion part for sampling, calculate the samp

30、le data and decide whether magnify or narrow the signal, then the signal-chip will adjust the pulse width modulation signals duty cycle. So, achieve the continuous adjustment of the signal amplitude, and the amplitude change related to the duty cycle of the pulse-width modulation signal.The key of s

31、ystem lies in the implement of the algorithm and the corresponding software design. The computer software main task is to complete the data analysis and the parameter display and so on 16. The analysis system includes four parts: information management, data acquisition, inspectional analysis, repor

32、t printing. System module structure is shown in Figure 6.Pulse wave detection and analysis systemInformation managementData acquisitionInspectional analysisReport printing Figure6. the framework of the system(1) data acquisition moduleWe designed the sampling frequency of 1000Hz to collect the finge

33、r tip pulse wave, that is, the sampling interval is set 1ms. In order to display the pulse wave in real-time require the computer be able to provide less than 1ms intervals to regularly query the USB port to read waveform data. Therefore, we designed a high precision timer control in the system soft

34、ware, using multimedia timer control to provide 0.1ms time interval. The timer control sent a time message per 0.1ms to carry out USB port inquiry, waveform data read 17. At the same time of read waveform data, started another thread to display the data. Because A/D converter produced a data every 1

35、ms, however, the timer control sent a message to query the USB port per 0.1ms, we can acquire continuity waveform. The data acquisition module block diagram as shown in Figure 7.Figure 7.the block of the data acquisition and control module(2) inspectional analysis moduleThe data processing module is

36、 started when the system start to collect data. This module includes three parts: the digital processing of pulse wave, extract and identify the pulse wave, calculate the parameters. The second part is the key of this automatic acquisition and analysis system, that is, to achieve automatic and accur

37、ately extract and identify out the single-wave in the circumstance of no human intervention.We adopted threshold method to extract single-wave 18. Through tested and analyzed the characteristics of the pulse wave, we found that motion artifacts often cause the pulse wave amplitude or rise time has a

38、 break. So we set the amplitude and rise time variation range to identify the interference wave and the normal wave, if the waveform amplitude and rise time fall in the set variation range we consider it is a normal wave, otherwise consider interference signal and be removed. Generally, we obtain th

39、e threshold value through the calculus of differences, the pulse wave will highlight the changes information after be difference calculated, thus, it is more easily detect out the single-wave 19. Use this method we should determine three thresholds: difference, pulse wave amplitude and rise time.V.

40、DETECTION PERFORMMANCE VALIDATIONIn order to verify the accuracy of the system, we use this deceive to detect and analyze human pulse signal 20. And the preliminary experiments show that the instrument testing repetition is good. Figure8. distribution of the deviation ratio of the parametersVI. CONC

41、LUSION Non-invasive pulse wave measurement is medical research topic for many years. But the high rate of cardiovascular disease prompts us to create a more convenient and accurate cardiovascular function detecting system. This system combines SMC and computer technology, it has the advantage of SCM

42、s miniaturization and the mass data processing function of computer. It can be used in any computer based on the SCMDAQ and the matching application software. Beside this mode can leave certain expansion space. Due to the computers multiple serial ports, we can switch on sets of data collector and i

43、mplement multiple detections 21. But because of some effect factors, this system also needs to be improved.REFERENCES1L. Zhichang, et al., "Engineering Analysis for Pulse Wave and its Application in Clinical Practice," Beijing: science pulishing company, 2006.2Y. Chen and C. Wen, "A N

44、ew Methodology of Continuous and Noninvasive Blood Pressure Measurement by Pulse Wave Velocity," 11th Int. Conf. Control, Automation, Robotics and Vision, 2010.3Z.B.Zhang and T. H. Wu, "The Techniques of Non-invasive Blood Pressure Measurement and its Development," Chinese Journal of

45、Medical Instrumentation, pp. 196-199, 2003.4Y. Chen, et al., "Continuous and noninvasive blood pressure measurement: a novel modeling ethodology of the relationship between blood pressure and pulse wave velocity," Annals of Biomedical Engineering, vol. 37, 2009.5Z. Xueling, "Several K

46、ey Issues of Noninvasive Detection of Cardiovascular Health State Based on Pulse Wave," Tian Jin University, 2009.6G.Pressman and P.Newgard, "A transducer for the continuous external measurement of arterial blood pressure," IEEE Transactions on Biomedical Engineering, 1963.78H. Inukai

47、 and T. Oka, " Blood pressure measuring apparatus," United States Patent Application Publication, Jan.18 2007.9C. Jianxin and Z. Weizhen, "Boimedical electronics," Peking University Press, 1996.10W. Jicun, "Methods of Noninvasive Continuous Blood Pressure Measurement Based on Pulse Wave," Tian Jin University, Jun.1 2009., et al., "Automatic noninvasive measurement of systoli