BLOOD PRESSURE MEASURING DEVICE, MODEL SETTING DEVICE, AND BLOOD PRESSURE MEASURING METHOD

The blood pressure measuring device enhances precision by classifying attributes and using personalized models based on vascular age and gender, addressing the inaccuracies of conventional methods.

JP7591076B2Active Publication Date: 2024-11-27SHARP KK
View PDF 14 Cites 0 Cited by

Patent Information

Application Number
JP2023033423
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-02-01
Filing Date
2023-03-06
Publication Date
2024-11-27
Estimated Expiration
2040-01-29

AI Technical Summary

Technical Problem

Existing methods for measuring blood pressure lack precision and do not adequately account for individual differences in subjects, such as age and gender, leading to inaccuracies in blood pressure estimation.

Method used

A blood pressure measuring device that utilizes a pulse wave acquisition unit to calculate pulse wave parameters, classifies attributes based on vascular age and gender, and selects a personalized blood pressure estimation model for accurate measurement.

Benefits of technology

The device achieves higher precision in blood pressure measurement by creating individualized models that account for unique vascular conditions, improving accuracy over conventional methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007591076000005
    Figure 0007591076000005
  • Figure 0007591076000006
    Figure 0007591076000006
  • Figure 0007591076000007
    Figure 0007591076000007
Patent Text Reader

Abstract

To measure a living body's blood pressure with higher accuracy than conventional methods. [Solution] A blood pressure measurement device measures a first blood pressure of a subject. In the blood pressure measurement device, an attribute classification unit classifies attributes of the subject based on attribute information related to the vascular condition of the subject. The blood pressure measurement device is communicably connected to a model storage unit that stores in advance at least one blood pressure estimation model corresponding to the attribute classification result. The blood pressure measurement unit calculates the first blood pressure based on pulse wave parameters using at least one blood pressure estimation model corresponding to the classification result.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] One aspect of the present disclosure relates to a blood pressure measurement device that measures the blood pressure of a living body based on a pulse wave of the living body. This application claims priority based on Japanese Patent Application No. 2019-17189, filed on February 1, 2019, the contents of which are incorporated herein by reference. [Background technology]

[0002] In recent years, various techniques for measuring the blood pressure of a living body (e.g., a subject) have been proposed. As an example, Patent Document 1 discloses a technique for easily measuring (more precisely, estimating) the blood pressure of a living body. Specifically, the technique of Patent Document 1 estimates the blood pressure of the living body using a model (calculation formula) formulated in advance according to the age (actual age) and sex of the living body. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2010-220690 A [Patent Document 2] JP 2002-238867 A [Patent Document 3] JP 2015-40839 A [Patent Document 4] JP 2009-086901 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, as described below, there is still room for improvement in a specific technique for measuring the blood pressure of a living body with higher accuracy. An object of one aspect of the present disclosure is to measure the blood pressure of a living body with higher accuracy than conventional techniques. [Means for solving the problem]

[0005] In order to solve the above problems, a blood pressure measurement device according to one embodiment of the present disclosure is a blood pressure measurement device that measures a first blood pressure of a living body based on a pulse wave of the living body, and includes a pulse wave acquisition unit that acquires at least one pulse wave in a predetermined region on a body surface of the living body, a pulse wave parameter calculation unit that calculates at least one pulse wave parameter based on the at least one pulse wave, a blood pressure acquisition unit that acquires a second blood pressure of the living body, an attribute information acquisition unit that acquires attribute information that is information related to a vascular state of the living body, and an attribute classification unit that classifies an attribute of the living body based on the attribute information, wherein the blood pressure measurement device is communicatively connected to a model storage unit in which at least one blood pressure estimation model for estimating the first blood pressure corresponding to a classification result of the attribute is stored in advance, and the blood pressure measurement device further includes a first blood pressure measurement unit that calculates the first blood pressure based on the at least one pulse wave parameter by using the at least one blood pressure estimation model corresponding to the classification result of the attribute, and the attribute classification unit classifies the attribute of the living body based on a classification index calculated from the second blood pressure acquired by the blood pressure acquisition unit.

