KR101820511B1 - Apparatus and method for estimating blood pressure using biosignal based on activity model - Google Patents

Abstract

According to an embodiment of the present invention, a blood pressure estimating device includes: a measurement part measuring a preset kind of biosignal from a user; a state determining part determining which one of preset state types the users body state corresponds to the measured biosignal; a database storing corresponding relation information for making each of the state types correspond to at least one of preset blood pressure estimation functions for estimating the blood pressure of the user from the measured biosignal; and a blood pressure estimating part estimating the blood pressure of the user by using a blood pressure estimation function corresponding to a state type corresponding to the body state of the user based on the corresponding relation information. As such, the present invention is capable of estimating the blood pressure of the user more accurately.

Description

BACKGROUND OF THE INVENTION Field of the Invention [0001] The present invention relates to an activity model-based blood pressure estimation apparatus and method using bio-

The present invention relates to an activity-model-based blood pressure estimation apparatus and method for estimating a blood pressure of a user on the basis of a bio-signal measured by a user and estimating the blood pressure more accurately by reflecting the user's physical condition .

In living things including humans, a variety of electrical signals, which can be referred to as biosignals, are generated through the principles of electrical, mechanical, optical and the like. Examples of such bio-signals are electrocardiogram (ECG), ballistocardiogram (BCG), photoplethysmogram (PPG), etc. By analyzing these bio-signals, .

Among the information that can be estimated using bio-signals is the blood pressure of living organisms. Blood pressure is one of the important vital signs of blood pressure along the walls of blood vessels. High blood pressure, or abnormally low blood pressure, that is abnormally high in blood pressure is not only a disease that needs to be managed by itself, but also acts as a cause or risk factor for a variety of other diseases, so accurate observation of blood pressure is very important for maintaining health can do.

The technique of detecting the blood pressure using the bio-signals such as the electrocardiogram, the cardiac trough, and the optical pulse waveform is advantageous in that it can observe the change of the blood pressure easily while observing daily life while comparing with the existing blood pressure measuring method. Be in the spotlight. However, according to this technique, since a single algorithm is applied to estimate the blood pressure, there is a problem that the accuracy is not uniform depending on the physical condition of the user, such as the type of physical activity or the use of various drugs.

Korean Patent Publication No. 10-2010-0116880 (published on November 2, 2010)

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus and method for accurately estimating a blood pressure of a user irrespective of a physical condition of a user, such as a type of physical activity or a variety of drugs.

A blood pressure estimation apparatus according to an embodiment of the present invention includes a measurement unit that measures a predetermined type of living body signal from a user, a measurement unit that estimates a living body state of the user based on the measured living body signal, A database for storing correspondence relationship information in which each of the state types is associated with at least one of a plurality of predetermined blood pressure estimation functions for estimating a blood pressure of the user from the measured bio signal, And a blood pressure estimation unit for estimating the blood pressure of the user by using a blood pressure estimation function corresponding to the state type corresponding to the physical condition of the user based on the corresponding relationship information.

Further, the state type is classified according to the movement state of the user, and the measurement unit measures the electrocardiogram from the user as the predetermined type of living body signal, further measures the acceleration of the user, and the state determination unit Determining a state type corresponding to the physical condition of the user based on a result of comparing the heart rate and the acceleration with a predetermined reference heart rate and a predetermined reference acceleration from the electrocardiogram measurement result, .

The condition type is classified according to the user's medication usage state, and the measurement unit measures the cardiac trajectory and the optic-free pulse wave from the user as the predetermined type of living body signal, The state type corresponding to the physical condition of the user is determined based on the result of comparing the signal size of the trajectory and the signal amplitude of the optical pulse signal of the optical pulse width with the predetermined reference heartbeat signal magnitude and the predetermined magnitude of the reference pulse amplitude signal .

Further, the database further stores a plurality of measurement data including biometric signal data that is a result of measuring the predetermined type of living body signal from a person's body, and each of the plurality of measurement data includes one of the status types Wherein the state determination unit determines a state type corresponding to the user's body state based on a result of analyzing the plurality of measurement data using a machine running technique, Linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning.

