Measuring Blood Pressure Using a Photoplethysmography Approach ...

3 downloads 198 Views 287KB Size Report
Blood pressure is often measured using a device called a sphygmomanometer, a stethoscope, and a blood pressure cuff. All the existing manual or automatic ...
Measuring Blood Pressure Using a Photoplethysmography Approach M.K. Ali Hassan1, M.Y. Mashor1, N.F. Mohd Nasir2 and S. Mohamed3 1

Mechatronic Engineering Programme,School of Mechatronic Engineering, Universiti Malaysia Perlis, 02600 Jejawi, Arau, Perlis, Malaysia. 2 Biomedical Electronic Engineering Programme,School of Mechatronic Engineering, Universiti Malaysia Perlis, 02600 Jejawi, Arau, Perlis, Malaysia 3 Cardiology Programme,Department of Cardiology, Universiti Sains Malaysia,Kubang Kerian, Kelantan.

Abstract — Blood pressure is often measured using a device called a sphygmomanometer, a stethoscope, and a blood pressure cuff. All the existing manual or automatic measuring techniques of blood pressure are based on this principle, which is not convenient for continuous monitoring of blood pressure. In this paper, we proposed the regression model which could estimate unspecified people’s systolic blood pressure (SBP) conveniently and continuously and checked its accuracy with blood pressure cuff. The method for estimating each individual SBP by using only pulse wave transit time (PWTT) has been studied, but it is difficult to estimate unspecified people’s SBP with the method using only PWTT. This study examines the relationships between arterial blood pressure and certain features of the photoplethysmographic (PPG) signals from 10 healthy subjects. The experiment involved three sessions, which is the resting period, exercise period and recovery period. Keyword — Blood Pressure (BP), Pulse wave transit time (PWTT),Photoplethysmographic (PPG), Electrocardiography (ECG), Non invasive.

I. INTRODUCTION As we know blood pressure is often measured using a sphygmomanometer, a stethoscope, and a blood pressure cuff [1]. The cuff is placed around the upper arm and filled with air. This tightening effect is used to stop the blood from flowing through the brachial artery of the arm. The stethoscope is placed over the artery in front of the elbow and the pressure in the cuff is slowly released. No sound is heard until the cuff pressure falls below the systolic pressure in the artery, at this point, a pulse is heard. As the cuff pressure continuous to fall slowly, the pulse continues, first becoming louder, then dull and muffled. The cuff pressure at the point at which the first sounds are heard, is defined as the systolic blood pressure. The cuff pressure at the point at which the sounds stop, is defined as the diastolic blood pressure. The current study proposes a continuous blood pressure monitoring method based on pulse wave transit time (PWTT). Estimation of blood pressure using pulse wave

transit time PWTT has been studied extensively in the past decade [2-6]. PWTT is the time interval for the arterial pulse pressure wave to travel from the aortic valve to a peripheral site. An acute rise in BP causes vascular tone to increase and the arterial wall become stiffer causing the PWTT to shorten. It is widely accepted that PWTT varies inversely with blood pressure changes and can be used to develop for cuffless and continuous estimation of blood pressure [2]. Usually, time interval between ECG R-wave and the peak of peripheral pulse is selected as the PWTT [79]. Peripheral pulses are recorded at the fingertip using photoplethysmography (PPG) [10]. This study is to estimate systolic blood pressure continuously and noninvasively using a photoplethysmography approach. II. METHODOLOGY The BP device should have a minimum mean of error readout value of 5 mmHg with a standard deviation of error of ±8mmHg. Since SBP is inversely proportional to PWTT, the regression model using only PWTT parameter for estimating each individual SBP was successful. The current studies to estimated unspecified people’s systolic blood pressure (SBP) by using average of systolic blood pressure value from all subjects and create a new regression model for each subject. The procedure of regression analysis is shown Figure 1 The ECG and PPG signals were recorded simultaneously for 30 seconds with AD Instruments system and calculated Pulse Wave Transit Time (PWTT). Figure 2 shows a conventional PWTT definition.