[0006] In order to solve the above problem, a model setting device according to one aspect of the present disclosure is a model setting device communicably connected to a blood pressure measuring device that measures a first blood pressure of a living organism based on a pulse wave of the living organism, and includes a second blood pressure measuring unit that measures a second blood pressure of the living organism, a pulse wave acquiring unit that acquires at least one pulse wave in a predetermined area on a body surface of the living organism, a pulse wave parameter calculating unit that calculates at least one pulse wave parameter based on the at least one pulse wave, an attribute information acquiring unit that acquires attribute information that is information related to a vascular state of the living organism, and a classification unit that classifies attributes of the living organism based on the attribute information. and a gender classification unit, the model setting device being communicatively connected to a model storage unit capable of storing at least one blood pressure estimation model for estimating the first blood pressure corresponding to a classification result of the attribute, the model setting device further comprising a model creation unit which creates the at least one blood pressure estimation model based on the at least one pulse wave parameter and the second blood pressure and stores the at least one blood pressure estimation model in the model storage unit, and the attribute classification unit classifies the attributes of the living body based on a classification index calculated from the second blood pressure acquired by the blood pressure acquisition unit.

[0007] In order to solve the above-mentioned problems, a blood pressure measurement method according to one embodiment of the present disclosure is a blood pressure measurement method using a blood pressure measurement device that measures a first blood pressure of a living body based on a pulse wave of the living body, and includes a pulse wave acquisition step of acquiring at least one pulse wave in a predetermined area on a body surface of the living body, a pulse wave parameter calculation step of calculating at least one pulse wave parameter based on the at least one pulse wave, a blood pressure acquisition step of acquiring a second blood pressure of the living body, an attribute information acquisition step of acquiring attribute information which is information related to a vascular state of the living body, and an attribute classification step of classifying an attribute of the living body based on the attribute information, wherein the blood pressure measurement device is communicably connected to a model storage unit in which at least one blood pressure estimation model for estimating the first blood pressure corresponding to a classification result of the attribute is stored in advance, and the blood pressure measurement method includes a first blood pressure measurement step of calculating the first blood pressure based on the at least one pulse wave parameter by using the at least one blood pressure estimation model corresponding to the classification result of the attribute, and the attribute classification step further includes a step of classifying an attribute of the living body based on a classification index calculated from the second blood pressure acquired in the blood pressure acquisition step. Effect of the Invention

[0008] According to the blood pressure measurement device of the present disclosure, it is possible to measure the blood pressure of a living body with higher accuracy than in the past. The blood pressure measurement method of the present disclosure also provides the same effect. Moreover, the model setting device of the present disclosure also provides the same effect. [Brief description of the drawings]

[0009] [Figure 1] FIG. 2 is a functional block diagram showing the configuration of the main parts of the blood pressure measurement device of the first embodiment. [Diagram 2] 11 is a diagram illustrating an example of a process performed by a face image division unit. FIG. [Diagram 3] FIG. 2A is a diagram showing an example of an acceleration pulse wave waveform obtained from a subject with a low vascular age, and FIG. 2B is a diagram showing an example of an acceleration pulse wave waveform obtained from a subject with a high vascular age. [Figure 4]FIG. 13 is a diagram showing an example of a result of classifying a plurality of subjects based on waveform feature amounts. [Diagram 5] FIG. 1A is a diagram showing an example of the relationship between true blood pressure values ​​and predicted blood pressure values ​​in a common model, and FIG. 1B is a diagram showing an example of the relationship between true blood pressure values ​​and predicted blood pressure values ​​in an attribute-specific model. [Figure 6] 11 is a diagram for explaining an example of a process for setting a measurement model in a model evaluation unit. FIG. [Figure 7] FIG. 11 is a diagram illustrating a process flow of a measurement model setting method. [Figure 8] FIG. 4 is a diagram illustrating an example of a power spectrum of a pulse wave signal. [Figure 9] FIG. 13 is a diagram showing an example of the relationship between blood pressure and PWV for subjects having different attributes. [Figure 10] FIG. 11 is a functional block diagram showing the configuration of the main parts of a blood pressure measurement device according to a second embodiment. [Figure 11] FIG. 11 is a diagram showing an example of mean blood pressure values ​​and attributes. [Figure 12] FIG. 13 is a diagram showing an example of manual input of blood pressure values. [Figure 13] 13A is an example of information in which the name of a subject is associated with the subject's attributes, and FIG. 13B is an example of information in which the ID of a subject is associated with the subject's attributes. [Figure 14] 10 is a flowchart showing an example of a processing flow of the blood pressure measurement device. [Figure 15] 1A and 1B are diagrams illustrating an example of a scatter plot of mean blood pressure and pulse pressure, and attributes corresponding to the mean blood pressure and pulse pressure. [Figure 16] FIG. 13 is a diagram illustrating an example of a modified example using a cloud. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] [Embodiment 1] The blood pressure measuring device 1 of the first embodiment will be described below. For convenience, in each of the following embodiments, the same reference numerals are given to components having the same functions as those described in the first embodiment, and the description thereof will not be repeated. Descriptions of matters similar to those of known techniques will also be omitted as appropriate. Note that the device configurations shown in each drawing are merely examples for convenience of description. In addition, each numerical value described below in the specification is merely an example.

