JP6656005B2 - Sphygmomanometer - Google Patents

Description

  The present invention relates to a sphygmomanometer.

  Conventionally, there has been known a method of extracting a feature amount from a subject's pulse wave signal, applying the feature amount to a model, and estimating a blood pressure. For example, Patent Document 1 discloses the above blood pressure estimation method.

JP-A-9-220207

  The posture of the subject at the time of acquiring the pulse wave signal affects the pulse wave signal and the feature amount extracted therefrom. Therefore, a model corresponding to one posture may not be able to accurately estimate the blood pressure of a subject in another posture.

  The present invention has been made in view of such a problem, and an object of the present invention is to provide a sphygmomanometer capable of suppressing an error in an estimated value of blood pressure even when a posture of a subject changes.

  The sphygmomanometer of the present invention includes a pulse wave signal acquisition unit that acquires a pulse wave signal of a subject, a feature amount extraction unit that extracts a feature amount from the pulse wave signal acquired by the pulse wave signal acquisition unit, and determines a posture of the subject. A posture determination unit to perform, a storage unit storing a plurality of models that define the relationship between the feature amount and the blood pressure, and a plurality of models stored in the storage unit, which are previously associated with the determination result of the posture determination unit. The apparatus includes a selecting unit for selecting a model, and a blood pressure estimating unit for estimating a subject's blood pressure from the feature amount extracted by the feature amount extracting unit using the model selected by the selecting unit.

  The sphygmomanometer according to the present invention selects a model according to the posture of the subject, and estimates the blood pressure using the selected model. Therefore, according to the sphygmomanometer of the present invention, an error in the estimated value of the blood pressure can be suppressed even if the posture of the subject changes.

FIG. 2 is a block diagram illustrating a configuration of a sphygmomanometer 1. FIG. 3 is a block diagram illustrating a functional configuration of a control unit 3. 5 is a flowchart illustrating an entire process executed by the sphygmomanometer 1. 5 is a flowchart illustrating a posture determination process executed by the sphygmomanometer 1. It is explanatory drawing showing the rising point S and peaks a-f. 5 is a graph showing a correlation between an estimated blood pressure and an actually measured blood pressure. FIG. 2 is a block diagram illustrating a configuration of a sphygmomanometer 1. 5 is a flowchart illustrating a posture determination process executed by the sphygmomanometer 1.

An embodiment of the present disclosure will be described with reference to the drawings.
<First embodiment>
1. Configuration of Sphygmomanometer 1 The configuration of the sphygmomanometer 1 will be described with reference to FIGS. 1 and 2. As shown in FIG. 1, the sphygmomanometer 1 includes a control unit 3, a pulse wave sensor 5, an acceleration sensor 7, an operation unit 9, a display 11, and a communication device 13.

  The control unit 3 mainly includes a known microcomputer including a CPU 15 and a semiconductor memory (hereinafter, referred to as a memory 17) such as a RAM, a ROM, and a flash memory. Various functions of the control unit 3 are realized by the CPU 15 executing a program stored in a non-transitional substantial recording medium. In this example, the memory 17 corresponds to a non-transitional substantial recording medium storing a program. In addition, by executing this program, a method corresponding to the program is executed. The number of microcomputers constituting the control unit 3 may be one or more. The memory 17 corresponds to a storage unit. The memory 17 stores a plurality of models described later.

  As shown in FIG. 2, the control unit 3 includes a pulse wave signal acquisition unit 19, a feature amount extraction unit 21, a posture determination unit 23, and a selection unit. A unit 25, a blood pressure estimation unit 27, a display unit 28, and a communication unit 30 are provided. The method for realizing these elements constituting the control unit 3 is not limited to software, and a part or all of the elements may be realized using hardware in which a logic circuit, an analog circuit, or the like is combined.

  The pulse wave sensor 5 is a known sensor capable of detecting a pulse wave signal from a finger of the subject. The pulse wave sensor 5 includes a light emitting diode (LED) 29 and a photodiode (PD) 31. The light emitting diode 29 emits visible light to the skin of the subject's finger. The wavelength of this light is between 5000 ° and 7000 °. Part of the light emitted by the light emitting diode 29 is reflected in the capillaries of the skin. The photodiode 31 receives the light reflected in the capillaries out of the light emitted by the light emitting diode 29 and extracts the light as an electric signal. The extracted electric signal is a pulse wave signal that fluctuates by reflecting the pulse wave of the subject. Pulse wave sensor 5 sends a pulse wave signal to control unit 3.