In addition, the blood pressure estimation apparatus may further include an arithmetic section, wherein the database stores biometric signal data, which is a result of measurement of the predetermined type of biometric signal from a person's body, Wherein each of the plurality of measurement data is matched to any one of the state types, and the operation unit is operable to calculate a specific state type of one of the state types Calculates a function indicating a relation between the biometric signal data and the blood pressure data included in each of the measurement data matched to the specific state type using a plurality of predetermined machine learning techniques, And an error between the result of estimating the blood pressure based on the calculated function and the blood pressure data is minimum Can be set to a blood pressure estimation function corresponding to the particular status types.

In addition, the plurality of machine learning techniques may be selected from linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning.

A method for estimating a blood pressure according to an embodiment of the present invention includes the steps of measuring a predetermined type of a living body signal from a user, measuring a state of the user's body state based on the measured living body signal, Using the database storing the correspondence relationship information in which each of the state types is associated with at least one of a plurality of predetermined blood pressure estimation functions for estimating the blood pressure of the user from the measured bio signal, And estimating the blood pressure of the user using the blood pressure estimation function corresponding to the state type corresponding to the physical condition of the user based on the corresponding relationship information.

Further, the state type is classified according to the movement state of the user, and the measuring step includes the steps of measuring the electrocardiogram from the user as the predetermined type of living body signal, and further measuring the acceleration of the user Wherein the determining comprises determining a heart rate of the user based on the result of the electrocardiogram measurement and comparing the heart rate and the acceleration with a predetermined reference heart rate and a predetermined reference acceleration, And determining a status type corresponding to the status type.

Wherein the state type is classified according to the user's medication usage state, and the measuring step includes the step of measuring, from the user, cardiac trajectory and optical pulse wave as the predetermined type of living body signal, Wherein the determining step includes the step of comparing the signal amplitude of the cardiac trajectory and the signal amplitude of the optical pulse wave signal with the predetermined reference heartbeat signal magnitude and the predetermined magnitude of the reference optical pulse pulse signal, And determining the corresponding state type.

Further, the database further stores a plurality of measurement data including biometric signal data that is a result of measuring the predetermined type of living body signal from a person's body, and each of the plurality of measurement data includes one of the status types Wherein the determining comprises determining a state type corresponding to a user's physical condition based on a result of analyzing the plurality of measurement data using a machine running technique, The machine learning technique may be any one of linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning.

In addition, the database may further comprise: biometric signal data that is a result of measuring the predetermined type of biometric signal from a person's body; and a plurality of blood pressure data that is a result of measuring the blood pressure of the person at the time when the biometric signal data is acquired Wherein each of the plurality of measurement data is matched to any one of the state types and is included in each of the measurement data matched to the specific state type in one of the state types Calculating a function indicative of a relationship between the biomedical signal data and the blood pressure data using a plurality of predetermined machine learning techniques, calculating a blood pressure based on the calculated function among the calculated functions, The function having the smallest error between the data is determined as the blood pressure estimation function corresponding to the specific state type And a step of setting the display device.

In addition, the plurality of machine learning techniques may be selected from linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning.

According to an embodiment of the present invention, in estimating a blood pressure of a user using a bio-signal, data on a user's physical condition is obtained using a bio-signal measuring device and various sensors, and based on the obtained data After analyzing the user's body condition, the user's blood pressure can be accurately estimated by applying an algorithm according to the user's body condition among various blood pressure estimation algorithms. Accordingly, it is possible to solve the problem of the prior art that the accuracy of blood pressure estimation varies according to the user's body condition.

1 is a conceptual diagram illustrating an activity model-based blood pressure estimation technique using bio-signals according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating a configuration of an activity model-based blood pressure estimation apparatus using bio-signals according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating a procedure of a blood pressure estimation method performed by an activity-model-based blood pressure estimation apparatus using bio-signals according to an embodiment of the present invention.
4 is a view showing bio-signals used in an activity-model-based blood pressure estimation apparatus and method using bio-signals according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. To fully disclose the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims.

