Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference Shanghai, China, September 1-4, 2005

Fully Automatic System for Monitoring Blood Pressure from a Toilet-Seat Using the

Volume-Oscillometric Method

Shinobu Tanaka, Masamichi Nogawa, and Ken-ichi Yamakoshi, Member, IEEE

contacting with a toilet-seat to improve this system. Abstract- Daily monitoring of health condition at home is

very important subject not only as an effective scheme for early diagnosis and treatment of cardiovascular and other diseases, but also for prevention and control of such diseases. From this point of view, we have been developing a fully automated "non-conscious" monitoring system for home health care. In this paper are described, structural detail of a newly developed toilet-seat-installed blood pressure measurement system and some results obtained by the system. Also described is outline of a newly designed system for measuring hydrostatic pressure difference between the heart and the measuring site, i.e., thigh, during blood pressure measurement.

11. MATERIALS AND METHOD

A. Outline of the System

Fig. 1 shows the outline of the newly designed BP monitoring system installed in a toilet seat. Principle of BP measurement is based on the volume-oscillometric method previously developed [2]. A measuring site is limited to a portion where the skin contacts with the toilet-seat. Anatomically, there exist small arteries of perforation in the posterior of the thigh, which could therefore be given as the -

I. INTRODUCTION measuring site for the BP measurement. To detect the volume pulsation from the perforation arteries, a reflectance-type Dm to progress of the population in the photo-plethysmogaphic (PG) sensor is used As shown in the

the requirement of home during Upper right pa* of Fig. 1, it consists 6 high-luminance living has been increasingly raised. This health monitoring is near-inkared LEDs (TLN-102, Toshiba Co,, and effective for early diagnosis and treatment of cardiovascular photodiodes (1337-16BR, Hamamatsu Photonics Co., Japan) and other diseases, and for prevention and control of such embedded in a PVC pusher plate of 40 - diameter. The diseases, and for reduction of medical expenses. It would be LEDs are arranged at equal separation to surround the highly desirable if the monitoring could be done in a fully photodiodes placed triangularly on the plate. Applied automated manner without the attachment of any biological sensors to a subject's body as well as any troublesome operations for the measurement. With this concept, we have recently designed a monitoring system with a highly accurate weight measuring device installed in a lavatory floor around a toilet-bowl, which allows the automatic measurement of body weight and weight of evacuation together with the ballistocardiogram as an index of cardiac ejecting function [I]. In recognition of the importance to measure blood pressure (BP) during the use of toilet, we have contemplated introducing BP measurement at the posterior of the thigh

S. Tanaka is with the Division of Biological Measurement and Applications, Institute of Nature and Environmental Technology, Kanazawa University, Kanazawa, 920-8667 Japan (phone: +81-(0)76-234-4736; fax: +81-(0)76-234-4738; e-mail: shinobua t.kanazawa-u.ac.jp).

M. Nogawa and K Yamakoshi are with the Division of Innovative Technology and Science, Graduate School of Natural Science and Technology, Fig. 1 Structural outline of a newly designed automatic blood Kanazawa University, Kanazawa, 920-8667, Japan. pressure monitoring device installed in a toilet seat.

0-7803-8740-6105/$20.00 02005 IEEE.

counter-pressure to the skin is obtained by measuring air pressure in a 20 mm disk chamber (made of polypropylene film) placed on the center of the pusher plate. The plate can be pushed up and down through a hole (45 mm diameter) of the toilet-seat by regulating the current supply to a high torque DC motor which drive a specially designed helicoids type actuator, so that the local compression to the tissue is made by the plate to transmit the pressure to the arteries concerned. The pusher plate and the actuator are set in a space between the toilet-seat and the -bowl

Fig. 2 shows a typical recording of PG signal and pressure (PC) during the period of raising the pusher plate. As shown here, both of a maximum oscillation point (MAP) and a diminishing point (SEP) of PG signal are clearly observed, indicating applicability of the volume-oscillation method to the present system.

In the present system, thigh is used as a measuring site, and therefore, hydrostatic compensation is required to obtain BP values at heart level. Fig. 3 shows the outline of the method newly designed for this purpose. On the sealing of the toilet, a commercially available ultrasonic sensor (OM5-5SD, Ohmic Electronics Co. Ltd., Japan, frequency: 40 kHz, resolution: lcm, measurement range: 0.5-5m) is installed to measure the distance from the sensor to the subject's head (d). As the distance from the sensor to the toilet-seat (L) is hown, the vertical distance between the subject's head and the toilet-seat (4 can be obtained. Using I and h, i.e. the previously measured value of the subject's sitting height, we can estimate the inclination of the trunk (8). From B and the previously measured value ofy, i.e., the distance between heart and thigh in upright-sitting position, we can obtain the vertical distance between heart and thigh (w), and thus, the hydrostatic pressure difference between these two points can be monitored.