  The acceleration sensor 7 can detect the direction and magnitude of acceleration in a three-dimensional space. The acceleration sensor 7 sends a detection signal representing the detected acceleration to the control unit 3. The acceleration sensor 7 is attached to the body of the subject at least when determining the posture of the subject as described later. The acceleration sensor 7 may be constantly attached to the body of the subject, or may be limitedly attached when determining the posture of the subject.

  The operation unit 9 receives a user operation and sends a signal corresponding to the operation to the control unit 3. The user's operation includes an operation to instruct the start of blood pressure measurement, which will be described later. The user may be a subject or another person. Others include, for example, doctors, nurses, and the like.

The display 11 is a display device that can display an image. The communication device 13 is a device that can perform wireless communication with the outside.
Each component of the sphygmomanometer 1 may be housed in one housing, or may be housed separately in two or more housings. The sphygmomanometer 1 may be entirely worn on the body of the subject, or may be partially worn on the body of the subject. As a part to be worn on the body of the subject, for example, the pulse wave sensor 5 and the acceleration sensor 7 or the acceleration sensor 7 can be mentioned.

Examples of the body part of the subject who wears part or all of the sphygmomanometer 1 include a wrist, an arm, a torso, a leg, a neck, and a head.
2. Process Performed by Sphygmomanometer 1 The blood pressure measurement process performed by the sphygmomanometer 1 will be described with reference to FIGS. This blood pressure measurement process is executed when the user performs an operation corresponding to the start of the blood pressure measurement on the operation unit 9. When performing the blood pressure measurement process, the acceleration sensor 7 is attached to the chest of the subject. The subject's finger is in contact with the pulse wave sensor 5.

In step 1 of FIG. 3, the pulse wave signal acquisition unit 19 acquires the subject's pulse wave signal using the pulse wave sensor 5.
In step 2, the feature value extraction unit 21 extracts a feature value from the pulse wave signal acquired in step 1. The feature amounts to be extracted are bValue, dValue, dTime, eTime, and PPI. These features are extracted as follows.

  As shown in FIG. 5, the feature quantity extraction unit 21 creates a first-order differential signal and a second-order differential signal from the pulse wave signal. After the rising point S of the pulse wave signal, the first to sixth peaks in the second derivative signal are a, b, c, d, e, and f, respectively. bValue is the height of the peak of b. dValue is the height of the peak of d. dTime is the time from the rising point S to d. eTime is the time from the rising point S to e. dTime and eTime correspond to a feature amount relating to time. The feature amount relating to time is a feature amount whose unit is time. PPI is the pulse interval.

  In Step 3, the posture determination unit 23 determines the posture of the subject when the pulse wave signal is obtained. This processing will be described with reference to FIG. In step 11, a detection signal is obtained from the acceleration sensor 7. The detection signal includes information indicating in which direction and how much acceleration is applied.

  In step 12, it is determined that the posture of the subject is a posture corresponding to the detection signal acquired in step 11. The memory 17 stores in advance a map that associates the content of the detection signal with the posture of the subject. The posture determination unit 23 determines the posture of the subject by applying the detection signal obtained in step 11 to this map. The posture of the subject includes a sitting position and a supine position.

  Referring back to FIG. 3, in step 4, the selection unit 25 selects a model corresponding to the posture determined in step 3 from among a plurality of models stored in the memory 17 in advance. The memory 17 stores in advance a map that associates the posture of the subject with the model to be selected. The selection unit 25 applies the posture determined in step 3 to this map and selects a model. For example, if it is determined in step 3 that the subject is in the sitting position, the selection unit 25 selects a sitting model. When it is determined in step 3 that the subject is in the supine position, the selection unit 25 selects a supine position model.

  The model stored in the memory 17 is a model in which the feature amount is a displacement variable and the blood pressure is a target variable. When a feature is input to the model, blood pressure can be obtained. The model for the sitting position is the following equation (1). The model for the supine position is represented by the following equation (2).

BP in Expressions (1) and (2) is blood pressure. The sitting model corresponds to the first model. The supine model corresponds to the second model.
Α, β, γ, and δ in Equations (1) and (2) are coefficients. Const is a constant. α, β, γ, δ, and Const can be determined by learning.

  “DTime / PPI” and “eTime / PPI” in the supine model correspond to variables in which time-related features are normalized by pulse wave intervals. The contribution ratio of “dTime / PPI” and “eTime / PPI” to blood pressure is greater in the supine model than in the sitting model.

Here, the contribution rate is a simple contribution rate, and means a rate at which the objective variable is explained by each explanatory variable. The contribution rate is represented by Expression (3). S and S _bj in Expression (3) are represented by Expression (4).