[0011] (Overview of blood pressure measuring device 1) FIG. 1 is a functional block diagram showing the configuration of the main parts of a blood pressure measurement device 1 of the first embodiment. The blood pressure measurement device 1 measures the blood pressure (hereinafter simply referred to as blood pressure) of a subject H (living body) based on the pulse wave of the subject H. Specifically, the blood pressure measurement device 1 measures the blood pressure using a blood pressure measurement model (hereinafter also simply referred to as a "measurement model") set in a model setting device 100 described below. In this specification, a blood pressure estimation model described below is also simply referred to as an "estimation model". In addition, the measurement model and the estimation model may be collectively referred to simply as a "model".

[0012] In the following description, the blood pressure measuring device 1 will be described as a non-contact blood pressure measuring device (a measuring device capable of measuring blood pressure without contacting the subject H). In the first embodiment, a case where the subject H is a human will be exemplified. The blood pressure measuring device 1 measures blood pressure by treating a predetermined region on the body surface of the subject H as a ROI (Region of Interest). In the following description, a case where the ROI is a face will be exemplified. In this specification, the face of the subject H will also be simply referred to as the "face". The same applies to other notations.

[0013] The blood pressure measurement device 1 includes a model setting device 100, a model selection unit 60, a pulse wave signal quality evaluation unit 150, a blood pressure measurement unit 160 (first blood pressure measurement unit), and a blood pressure measurement result output unit 170. The model setting device 100 includes a blood pressure acquisition unit 2 (second blood pressure measurement unit), a pulse wave acquisition unit 10, a pulse wave parameter calculation unit 20, a vascular age calculation unit 21, a gender detection unit 22, an attribute classification unit 23, a model creation unit 30, a model evaluation unit 40, and a model storage unit 55.

[0014] 1, the model setting device 100 is provided inside the blood pressure measurement device 1. However, the model setting device 100 can also be provided outside the blood pressure measurement device 1 (see embodiment 2 described later).

[0015] The blood pressure acquiring unit 2 measures the blood pressure of the subject H. The blood pressure acquiring unit 2 is a contact type blood pressure monitor (e.g., a cuff blood pressure monitor). The blood pressure (hereinafter, BPm) measured by the blood pressure acquiring unit 2 is used as test data (or training data) in the model setting device 100. That is, BPm is used to set a measurement model in the model setting device 100 (more specifically, in the model evaluation unit 40). In addition, BPm is also used to create at least one estimation model in the model creation unit 30.

[0016] The blood pressure acquisition unit 2 outputs BPm to the model creation unit 30 and the model evaluation unit 40 (more specifically, the model evaluation index calculation unit 42 described below). The final blood pressure measurement result (P described below) in the blood pressure measurement device 1 is also referred to as the first blood pressure. To distinguish it from the first blood pressure, BPm is also referred to as the second blood pressure. As described below, P is measured (calculated) by the blood pressure measurement unit 160.

[0017] (Pulse wave acquisition unit 10) The pulse wave acquisition unit 10 acquires a pulse wave in an ROI (e.g., a face). The pulse wave acquisition unit 10 includes an imaging unit 11, a light source 12, a light source adjustment unit 13, a face image acquisition unit 14, a face image division unit 15, a skin region extraction unit 16, and a pulse wave calculation unit 17.

[0018] The imaging unit 11 is a camera including an image sensor and a lens. For example, a CMOS (Complementary Metal-Oxide Semiconductor) type or a CCD (Charge-Coupled Device) type image sensor may be used as the image sensor. The imaging unit 11 captures images of the subject H multiple times at a predetermined frame rate (i.e., at predetermined time intervals) and outputs the captured images of the subject H (hereinafter, subject images) to the face image acquisition unit 14. As an example, the frame rate is 300 fps (frames per second).

[0019] The imaging unit 11 may include a known color filter. It is preferable that the color filter has optical properties suitable for observing the increase or decrease in blood volume. Examples of suitable color filters include (i) an RGBCy (Red, Blue, Green, Cyan) color filter, and (ii) an RGBIR (Red, Blue, Green, Infrared) color filter. Alternatively, a Bayer array color filter may be used as the color filter. Thus, the imaging unit 11 may be an RGB camera or an IR camera.