ECG data PPG data SBP

PWTT Regression model

Figure 1: Procedure of regression analysis

N.A. Abu Osman, F. Ibrahim, W.A.B. Wan Abas, H.S. Abd Rahman, H.N. Ting (Eds.): Biomed 2008, Proceedings 21, pp. 591–594, 2008 www.springerlink.com © Springer-Verlag Berlin Heidelberg 2008

592

M.K. Ali Hassan, M.Y. Mashor, N.F. Mohd Nasir and S. Mohamed

SBP(mmHg) vs PWTT(ms) 160

Series1

150

Linear (Series1)

SBP(mmHg)

140 130 120 110 100 150

y = -0.6043x + 256.15 R2 = 0.9576

170

190

210

230

PWTT(ms)

Figure 2: An illustration of the definition

Figure 3: Data of Subject 1

of Pulse Wave Transit Time (PWTT)

SBP(mmHg) vs PWTT(ms) 132 130 128 SBP(mmHg)

Blood pressure was measured by a standard sphygmomanometer. The experimental session involved three sessions: resting period, exercise period and recovery period. The first experiment to measure systolic blood pressure value was conducted using only Pulse Wave Transit Time (PWTT).

126

Series1

124

Linear (Series1)

122 120 118 116 100

1. Combination of regression model using PWTT parameter.

y = -0.1944x + 159.15 R2 = 0.7621 150

200

250

PWTT(m s)

Step 1) The subject was let to relax for 5 minutes.

Figure 4: Data of Subject 2 SBP(mmHg) vs PWTT(ms) 140 Series1

130 SBP(mmHg)

In this study, 10 healthy subjects were involved in this experiment. Both ECG and PPG signal were sampled at 1 kHz and recorded by data acquisition system (AD Instrument System) simultaneously for 45 seconds. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured on subject’s right arm by using sphygmomanometer. The experimental session involved three sessions as mentioned previously.

Linear (Series1)

120 110 100 y = -0.6024x + 233.58 R2 = 0.8646

90

Step 2) Systolic blood pressure (SBP) was measured using sphygmomanometer.

80 150

In this experiment, 60 data points for each subject were recorded. We used 50 data points for making a regression model and 10 data points for validating its accuracy using sphygmomanometer. The first three sample data is shown in Figure 3, 4, 5

_______________________________________________________________

210

230

Figure 5: Data of Subject 3

Step 4) Exercised the subject by jogging about 2 to 3 minutes.

Step 6) Step 2 to 3 was repeated for recovery session.

190 PWTT(ms)

Step 3) ECG and PPG values were measured using AD Instruments and PWTT was calculated.

Step 5) Step 2 and 3 was repeated.

170

III. RESULT In this section, all regression models are combined to get a new regression model for each subject. We used the average of the slope values from all regressions model. A new slope will become reference slope for all regression models. Now, we have derived a new regression model from all the subjects which are shown in Table 1. This model is based on the linear equation, Y=mX + c. We see the accuracy after applying this method to measure systolic blood pressure of Subject 1. Table 2,3,4 shows the new data of Subject 1,2 and 3 after the application

IFMBE Proceedings Vol. 21

_________________________________________________________________

Measuring Blood Pressure Using a Photoplethysmography Approach

Table 1: New Regression model Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10

Old Regression Model y = -0.6043x + 256.15 y = -0.1944x + 159.15 y = -0.6024x + 233.58 y = -0.3375x + 191.91 y =-0.625x + 285 y = -0.24x + 171.22 y = -0.4762x + 237.28 y = -0.2183x + 178.66 y = -0.4641x + 259.66 y = -0.386x + 198.975

New Regression Model y = -0.41482x + 245.2978 y = -0.41482x + 235.6905 y = -0.41482x + 219.2716 y = -0.41482x + 205.8532 y = -0.41482x + 265.523 y = -0.41482x + 171.22 y = -0.41482x + 233.58 y = -0.41482x + 159.15 y = -0.41482x + 256.15 y = -0.41482x + 198.975

593

of the new method. The value of new SBP is closer to the value using sphygmomanometer. Hence, the SBP values for other subjects were measured using this method. Step 1) Measured systolic blood pressure (SBP) using sphygnanometer Step 2) Measured ECG and PPG data and calculated PWTT value. Step 3) Measured slope value based on regression model.

Table 2: The new SBP data of Subject 1 after applying new method PWTT (ms) 290 270 280 250 273 288 285 265 255 259

SBP (mmHg)

New SBP (mmHg)

Different (mmHg)

125 132 130 145 135 127 128 133 140 138

125 133.2964 129.1482 141.5928 132.0519 139.5187 125.8296 127.0741 135.3705 139.5187

0 1.2964 -0.8518 -3.4072 -2.94806 -0.4813 -1.17036 -0.9259 2.3705 -0.4813

Step 4) Generated a new regression model based on new slope value that we got from Step 3. Step 5) Start using this regression model to measure SBP continuously. We had checked the accuracy of the regression model using PWTT with 10 data points which was not included for making the regression. According to Table 4, the regression model using PWTT parameters for estimating unspecified people’s SBP is appropriate and accurate if we compare with sphygmomanometer.