Fig. 4 shows an example of hydrostatic pressure recording using the present method. During the measurement, trunk angle was changed from 63 to 29 degree and simultaneously video recorded. As shown in the figure, trunk angle obtained by the present method (&) coincide well with those obtained from VTR (&). Altogether six subjects were involved in this experiments (four males and two females, 21-23 yrs old), and a linear relationship between these two values were obtained (Bu =1.13 9. -2.9, r=0.96) indicating reasonable accuracy of the present method in measuring hydrostatic pressure difference between heart and thigh.

B. Subjects and method

In order to evaluate accuracy of the present system in measuring BP in thigh, simultaneous measurements of wrist BP were carried out using a commercially available BP monitor (HEM-6371T, OMRON Co. Ltd., Japan). Subjects

0 1 0 2 0 3 0

Fig. 2 Typical recording of photoplethysmogram (PG) and cuff pressure (PC) during blood pressure measurement in thigh.

Fig. 3 Outline of a method for measuring hydrostatic pressure difference between heart and thigh.

)4 3 40 subject: male, 22ys

L g 35 E $ 30 G g 25 2 2 20 P

2 I5 m *

50 100 150 200

Time[s]

Fig. 4 Example of recording of hydrostatic pressure change measured by the present method. Square plots indicate those obtained fiom VTR data.

involved were ten young healthy subjects (seven males and three females) aged from 21 to 25 years old. At least, ten times measurement were performed in each subject (max., nineteen times in different trunk angle).

In Table 1 are summarized mean and systolic blood pressure (MBP and SBP) obtained by the present system (Thigh) and those obtained by a commercially available BP monitor (Wrist). For the BP values obtained by the present system, hydrostatic compensation was made. As shown in this table, difference between mean values of "Thigh" and "Wrist" are less than 10 mmHg and the values of S.D. are about 5 mmHg both in MBP and SBP.

Fig. 5 shows the Bland-Altman plots of the results. In the result of SBP (upper fig.), no significant bias was seen, and almost all plots were within +I-1.96 S.D., indicating reasonable accuracy of the present system in measuring SBP. On the other hand, in the result of MBP (lower fig.), slight difference (mean=4.95 mmHg) between these two methods was seen (under-estimation of the present system). This difference may arise from the method for estimating MBP in the commercially available BP monitor. Using this monitor, we could not obtain MBP, but diastolic pressure (DBP), and therefore, we estimated MBP using a conventional equation [MBP = DBP + (SBP-DBP)/3]. This may be one of the causative factors of the difference. To avoid this problem, simultaneous measurement of direct BP or indirect instantaneous BP [3] will be preferable.

IV. CONCLUSION

From the results obtained in this study, it was clearly indicated that the present system is available for non-invasive measurement of BP in thigh, and will be a useful mean for non-conscious physiological monitoring under daily life. Conclusively, this BP measurement technique from the toilet-seat would be available for health monitoring in daily living at home in conjunction with the use of the previously developed toilet monitoring system [I]

This work is partly supported by the Knowledge-based Cluster Creation Project (Ishikawa High-tech Sensing Cluster), Ministry of Education, Culture, Sports, Science and Technology.

Table 1 Results of blood ressure measurements by the resent system using young \ealthy subjects. Data obtained Ey simultaneous measurements using a commercially

available BP monitor (cuff-oscillometnc method) are also shown.

Sub] c t

A

3 e 0 -30

80 90 100 110 120 1% Average [mmHg]

MBP

30 SBP n=122, mean=0.89, S.D.=8.19

20 I . r +1.96SD

3 - a I: mean

8 -10 .* a a. -

- " -;.-- +1.96SD

mean

- * .- - a -20

-1.96SD

h B P inean (S D) high 1 ' ~ ~ r i s t

0 0 ( 7 4 ) 1 82(33)

i? 2 -20

b' -40 I

50 60 70 80 90 100 Average [mmHg]

SBP mean (S D ) Tnlgh 1 mrrlst

110(61) 1 1 1 2 ( 4 1 )

Fig. 5 Bland-Altman plots of BP measured by a commercially available wrist- e BP monitor and the present system (upper; systolic #, lower; mean BP).

.

[2] K Yamakoshi et al, "New oscillometric method for indirect measurement of systolic and mean arterial pressure in the human finger, Part I & 11," in Afed. & Biol. Eng. & Cornput., vol. 20, 1982, pp. 307-3 13 &314-318

[3] S. Tanaka, et al, "Development of an Instrument for Non-invasive Measurement of Instantaneous Blood Pressure in the Radial Artery Based on the Volume-Compensation Method -Including the design and performance of a partially occlusive pad-type cuff and a nozzle-flapper type small electro-pneumatic converter-," in IEEE E17g. Med and Biol. hlagazirze (in press),.

-1.96SD

[I] K Yamakoshi, "Unconstrained physiological monitoring in daily living for health care," in Frontiers Med Biol. Erig., vol. 10,2000, pp. 239-259