[0020] When the imaging unit 11 images the subject H, the light source 12 irradiates light onto the subject H. The light source adjustment unit 13 adjusts the light source 12. As an example, it is preferable that the light source adjustment unit 13 adjusts the light source so that the pulse wave propagation time (an example of a pulse wave parameter) between regions used in the measurement model set by the model setting device 100 can be calculated with high accuracy.

[0021] Specifically, the light source adjustment unit 13 adjusts the light source 12 so that a pulse wave having a certain signal quality can be detected in the relevant region. A "pulse wave having a certain signal quality" is, for example, a "pulse wave with a high SNR (Signal-to-Noise Ratio)". More specifically, the light source adjustment unit 13 adjusts at least one of (i) the light intensity of the light source 12, (ii) the optical spectrum of the light source 12, and (iii) the irradiation angle with respect to the skin of the subject H.

[0022] The pulse wave acquiring unit 10 does not necessarily have to be provided with the light source 12 and the light source adjusting unit 13. When the light source 12 and the light source adjusting unit 13 are not provided, the imaging unit 11 may image the subject H using only the ambient light.

[0023] The facial image acquiring section 14 extracts the facial region of the subject H from the subject image captured by the imaging section 11. The facial image acquiring section 14 acquires the image after the facial region is extracted as a facial image (an image showing the face of the subject H). The facial image is an example of an image including an image of an ROI. As an example, the facial image acquiring section 14 may extract the facial region at a predetermined frame interval of a moving image (a moving image composed of a plurality of subject images) showing the subject by performing face tracking on the moving image.

[0024] However, the face image acquiring unit 14 can extract the face area without necessarily performing face tracking. For example, (i) the subject H may be made to place his / her face in a preset frame, and (ii) the subject image may be captured by the imaging unit 11 in a state in which the face and the imaging unit 11 are fixed. In such a case, face blurring in the subject image can be suppressed, and therefore face tracking is not necessary.

[0025] The face image division unit 15 divides the face image extracted by the face image acquisition unit 14 into a plurality of regions (partial regions). In the following description, for convenience, the face image is also referred to as "IMG." FIG. 2 is a diagram for explaining an example of the processing of the face image division unit 15. FIG. 2 shows an example of an IMG after division by the face image division unit 15. The IMG in FIG. 2 is an example of a face image showing a face facing forward.

[0026] In the example of Fig. 2, the facial image division unit 15 divides the IMG into 10 parts each in the vertical and horizontal directions (e.g., 10 equal parts each). That is, the facial image division unit 15 divides the IMG into 100 partial regions (partial regions 1 to 100). However, the method of dividing the IMG by the facial image division unit 15 is not limited to the example of Fig. 2. For example, the size of each partial region does not necessarily have to be the same.

[0027] The skin region extraction unit 16 extracts a skin region (a region in which at least a part of the skin is shown) from each partial region. A skin region is a region in which the skin is not completely covered by a covering (e.g., hair or sunglasses). A skin region can also be expressed as a region in which a pulse wave can be detected (calculated). In the example of FIG. 2, the skin region is shown as an unshaded region among the partial regions. In the example of FIG. 2, the skin region extraction unit 16 extracts 52 skin regions (52 locations) from the 100 partial regions.

[0028] Pulse wave calculation unit 17 calculates a pulse wave (more precisely, a pulse wave signal) for each skin region extracted by skin region extraction unit 16. A known method (e.g., a method using independent component analysis) may be applied as a method for calculating a pulse wave in pulse wave calculation unit 17. As an example, when face image division unit 15 is not provided, one pulse wave may be calculated by pulse wave calculation unit 17. It is sufficient for pulse wave calculation unit 17 to calculate at least one pulse wave.

[0029] Pulse wave parameter calculation section 20 calculates pulse wave parameters based on the pulse wave of each skin region calculated by pulse wave calculation section 17. In this specification, "pulse wave parameters" collectively refers to explanatory variables (also referred to as independent variables) used in measuring (calculating) blood pressure based on a measurement model. Pulse wave parameter calculation section 20 may calculate at least one pulse wave parameter based on at least one pulse wave.

[0030] An example of a pulse wave parameter is the pulse wave transit time (PTT) between each skin region. In the first embodiment, the pulse wave parameter calculation unit 20 calculates the PTT based on the pulse wave of each skin region using a known method. The PTT between region A (any one skin region) and region B (a skin region other than region A) is also represented as PTT(AB). For example, the PTT between regions 23 and 24 in FIG. 2 is represented as PTT(23-24).