Table 3: The new SBP data of Subject 2 after applying new method PWTT (ms) 205 199 193 193 198 210 205 211 190 191

SBP (mmHg)

New SBP (mmHg)

Different (mmHg)

121 121 121 121 121 121 121 119 121 120

121.0000 123.2442 125.4884 125.4884 123.6183 119.1299 121.0001 118.7559 126.6105 126.2365

0 2.24423 4.48841 4.48841 2.61826 -1.8701 5E-05 -0.24413 5.6105 6.23647

IV. CONCLUSION The model using Pulse Wave Transit Time (PWTT) alone is adequate to estimate SBP for a personal use. However it is not appropriate for the public to be applied on unspecified people. After applying new method based on the average value of slope from regression models, the result has shown that estimating unspecified people’s SBP using this method is appropriate. Although the regression model using PWTT gives acceptable results, it requires more study to make it more accurate and robust.

Table 4: The new SBP data of Subject 3 after applying new method PWTT (ms) 205 206 208 204 206 205 208 210 209 213

SBP (mmHg)

New SBP (mmHg)

Different (mmHg)

108 108 107 108 107 108 107 108 106 107

108.0000 107.626 106.878 108.3741 107.626 108.0001 106.878 106.1299 106.5039 105.0078

0 -0.37398 -0.12204 0.37408 0.62602 5E-05 -0.12204 -1.8701 0.50393 -1.99219

_______________________________________________________________

ACKNOWLEDGMENTS This study was supported by the University Malaysia Perlis Fund, grant no. 9005-00024.

REFERENCES 1. 2.

I. G. Webster, Medical Instrumentation: Application and design, third edition, New York: Wiley, 1998. [M. H. Pollak and P. A. Obrist, “Aortic-radial pulse transit time and ECG Q-wave to radial pulse wave interval as indices at beat by-beat blood pressure change”, Psychophysiology, vol. 20, pp. 21-28, 1983.

IFMBE Proceedings Vol. 21

_________________________________________________________________

594 3. 4. 5.

6.

7.

M.K. Ali Hassan, M.Y. Mashor, N.F. Mohd Nasir and S. Mohamed L. A. Geddes, M. H. Voelz, C. F. Babbs, J. D. Bourland and W. A. Tacker, “Pulse transit time as an indicator of arterial blood pressure”, Psychophysiology, vol. 18, pp. 71-74, 1981 A. A. Robert A. S. John. M.D. Dennis, A. W. Mark and C. B. Taylor, “The covariation of blood pressure and pulse transit time in hypertensive patients”, Psychophysiology, vol. 18, pp. 301- 306, 1981. G. V. Mane, C.R. La, 1. Van lanes and D.W. Johnston, “The relationship between arterial blood pressure and puke transit time during dynamic and static exercise”, Psychophysiology, vol. 21, pp. 521-527, 1984, I. Kerola, V. Konna and R. Sepponen, “Noninvasive blood pressure data acquisition employing pulse transit time detection,” in Proceedings of the 18i’h Annual International Conference o/ the IEEE Medicine and Biology Society, 1996, vol.: 3, pp: 1308 - 1309. Allen J and Murray A 1993 Development of a neural network screening aid for diagnosing lower limb peripheral vascular disease from photoelectric plethysmography pulse waveforms Physiol. Meas. 14 13–22

_______________________________________________________________

8.

K. W. Chan and Y. T. Zhang, “Noninvasive and cuffless measurement of blood pressure for telemedicine,” 2001 Proceeding of the 23 rd Annual EMBS International Conference, Istanbul, Turkey, pp. 3592-3593, 2001 9. A. Ligtenberg and M. Kunt, “ A robust-digital QRS-detection algorithm for arrhythmia monitoring.” Comput. Biomed. Res., vol.16, pp. 273-286,1983 10. Jago J R and Murray A 1988 Repeatability of peripheral pulse measurements on ears, fingers and toes using photoelectric plethysmography Clin. Phys. Physiol. Meas. 9 319–29 Author: Institute: Street: City: Country: Email:

IFMBE Proceedings Vol. 21

Muhamad Khairul Bin Ali Hassan Universiti Malaysia Perlis Kangar, Perlis Malaysia [email protected]

_________________________________________________________________