JPS6349132A - Blood pressure value determining method - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field The present invention relates to a method for determining a blood pressure value from a blood pressure value evaluation curve formed as a cuff pressure changes using data points indicating the magnitude of a pulse synchronized wave.

Conventional technology When automatically determining blood pressure values, the sound method detects the so-called Korotkoff sound and determines the blood pressure value based on the change in the volume of the Korotkoff sound, and the cuff pressure that is generated in synchronization with the pulse. Oscillometric method that detects vibrations and determines blood pressure based on changes in the amplitude of pulse waves; detects vibrations in arterial walls using the Doppler effect of ultrasound and measures the amplitude of the vibrations in the arterial walls. Ultrasound methods are used to determine blood pressure values based on changes in blood pressure. In each of the above methods, the blood pressure value formed in response to changes in cuff pressure is determined by data points representing the magnitude of pulse-synchronous waves generated in synchronization with pulses, such as Korotkoff sounds, cuff pressure vibrations, or arterial wall vibrations. Blood pressure values are automatically determined from the evaluation curve. For example, when the blood pressure value evaluation curve exceeds a predetermined dark value,
Alternatively, the cuff pressure value when the difference between the data points is maximum is determined and determined as the systolic blood pressure value or the diastolic blood pressure value.

Problems to be Solved by the Invention By the way, although the conventional method achieves a certain degree of accuracy for smooth evaluation curves, the data point values collected during blood pressure measurement may include variations due to various causes. As this was unavoidable, sufficient blood pressure measurement accuracy could not be obtained in many cases.

Means for Solving the Problems The present invention was made against the background of the above circumstances.
The gist of this is that in the blood pressure evaluation curve formed as the cuff pressure changes, the data points indicate the magnitude of the pulse-synchronized wave that is generated in synchronization with the pulse when a part of the living body is compressed. A method for determining a blood pressure value by determining the rising point of the regression line, comprising (11) a regression line for determining a regression line based on a predetermined number of three or more consecutive data points among the data points; (2) a prediction interval calculation step of calculating a prediction interval based on a plurality of data points constituting the data point group at a position on the regression line where an adjacent data point adjacent to the data point group should exist; and (3) a blood pressure value determining step of determining a blood pressure value based on the fact that the adjacent data point deviates from the predicted interval.

Operation and Effect of the Invention In this manner, a straight line indicating the overall change trend of a predetermined number of adjacent data point groups can be obtained by the regression line, and on this regression line, a straight line can be obtained based on the data point group. The blood pressure value is determined based on the fact that the adjacent data points adjacent to the data point group are outside the predicted interval.

Here, the above prediction interval is one that falls within the expected interval with a high probability when adjacent data points adjacent to a data point group occur with the same variation as the data points that make up the data point group. In some cases, it is statistically considered that other variables other than the same variation as the data points are included. Therefore, the cuff pressure value near the point where the adjacent data point deviates from the predicted interval due to the rise of the blood pressure value evaluation curve accurately corresponds to the blood pressure value.

EXAMPLES Hereinafter, some examples of the present invention will be explained in detail based on the drawings.

In FIG. 1, a pressure sensor 12, a rapid exhaust valve 14, a rate-limiting exhaust valve 16, and an electric pump 18 are connected via piping 20 to a rubber bag-shaped cuff 10 that is wrapped around a living body's arm to compress it. connected. The pressure sensor 12 detects the pressure inside the cuff 10, and the cuff pressure detection circuit 2
The pressure signal SP is supplied to the pulse wave detection circuit 2 and the pulse wave detection circuit 24.

Cuff pressure detection circuit 221J includes a low-pass filter for distinguishing the static pressure inside the cuff 10 from the pressure signal sp, and detects a cuff pressure signal S P representing the static pressure inside the cuff 10 (cuff pressure) Pc.
e is supplied to the CPU 28 via the A/D converter 26.

The pulse wave detection circuit 24 includes a bandpass filter that discriminates the dynamic pressure within the cuff 10, that is, the pressure vibration component (pulse wave) generated in synchronization with the pulse, from the pressure signal SP, and detects the pulse wave representing the pressure vibration. Wave signal (oscillometric signal) SP
o is supplied to the CPU 28 via the A/D converter 30.