In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The following terms are defined in consideration of the functions in the embodiments of the present invention, which may vary depending on the intention of the user, the intention or the custom of the operator. Therefore, the definition should be based on the contents throughout this specification.

1 is a conceptual diagram illustrating an activity model-based blood pressure estimation technique using bio-signals according to an embodiment of the present invention. Referring to FIG. 1, in a blood pressure estimation technique according to an embodiment of the present invention, a bio-signal measured from a user may be used not only for estimating a blood pressure but also for determining a user's physical condition. Here, the biological signal may be an electrocardiogram, cardiac trajectory, optical pulse-wave pulse wave, or the like. However, in order to determine the physical condition, heart rate of a user obtained through a method such as analyzing an electrocardiogram and acceleration of a user obtained through an acceleration sensor or the like may be additionally used. Further, in estimating the blood pressure, in addition to the measured biological signal itself, a secondary biological signal PTT (pulse transit time) obtained by analyzing the measured biological signal, a pre-ejection period (PEP) have.

The user's physical condition determined using the necessary information can be classified according to a plurality of status types. Such a state type includes a state indicating a state of the user's motion such as a state of taking a sleep or rest, a state of isotonic movement such as walking or running, a state of isometric movement such as sitting for a long time or lifting a heavy object, There can be types. On the other hand, there may be state types that indicate drug-taking states related to whether or not they have taken drugs such as glyceryl trinitate (GTN), angiotensin, norepinephrine or midodrine .

If it is determined that the user's body state corresponds to any one of the state types, a blood pressure estimation function corresponding to the state type is selected, and the measured user's biological signal is used as a factor of the selected blood pressure estimation function The value obtained by the selected blood pressure estimation function can be estimated as the blood pressure of the user. That is, according to one embodiment of the present invention, there is a blood pressure estimation function corresponding to each of various state types representing the user's body state, and the blood pressure estimation function corresponding to the state type corresponding to the state type, The user's blood pressure can be estimated through the estimation function.

FIG. 2 is a diagram illustrating a configuration of an activity model-based blood pressure estimation apparatus using bio-signals according to an embodiment of the present invention. 2 may include a measurement unit 110, a state determination unit 120, a database 130, a blood pressure estimation unit 140, an operation unit 150, and an output unit 160 . However, since the blood pressure estimation apparatus 100 of FIG. 2 is only an embodiment of the present invention, the concept of the present invention is not limited to FIG.

The measurement unit 110 can measure a predetermined type of living body signal from the user. Such bio-signals may be electrocardiogram, heart trajectory, pulse wave pulse wave, etc., but the present invention is not limited thereto. The measuring unit 110 may include an electrode for measuring a biological signal and an interface for the electrode. In addition, the measuring unit 110 may further measure the acceleration of the user. In this case, the measuring unit 110 may include an acceleration sensor.

The state determination unit 120 may determine which of a plurality of state types the user's body state based on the bio-signal measured by the measurement unit 110 is. Specifically, the state determiner 120 can determine a heart rate indicating how many times the user's heart (for example, minute) jumps (i.e., how fast the heart is running) from a bio-signal such as an electrocardiogram. The state determiner 120 can determine the state type of the user's body state using the thus derived heart rate of the user and the acceleration of the user measured by the measuring unit 110. [

In addition, the state determiner 120 may determine which state type the user's body state corresponds to, using the cardiac tilt measured by the measuring unit 110 and the signal amplitude of the optical pulse wave. Various machine learning techniques such as linear regression analysis, support vector machine (SVM), artificial neural network (ANN), or deep learning may be used. The state determining unit 120 may determine a state of the user by using a microprocessor together with the blood pressure measuring unit 140 and the calculating unit 150 to be described later. May be implemented by an arithmetic and logic unit that includes a processor.