R ² in the equation (3) is a multiple contribution ratio. Equation (3), S in the formula (4) is the sum of the partial regression sum of squares, S _bj is partial regression sum of squares of each explanatory variable.
In step 5, the blood pressure estimation unit 27 estimates the blood pressure of the subject by applying the feature amount extracted in step 2 to the model selected in step 4.

In step 6, the display unit 28 displays the blood pressure estimated in step 5 on the display 11.
In step 7, the communication unit 30 transmits the blood pressure estimated in step 5 to the outside using the communication device 13.

3. Effect of Sphygmomanometer 1 (1A) Sphygmomanometer 1 selects a model according to the posture of the subject, and estimates the blood pressure using the selected model. Therefore, the blood pressure can be accurately estimated regardless of the posture of the subject.

  (1B) The sphygmomanometer 1 can determine that the subject's posture is a sitting position and a supine position. Further, the sphygmomanometer 1 includes a model for sitting and a model for supine. The sphygmomanometer 1 selects the model for the sitting position when it is determined that the subject is in the sitting position, and selects the model for the supine position when it is determined that the subject is in the supine position.

  The model for the supine position has a larger contribution rate to blood pressure than the model for the sitting position, in which the variable in which the characteristic amount relating to time is normalized by the pulse wave interval. Therefore, it is possible to more accurately estimate the blood pressure regardless of whether the subject is in the sitting position or the supine position.

  (1C) The model for the supine position includes “dTime / PPI” and “eTime / PPI”. These are variables in which “dTime” and “eTime” included in the sitting model are normalized by the pulse wave interval. Thus, the blood pressure can be more accurately estimated regardless of the posture of the subject.

  (1D) The sphygmomanometer 1 determines the posture of the subject using the detection signal acquired from the acceleration sensor 7 attached to the body of the subject. Therefore, the user does not have to input the posture of the subject.

4. Test for Confirming Effect of Sphygmomanometer 1 Using the sphygmomanometer 1, blood pressures of a plurality of subjects were estimated. Hereinafter, the blood pressure estimated using the sphygmomanometer 1 is referred to as an estimated blood pressure. The subject includes a subject whose posture is a sitting position and a subject whose posture is a supine position. For the test subject whose posture is sitting, a model for sitting was used. For subjects in a supine position, a model for the supine position was used.

The blood pressure of the same subject was measured using a cuff-type sphygmomanometer. In the following, the blood pressure measured using a cuff-type sphygmomanometer is defined as an actually measured blood pressure.
FIG. 6 shows the correlation between the estimated blood pressure and the actually measured blood pressure. The vertical axis in FIG. 6 is the estimated blood pressure, and the horizontal axis is the measured blood pressure. The triangle markers in FIG. 6 represent subjects whose posture is sitting, and the diamond markers represent subjects whose posture is supine.

  Table 1 shows the correlation coefficient, error average, and error standard deviation between the estimated blood pressure and the measured blood pressure.

As shown in Table 1, the correlation coefficient was large and the error average and error standard deviation were small regardless of the posture of the subject.
As a comparative example, the blood pressure of the subject was estimated by another method using the sphygmomanometer 1. Specifically, a model for sitting was used for a subject whose posture was supine, and a model for sitting was used for a subject whose posture was supine. Table 2 shows a correlation coefficient, an error average, and an error standard deviation between the estimated blood pressure and the actually measured blood pressure in this case.

As shown in Table 2, the error average and the error standard deviation were larger than those in Table 1.
From the above results, it was confirmed that the blood pressure could be accurately estimated by using the blood pressure monitor 1 correctly regardless of the posture of the subject.
<Second embodiment>
1. Differences from the First Embodiment The second embodiment has the same basic configuration as that of the first embodiment. Therefore, the description of the common configuration will be omitted, and differences will be mainly described. The same reference numerals as in the first embodiment denote the same components, and refer to the preceding description.

  As shown in FIG. 7, the sphygmomanometer 1 does not include the acceleration sensor 7. The operation unit 9 includes a sitting position button 35 and a supine position button 37. The sitting position button 35 and the supine position button 37 are button switches that can be operated by the user. When the user turns on the sitting button 35, the memory 17 stores that the subject's posture is sitting. When the user turns on the supine position button 37, the memory 17 stores that the subject is in the supine position.

The operation unit 9 corresponds to an input device. The storage in the memory 17 corresponds to the input result in the input device.
2. Process Performed by Sphygmomanometer 1 The sphygmomanometer 1 performs the process illustrated in FIG. 8 as the posture determination process in step 3. At step 21 in FIG. 8, the posture determination unit 23 reads the posture of the subject from the memory 17. Note that the memory of the subject's posture in the memory 17 is stored by operating the sitting position button 35 or the supine position button 37.