C, PU 28 constitutes a control device of the blood pressure measuring device together with ROM 32 and RAM 34, processes input signals according to a program stored in advance in ROM 32 while utilizing the storage function of RAM 34, and controls the rapid exhaust valve 14 and the rate-limiting exhaust valve. 16 and the electric pump 18 via the output interface 36, and causes the blood pressure display 38 to display the blood pressure value.

Hereinafter, the blood pressure measurement operation will be explained according to the flowchart shown in FIG.

First, step S1 is executed to determine whether a start signal has been input. This starting signal is supplied from a starting switch 40 (shown in FIG. 1) that is operated at startup, or periodically from a starting circuit (not shown) prepared to start repeated blood pressure measurements. If the determination in step S1 is negative, the system is put on standby, but if the determination is positive, step S2 is executed, and the rapid exhaust valve 14 and the rate-limiting exhaust valve 16 are closed, and the electric pump 18 is driven. Pressurization of the cuff 10 is started. Then, in step S3, the cuff pressure Pc is a predetermined target cuff pressure P.

It is determined whether or not it has been reached. This target cuff pressure Po is a pressure that is sufficiently higher than the systolic blood pressure value of the subject, for example, 1
It is set to about 80118g. Steps S2 and S3 are repeatedly executed until the cuff pressure Pc reaches the target cuff pressure P1, but once the cuff pressure Pc reaches the target cuff pressure P1, step S4 is executed and the electric pump 18 is stopped, and step S5
is executed and the rate-limiting exhaust valve 16 is opened. This starts lowering the pressure of the cuff 10. The dot in FIG. 3 indicates this state.

During the pressure reduction process of the cuff 10, the flow resistance of the rate-limiting exhaust valve 16 is determined so that the pressure changes at a rate suitable for blood pressure measurement, for example, 3 to 4** IIg/sec.

In the above state, in step S6, the pulse wave signal 5
Every time it is determined that P00 has occurred, step S7 is executed, and the peak value of the pulse wave is stored together with the actual cuff pressure value or the time of occurrence.

This peak value and the actual cuff pressure Pc or the time of occurrence are 1
Configure one data point. Figure 5 shows a large number of data points collected over time or changes in cuff pressure Pc, and a blood pressure value evaluation curve KL made up of them.
An example is shown. Then, in step S8, it is determined whether the cuff pressure Pc has become less than a predetermined constant value 27. This value P7 is a value sufficiently lower than the diastolic blood pressure value of the subject, for example, about 40 mm (g).Initially, the cuff pressure P is larger than the constant value P, so the above step The judgment in S8 is denied, and the peak value of the pulse wave is stored during this time.Therefore, until the cuff pressure Pc reaches a certain value P, l, there are enough data points to determine the blood pressure value. is memorized.

Thereafter, by executing a blood pressure value determination routine in step S9, a blood pressure value is determined based on a blood pressure value evaluation curve composed of data points sequentially stored in step S7, and step 310 is executed to determine the systolic blood pressure. value, diastolic blood pressure value, etc. are displayed on the blood pressure display 38.

In the blood pressure value determination routine, as shown in FIG. 4, a new data point group consisting of three data points is selected by executing step SRI. In this step SRI, for example, a data point group consisting of the three leftmost data points in FIG. After , is determined, the data point group consisting of the rightmost three data points in FIG. 5 is initially selected, and in subsequent cycles, data points shifted one position to the left are selected.

In the following step SR2, as shown in FIG. 6, the three points selected in step SRI, for example, P+, P
A regression line RL of the data point group consisting of z and P3 is obtained according to the following equation (1).

Y=aX+b...(11 However, (112), Y is a variable indicating the magnitude of the pulse wave, and X is a variable indicating time or cuff pressure.

In addition, a is a proportionality constant indicating the slope of the regression line (2
), where b is the regression straight '+M
This is the intercept of RL and is determined by equation (3).

However, xk and y are the X-coordinate position and Y-coordinate position of each data point constituting the data point group.