The database 130 may store various data necessary for the operation of the blood pressure estimation apparatus 100. [ Such data may include information on a plurality of state types for classifying the user's physical condition. In addition, the database 130 may store a plurality of blood pressure estimation functions for estimating the blood pressure of the user from the bio-signals measured by the measurement unit 110. Specific details of the blood pressure estimation function will be described later. In addition, the database 130 may store correspondence relationship information that makes each of the plurality of state types correspond to at least one of the plurality of blood pressure estimation functions.

In addition, the database 130 may include a plurality of measurement data. Each of the measurement data includes biometric signal data that is a result of measurement of the predetermined type of biometric signal from the user or other persons and biometric signal data that indicates the blood pressure of a person who is a source of the biometric signal data at the time point when the biometric signal data is acquired And blood pressure data as a result of the measurement can be included. That is, the database 130 stores a plurality of measurement data in which a result of measuring a person's biological signal and blood pressure simultaneously is recorded. Such a measurement data can be analyzed through a machine learning technique after a sufficient number of accumulated data. Details of such an analysis technique will be described later.

Such a database 130 may be embodied as a computer readable recording medium. Examples of such a computer readable recording medium include magnetic media such as a hard disk, a floppy disk and a magnetic tape, a CD-ROM, a DVD Magneto-optical media such as optical media, floptical disks, and hardware devices that are specially configured to store and execute program instructions such as flash memory. have.

The blood pressure estimation unit 140 selects a blood pressure estimation function corresponding to the state type corresponding to the user's body condition based on the corresponding relationship information stored in the database 130 and uses the measured biometrics signal as a factor of the selected function The user's blood pressure can be estimated.

The calculating unit 150 may calculate a function indicating the relationship between the biometric signal data and the blood pressure data included in each of the measurement data matched to the specific state type in one specific state type of the state types, Respectively. The calculating unit 150 may set an optimal function among the calculated functions (for example, a function that minimizes an error between the estimated blood pressure value and the blood pressure data by the function) as a blood pressure estimation function corresponding to the specific state type. The calculation unit 150 can receive specific information from the portable device through communication with the portable device such as a mobile phone. For example, the blood pressure estimation apparatus 100 may replace the function of the measurement unit 110 by a method such as receiving the user's bio-signal acquired by the user's portable device from the portable device. In addition, the arithmetic unit 150 may serve as a controller for controlling other components of the blood pressure estimation apparatus 100 as a whole.

The output unit 160 can output the specific information in a form recognizable to the user based on the control of the blood pressure estimating unit 140. [ The specific information may represent the blood pressure of the user, which is typically estimated by the blood pressure estimating unit 140. The output unit 150 may be implemented by a display device such as a liquid crystal display (LCD) or an organic light emitting diode (OLED), or an audio output device such as a speaker.

FIG. 3 is a diagram illustrating a procedure of a blood pressure estimation method performed by an activity-model-based blood pressure estimation apparatus using bio-signals according to an embodiment of the present invention. 4 is a diagram illustrating bio-signals used in an activity-model-based blood pressure estimation apparatus and method using bio-signals according to an embodiment of the present invention.

4, in the blood pressure estimation method according to the embodiment of the present invention, in addition to the ECG, the cardiac trajectory (BCG), and the light pulse width (PPG), a secondary signal PTT and PEP may be additionally used. The PTT corresponds to the "pulse wave delivery time", and more specifically, the time at which the pulse is delivered from the heart to other parts of the body. Such a PTT can be obtained by comparing an ECG, heart trajectory or optical pulse wave signal measured near the heart with an optical pulse wave signal measured at other parts of the body (for example, a finger or a toe). The use of the optic pulse wave signal can be measured even if it is not near the heart. Specifically, the difference in phase (for example, the time difference between the peaks of two signals) in the time axis of the waveform of the two signals to be compared can be the PTT. PEP refers to the time between the electrical signal of the heart and the physical response of the heart, specifically, by comparing the electrocardiogram signal with the cardiac tidal signal (e.g., calculating the time difference between the peak of the electrocardiographic signal and the peak of the cardiac tidal signal) . These PTT and PEP signals can be factors of the blood pressure estimation function to be described later together with electrocardiogram, heart balloon, light pulse wave and heart rate. On the other hand, the signal amplitude (amplitude) or waveform of the bio-signals can be used as a characteristic of the bio-signal.