In step 22, the posture determination unit 23 determines that the posture of the subject is the posture read in step 21.
3. Effects of Sphygmomanometer 1 According to the second embodiment described in detail above, the following effects are obtained in addition to the effects (1A) to (1C) of the first embodiment described above.

(2A) The sphygmomanometer 1 determines the posture of the subject using the result of the user operating the sitting position button 35 or the supine position button 37. Thereby, the posture of the subject can be more accurately determined. Further, the configuration of the sphygmomanometer 1 can be simplified.
<Other embodiments>
The embodiments for carrying out the present invention have been described above. However, the present invention is not limited to the above-described embodiments, and can be implemented with various modifications.

  (1) The sphygmomanometer 1 may also determine that the subject's posture is standing by using the acceleration sensor 7 or by using the operation result of the operation unit 9. In this case, the sphygmomanometer 1 can use a standing model. The standing model may be the same model as the sitting model, or may be a different model.

(2) The model may be a model other than those represented by the above formulas (1) and (2). The feature amount included in the model may be another feature amount.
(3) In the above equation (2), a value obtained by multiplying “dTime” by the pulse rate per unit time may be used instead of “dTime / PPI”. In the above equation (2), a value obtained by multiplying “eTime” by the pulse rate per unit time may be used instead of “eTime / PPI”. Since the pulse rate is a value obtained by dividing 60 seconds by the PPI, multiplying the pulse rate has the same meaning as the PPI division.

(3) The method of determining the posture of the subject may be another method. For example, the posture of the subject may be determined by photographing the subject using a camera and analyzing the image of the camera.
(4) A plurality of functions of one component in the above embodiment may be realized by a plurality of components, or one function of one component may be realized by a plurality of components. . Also, a plurality of functions of a plurality of components may be realized by one component, or one function realized by a plurality of components may be realized by one component. Further, a part of the configuration of the above embodiment may be omitted. Further, at least a part of the configuration of the above-described embodiment may be added to or replaced with the configuration of another above-described embodiment. In addition, all aspects included in the technical idea specified only by the language described in the claims are embodiments of the present invention.

  (5) In addition to the above-described sphygmomanometer 1, a system including the sphygmomanometer 1 as a component, a program for causing a computer to function as the control unit 3, a non-transitional actual recording medium such as a semiconductor memory storing the program The present invention can also be implemented in various forms, such as a blood pressure estimation method.

DESCRIPTION OF SYMBOLS 1 ... Sphygmomanometer, 17 ... Memory, 19 ... Pulse wave signal acquisition unit, 21 ... Feature amount extraction unit, 23 ... Posture determination unit, 25 ... Selection unit, 27 ... Blood pressure estimation unit

Claims (4)

A pulse wave signal acquisition unit that acquires a pulse wave signal of the subject,
A feature value extraction unit that extracts a feature value from the pulse wave signal acquired by the pulse wave signal acquisition unit,
A posture determination unit for determining the posture of the subject,
A storage unit storing a plurality of models defining the relationship between the feature amount and the blood pressure,
From the plurality of models stored in the storage unit, a selection unit that selects the model previously associated with the determination result of the posture determination unit,
Using the model selected by the selection unit, a blood pressure estimation unit that estimates the blood pressure of the subject from the feature amount extracted by the feature amount extraction unit,
Equipped with a,
The posture determination unit is configured to be able to determine at least that the posture is a sitting position or a standing position, and that the posture is a supine position,
The plurality of models stored by the storage unit include a first model and a second model in which a contribution ratio of a variable obtained by normalizing the feature amount with respect to time with a pulse wave interval to the blood pressure is larger than that of the first model. The model and include
The selection unit selects the first model when the posture determination unit determines that the posture is sitting or standing, and when the posture determination unit determines that the posture is supine. A sphygmomanometer for selecting the second model . The sphygmomanometer according to claim 1 , wherein
The sphygmomanometer according to claim 2, wherein the second model is a model in which at least a part of the time-related feature amount in the first model is normalized by a pulse wave interval. The blood pressure monitor according to claim 1 or 2 ,
The sphygmomanometer, wherein the posture determination unit determines the posture using a detection signal obtained from an acceleration sensor attached to the body of the subject. The blood pressure monitor according to claim 1 or 2 ,
The sphygmomanometer, wherein the posture determination unit determines the posture using an input result of an input device by which a user inputs the posture.

Images