Moreover, and y are the respective average values of the X coordinate position and Y coordinate position of each data point. Therefore, step SR2 corresponds to the regression line calculation step of this application example.

In the subsequent step SR3, the standard deviation σ of the data point group with respect to the regression line RL determined in step SR2 is determined from the following equation (4), and a prediction interval (PIT) is determined based on this standard deviation σ. In this application example, step SR3 corresponds to a predicted section calculation step.

...(4) ・ ・ ・ ・(5) However, n above is the number of data points composing the data point group, L is known as Student's t, and normal population N(μ, This is a statistic expressed by the following equation (6), which is created from the mean or the square root of the mean square S for a sample of size n from σ1).

Therefore, the predicted interval (PIT) is CB in FIG.

As shown in FIG. 6, adjacent data points (in FIG. 6 P, ) with a statistically high probability, for example, 95% probability.

As described above, the prediction interval, that is, Predicti
When on Interval is determined, step SR
In step 4, it is determined whether the adjacent data point (R4) adjacent to the data point group selected in step SRI is within the predicted interval. Normally, since the adjacent data points are initially within the expected interval, steps SR1 and subsequent steps are executed again.

While repeating the above steps, when the adjacent data point comes to be located at the rising part of the blood pressure value evaluation curve KL, in the step SR4, the adjacent data point adjacent to the new data point group is When it is determined that the vehicle has deviated from the calculated predicted interval (PIT), step SR5 is executed, and it is determined whether the number of times that the vehicle has deviated from the predicted interval (PIT) is three times in a row. Initially, the judgment in step SR5 is negative, so steps SR1 and subsequent steps are executed again, so that the adjacent data points of the data point group selected in this cycle are calculated from the predicted interval (PIT) calculated in the previous cycle. It is determined whether it has come off or not. The predicted interval (PIT) at this time is CB t in Figure 6.
As shown in , it is set above the current cycle's special price.

In step SR5, if it is determined that the adjacent data point following the data point group has deviated from the predicted interval (PIT) three times in a row, step SR
6 is executed to determine the blood pressure value. For example, one of three types of cuff pressure values corresponding to three adjacent data points outside the predicted interval (PIT) is set as the blood pressure value.

Which of these three cuff pressure values is to be determined as the blood pressure value is determined experimentally. The blood pressure value determined in this way is stored as the systolic blood pressure value in the rising portion of the first half of the blood pressure value evaluation curve KL as shown in part A of FIG. In addition, blood pressure value evaluation vA as shown in part B of FIG.
In the rising portion of the latter half of KL, the data points are selected sequentially from the right end of FIG.
The blood pressure value determined in step 6 is stored as the diastolic blood pressure value.

Therefore, in this application example, step SR6 corresponds to the blood pressure value determination step.

As mentioned above, in this application example, the blood pressure value evaluation curve K
A straight line indicating the change trend of the entire data point group consisting of three consecutive points selected from a large number of data points constituting L is obtained by a regression line RL, and on this regression line RL, the above data point group is The systolic blood pressure value and the diastolic blood pressure value are determined based on the fact that the adjacent data points adjacent to the data point group are outside the predicted interval (PIT) obtained based on the prediction interval (PIT). Here, the above prediction interval (PIT) is one that falls within the predicted interval (PIT) with an extremely high probability when adjacent data points adjacent to a data point group occur with the same variation as the data points that make up the data point group. If this is the case, it is statistically considered that other variables other than the same variation as the data point group are included. Therefore, the rising point of the blood pressure value evaluation curve KL is accurately determined when the adjacent data point deviates from the predicted interval, and the cuff pressure value at that point is accurately measured as the blood pressure value.

Next, another application example of the present invention will be explained. In addition, in the following description, the same reference numerals are given to the parts common to the above-mentioned application example, and the description will be omitted.

As shown in FIG. 8, instead of the pulse wave detection circuit 24, microphones 50 and I (sound detection circuit 52 may be provided) for detecting pulse sounds generated from the portion compressed by the cuff 10. The signal output from 50 is supplied to CPU 28 through sound detection circuit 52 and A/D converter 30, which are equipped with a bandpass filter of an appropriate frequency band for discriminating so-called Korotkoff sounds.