3, each step of the blood pressure estimation method according to the embodiment of the present invention will be described below. The description of the parts overlapping with those of FIG. 1 and FIG. 2 may be omitted . In addition, each of the following steps need not be performed in order, and the order may be changed as necessary.

First, a function indicating the relationship between the biometric signal data and the blood pressure data can be calculated for a specific state type (S110). More specifically, the state determiner 120 may select one of the plurality of state types stored in the database 130. For example, Such a state type may be related to a user's exercise state such as rest, isotonic exercise, isometric exercise, or the like, and may be related to a user's medication taking state indicating whether or not to take various medicines.

Suppose you choose isotonic exercise below. Then, the operation unit 150 can call the stored measurement data that matches the isotonic motion in the database 130. That is, the measurement data is data generated by measuring a biological signal and blood pressure when a subject (user or other person) is performing isotonic motion. In order to analyze using a machine learning technique, which will be described later, It can be accumulated. Each of these measurement data may include biometric signal data and blood pressure data based on actual measurement results.

By analyzing a plurality of measurement data matched with the isotonic motion using each of a plurality of machine learning techniques, a blood pressure estimation function corresponding to each machine learning technique can be calculated. That is, assuming that the machine learning method uses four kinds of linear regression analysis, SVM, ANN, and deep learning, there are four blood pressure estimation functions.

Next, the optimal function among the calculated functions may be set as a blood pressure estimation function corresponding to the specific state type (S120). In the above example, the bio-signal data (electrocardiogram, cardiac trajectory, optical pulse voltage, PTT, PEP, heart rate, etc.) included in any one of the measurement data matched to the isotonic motion is converted into a blood pressure estimation function , The blood pressure estimation value will be output as the output value of the function. Then, the estimation value can be compared with the biometric signal data and the blood pressure data included in the measurement data, and the error can be obtained. The overall error can be obtained by performing this process on all of the measurement data matched to the isotonic motion. Such an overall error can be obtained, for example, by using a least squares approximation, but is not limited thereto. The above-described process can be performed for each of the blood pressure estimation functions calculated using three different machine learning techniques other than the linear regression analysis. Then, for example, the blood pressure estimation function calculated by the SVM may be selected as the blood pressure estimation function having the smallest overall error, and the blood pressure estimation function calculated by the SVM may be used as the blood pressure estimation function for the state type. Hereinafter, when the user is determined to be in the isotonic motion through steps S130 to S150, which will be described later, the user's blood pressure can be estimated using the blood pressure estimation function calculated by the SVM.

Next, the predetermined type of living body signal can be measured from the user (S130). Since the types of biological signals have already been described, detailed description thereof will be omitted here. In addition, the information such as the heart rate and the acceleration of the user along with the bio-signal can be obtained as described above. Subsequently, based on the measured bio-signal, it is possible to determine which of the state types stored in the database 130 the user's body state (S140).

The body condition judging process will be described in more detail as follows. First, the physical condition of the user can be determined by observing how the heart rate, acceleration, signal amplitude of the cardiac trajectory, signal amplitude of the optical pulse wave, etc., has changed compared to the reference state (e.g., rest). Specific criteria for this are summarized in Table 1 below.

Biological activity Heart rate acceleration Heart ballistics Optic pulse wave Relaxation (standard condition) equal equal equal equal Isotonic exercise Increase Increase - - Isometric exercise Increase equal - - Taking GTN equal equal - Increase Taking angiotensin equal equal Increase degradation Taking norepinephrine equal equal equal degradation Taking midodrine equal equal degradation degradation

According to the above table 1, for example, when the heart rate rises compared to the resting state in the reference state, the acceleration is equal (the minute rise or fall of the rise or fall ratio can be regarded as equivalent) It can be judged to be isometric exercise. In addition, if the heart rate and acceleration are equal to the rest state but the signal amplitude of the cardiac trajectory and the optical pulse wave falls, it can be determined that the user has taken the same drug as the unadorned state.