In this application example, the flowcharts shown in FIG. 2 and FIG.
is changed to steps 86' and S7' shown in FIG. That is, in step 36', it is determined whether a Korotkoff sound has occurred, and in step 37', the magnitude of the Korotkoff sound is stored together with the cuff pressure at that time or the time of occurrence.

Further, as shown in FIG. 10, a second microphone 56 is added to the blood pressure measuring device of FIG.
Further, a pulse sound detection circuit 58 and an A/
A D converter 60 may also be provided.

In this application example as well, the flowcharts shown in FIG. 2 and FIG.
The steps are changed to steps SPI to SP6 shown in FIG. That is, each time steps SP1 and SF3 are executed and an upstream pulse sound is determined by the second microphone 56 disposed upstream of the artery, the time of occurrence of the upstream pulse sound is stored, and steps SP3 and In SF3, each time a Korotkoff sound occurs, its amplitude value is stored together with the cuff pressure or the time of occurrence. Thereafter, the generation time difference between the upstream pulse sound and the Korotkoff sound is calculated in step SP5, and product data, which is the product of the time difference and the amplitude value of the Korotkoff sound, is calculated in step SP6. As a result, when the cuff pressure becomes smaller than the constant value P7, a blood pressure value evaluation curve consisting of a large number of data points representing the above performance is formed.

Although some examples of the present invention have been described above, the present invention can also be applied to other aspects.

For example, in the above-mentioned application example, after all the data points constituting the blood pressure evaluation curve have been collected, the algorithm for determining the rising point of the blood pressure evaluation curve as shown in Figure 4 is executed. It may be executed every time the sample is collected.

In addition, in step SR2 of the application example described above, the data point group for calculating the regression line is composed of three data points, but it may be composed of four or more data points. The data points forming the group may be increased sequentially.

Furthermore, in the application example described above, an algorithm for determining the rising point of the blood pressure value evaluation curve as shown in FIG. 4 is used to determine the systolic blood pressure value and the diastolic blood pressure value.
It may be used only to determine either the systolic blood pressure value or the diastolic blood pressure value, and a conventional determination algorithm may be used for the other.

It should be noted that the above description is merely a partial example of the present invention, and the present invention can be modified in various ways without departing from its spirit.

[Brief explanation of drawings]

FIG. 1 is a block diagram for explaining the configuration of a blood pressure measuring device to which the present invention is applied. 2 and 4 are flowcharts illustrating the operation of the application example of FIG. 1. FIG. 3 is a diagram showing changes in cuff pressure in the application example of FIG. 1. FIG. 5 is a diagram showing an example of a blood pressure value evaluation curve obtained in the application example of FIG. 1. Figures 6 and 7 are
It is a figure explaining the blood pressure value determination process in the application example of a figure, respectively. FIG. 8 is a diagram corresponding to FIG. 1 in another application example of the present invention, and FIG. 9 is a flowchart showing the main part of the operation of the general example of FIG. FIG. 10 is a diagram corresponding to FIG. 1 in another application example of the present invention, and FIG. 11 is a flowchart showing a main part of the operation of the application example of FIG. 10: Cuff KL: Blood pressure value evaluation curve RL: Regression line Applicant Nippon Corin Co., Ltd. Figure 2 Figure 3 Figure 5 A B Figure 4 Figure 6 Figure 7 Figure 9 Figure 11

Claims (1)

[Claims] In a blood pressure value evaluation curve formed in accordance with a change in cuff pressure using data points indicating the magnitude of a pulse-synchronized wave generated in synchronization with a pulse when a part of a living body is compressed, A method for determining a blood pressure value by determining a rising point, the step comprising: calculating a regression line based on a predetermined number of three or more consecutive data points among the data points; a prediction interval calculation step of calculating a prediction interval based on a plurality of data points constituting the data point group at a position on the regression line where an adjacent data point adjacent to the data point group should exist; A blood pressure value determining method, comprising: determining the blood pressure value based on the fact that the blood pressure value deviates from the expected range.