Alternatively, the state of the user may be determined using a machine learning technique such as linear regression analysis, SVM, ANN, and deep running as described above. Similar to S110 and S120 described above, it is possible to derive a function that outputs heart rate, acceleration, cardiac trajectory, and optical pulse wave as factors and outputs a state type by performing machine learning on the measurement data of the database 130. [ Then, by inputting factors such as heart rate, acceleration, heart trajectory, and optical pulse wave measured by the user to the function, it is possible to determine what state type the user's body state corresponds to.

Finally, the blood pressure of the user can be estimated using the blood pressure estimation function corresponding to the state type corresponding to the current physical condition of the user (S150). For example, when it is determined that the user is currently performing the isotonic motion, the blood pressure estimation function calculated by the SVM determined to correspond to the isotonic motion as exemplified in steps S110 and S120 is used to calculate the blood pressure Can be estimated.

As described above, according to an embodiment of the present invention, in estimating a blood pressure of a user using a bio-signal, the user's physical condition is analyzed based on data on the user's physical condition, The user's blood pressure can be accurately estimated by applying an algorithm that matches the user's physical condition. Accordingly, it is possible to solve the problem of the prior art that the accuracy of blood pressure estimation varies according to the user's body condition.

Combinations of each step of the flowchart and each block of the block diagrams appended to the present invention may be performed by computer program instructions. These computer program instructions may be loaded into a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus so that the instructions, which may be executed by a processor of a computer or other programmable data processing apparatus, And means for performing the functions described in each step are created. These computer program instructions may also be stored in a computer usable or computer readable memory capable of directing a computer or other programmable data processing apparatus to implement the functionality in a particular manner so that the computer usable or computer readable memory It is also possible for the instructions stored in the block diagram to produce a manufacturing item containing instruction means for performing the functions described in each block or flowchart of the block diagram. Computer program instructions may also be stored on a computer or other programmable data processing equipment so that a series of operating steps may be performed on a computer or other programmable data processing equipment to create a computer- It is also possible that the instructions that perform the processing equipment provide the steps for executing the functions described in each block of the block diagram and at each step of the flowchart.

In addition, each block or each step may represent a module, segment, or portion of code that includes a plurality of executable instructions for executing the specified logical function (s). It should also be noted that in some alternative embodiments, the functions mentioned in the blocks or steps may occur out of order. For example, two blocks or steps shown in succession may in fact be performed substantially concurrently, or the blocks or steps may sometimes be performed in reverse order according to the corresponding function.

The above description is merely illustrative of the technical idea of the present invention, and various modifications and changes may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents thereof should be construed as falling within the scope of the present invention.

100: blood pressure presumption device
110:
120:
130: Database
140: blood pressure estimation unit
150:
160: Output section

Claims (14)

A measurement unit for measuring a predetermined type of bio-electrical signal from a user;
A state determiner for determining, based on the measured bio-signal, which of the plurality of state types the user's body state corresponds to;
A database that stores correspondence relationship information that makes each of the state types correspond to at least one of a plurality of predetermined blood pressure estimation functions for estimating the blood pressure of the user from the measured bio signal; And
And a blood pressure estimation unit for estimating a blood pressure of the user by using a blood pressure estimation function corresponding to a state type corresponding to the physical condition of the user based on the corresponding relationship information,
Wherein the status type is classified according to a drug taking state of the user,
Wherein the database stores biometric signal data that is a result of periodically measuring the predetermined type of biometric signal from a plurality of people's body and blood pressure data that is a result of periodic measurement of the blood pressure of the plurality of persons at the time when the biometric signal data is acquired Further storing a plurality of measurement data,
Each of the plurality of measurement data being matched to any of the status types,
A plurality of functions indicating a relationship between the biometric signal data and the blood pressure data included in each of the measurement data matched to the specific state type in a specific state type of the state type, And a calculation unit for setting a function having the smallest error between the blood pressure data and the result of the estimation of the blood pressure based on the calculated function among the calculated functions as the blood pressure estimation function corresponding to the specific state type Included
Blood pressure estimation device. The method according to claim 1,
Wherein the state type is classified according to a motion state of the user,
Wherein the measuring unit measures the electrocardiogram from the user as the predetermined type of living body signal, further measures the acceleration of the user,
Wherein the state determination unit determines the state of the user based on the result of comparing the heart rate and the acceleration with a predetermined reference heart rate and a predetermined reference acceleration from the electrocardiogram measurement result, To determine the type
Blood pressure estimation device. The method according to claim 1,
Wherein the measuring section measures the cardiac trajectory and the optical pulse wave from the user as the predetermined type of living body signal,
Wherein the state determining unit determines the state of the user based on a result of comparing the signal amplitude of the cardiac trajectory and the signal amplitude of the optical pulse wave with the predetermined reference heartbeat signal magnitude and the predetermined magnitude of the reference optical pulse wave signal, Determine the corresponding status type
Blood pressure estimation device. The method according to claim 1,
Wherein the state determination unit determines a state type corresponding to a physical state of the user based on a result of analyzing the plurality of measurement data using a machine running technique,
The machine learning technique is one of linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning
Blood pressure estimation device. delete The method according to claim 1,
The plurality of machine learning techniques may be selected from among linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning
Blood pressure estimation device. Measuring a predetermined type of bio-electrical signal from a user;
Determining whether the physical condition of the user corresponds to one of a plurality of predetermined state types based on the measured bio-signal; And
A database storing correspondence relation information for associating each of the state types with at least one of a plurality of predetermined blood pressure estimation functions for estimating the blood pressure of the user from the measured bio signal, And estimating the blood pressure of the user by using a blood pressure estimation function corresponding to a state type corresponding to the user's physical condition,
Wherein the status type is classified according to a drug taking state of the user,
Wherein the database stores biometric signal data that is a result of periodically measuring the predetermined type of biometric signal from a plurality of people's body and blood pressure data that is a result of periodic measurement of the blood pressure of the plurality of persons at the time when the biometric signal data is acquired Further storing a plurality of measurement data,
Each of the plurality of measurement data being matched to any of the status types,
A plurality of functions indicating a relationship between the biometric signal data and the blood pressure data included in each of the measurement data matched to the specific state type in a specific state type of the state type, ; And
Setting a function having a minimum error between the blood pressure data and a result of estimating the blood pressure based on the calculated function among the calculated functions as a blood pressure estimation function corresponding to the specific state type
Blood pressure estimation method. 8. The method of claim 7,
Wherein the state type is classified according to a motion state of the user,
Measuring the electrocardiogram from the user as the predetermined type of living body signal; And
Further comprising measuring the acceleration of the user,
Wherein the determining step comprises the steps of: determining a heart rate of the user from the electrocardiogram measurement result; comparing the heart rate and the acceleration with a predetermined reference heart rate and a predetermined reference acceleration, Determining a state type
Blood pressure estimation method. 8. The method of claim 7,
Wherein said measuring step includes the step of measuring cardiac trajectory and optically-impulse pulse wave from said user as said predetermined type of biometric signal,
The method of claim 1, wherein the determining comprises: comparing a signal amplitude of the cardiac trajectory and a signal amplitude of the optical pulse wave signal with a predetermined reference heartbeat signal amplitude and a predetermined reference optical pulse wave signal amplitude, And determining a state type corresponding to
Blood pressure estimation method. 8. The method of claim 7,
Wherein the determining includes determining a state type corresponding to a physical condition of the user based on a result of analyzing the plurality of measurement data using a machine running technique,
The machine learning technique is one of linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning
Blood pressure estimation method. delete 8. The method of claim 7,
The plurality of machine learning techniques may be selected from among linear regression analysis, support vector machine (SVM), artificial neural network (ANN), and deep learning
Blood pressure estimation method. 12. A program stored in a computer-readable medium for performing the respective steps according to the method of any one of claims 7 to 10 and 12. 13. A computer-readable medium having recorded thereon instructions for performing the respective steps according to the method recited in any one of claims 7 to 10 and 12.

Images