[0031] In the example of FIG. 2, the pulse wave parameter calculation unit 20 selects a combination of any two skin regions from the 52 skin regions. 52 C2 = 1326 combinations are selected. Pulse wave parameter calculation section 20 then calculates the PTT for each combination. In this manner, pulse wave parameter calculation section 20 calculates 1326 PTTs, that is, PTT(23-24) to PTT(96-97).

[0032] Another example of the pulse wave parameter is the waveform feature of the pulse wave in each skin region. Thus, the pulse wave parameter calculation unit 20 may calculate the waveform feature using a known method. As an example, the waveform feature may be calculated based on (i) the pulse wave waveform, (ii) the velocity pulse wave waveform (a waveform obtained by differentiating the pulse wave signal once), or (iii) the acceleration pulse wave waveform (a waveform obtained by differentiating the pulse wave signal twice).

[0033] As an example, the pulse wave parameter calculation section 20 may derive an acceleration pulse wave waveform from the pulse wave signal and calculate the waveform feature amount by analyzing the acceleration pulse wave waveform. An example of an acceleration pulse wave waveform is shown in FIG. 3, which will be described later. A to D in FIG. 3 show feature points of the acceleration pulse wave waveform. Specifically, Characteristic point a: the first maximum point of the accelerated plethysmogram waveform; · Feature point b: the first minimum point of the accelerated plethysmogram waveform; Characteristic point c: the second maximum point of the accelerated plethysmogram waveform; · Characteristic point d: the second minimum point of the accelerated plethysmogram waveform; In the following explanation, for simplicity, the amplitude of the accelerated pulse waveform at characteristic point a will be simply referred to as "amplitude a" (or simply "a"). The same applies to b to d.

[0034] As one example, pulse wave parameter calculation section 20 may (i) calculate each amplitude (a to d) as the waveform feature amount. Alternatively, pulse wave parameter calculation section 20 may calculate the ratio of each amplitude (e.g., b / a) as the waveform feature amount (see also FIG. 4). Pulse wave parameter calculation section 20 may also calculate the time difference between each feature point (e.g., the time difference between feature point a and feature point b) as the waveform feature amount.

[0035] (Vascular age calculation unit 21) The vascular age calculation unit 21 calculates the vascular age (hereinafter simply referred to as vascular age) of the subject H based on the pulse wave calculated by the pulse wave calculation unit 17. For example, the vascular age calculation unit 21 calculates the vascular age by analyzing the pulse wave. A known method (for example, see Patent Document 2) may be used to calculate the vascular age.

[0036] The information indicating the vascular age (vascular age information) is an example of attribute information of the subject H (hereinafter simply referred to as attribute information). In this specification, attribute information collectively means information related to the vascular condition of the subject H (vascular condition related information). The vascular condition related information is an example of information related to a characteristic (more specifically, constitution) unique to the subject H. In this specification, the functional unit that acquires the attribute information is collectively referred to as an attribute information acquisition unit. Therefore, the vascular age calculation unit 21 is an example of an attribute information acquisition unit.

[0037] FIG. 3 is a graph showing an example of the relationship between vascular age and accelerated pulse wave waveform (see also Patent Document 2). FIG. 3(a) shows an example of an accelerated pulse wave waveform obtained from a subject with a low (young) vascular age. In contrast, FIG. 3(b) shows an example of an accelerated pulse wave waveform obtained from a subject with a high (old) vascular age. Specifically, a subject with a low vascular age refers to a subject in which arteriosclerosis has not progressed much. On the other hand, a subject with a high vascular age refers to a subject in which arteriosclerosis has progressed to a certain degree.

[0038] It is known that the accelerated pulse waveform changes according to the vascular age. As shown in Fig. 3(a), when the vascular age is young, the amplitude b is relatively large and the amplitude d is relatively small. In contrast, as shown in Fig. 3(b), as the vascular age increases, the amplitude b becomes smaller and the amplitude d becomes larger.

[0039] Therefore, the vascular age calculation unit 21 calculates the following waveform index (WI): WI=d / ab / a The vascular age may be calculated based on the WI. The WI is one of the effective indicators of the vascular age. When the vascular age calculation unit 21 calculates the WI in a plurality of skin regions, the vascular age calculation unit 21 may calculate the vascular age using a representative value (e.g., an average value or a median value) of each WI.

[0040] (Gender detection unit 22) The gender detection unit 22 detects (determines) the gender of the subject H. As an example, the gender detection unit 22 determines the gender of the subject H by analyzing the IMG. In the first embodiment, the gender detection unit 22 uses a known deep learning technique to identify whether the gender of the subject H shown in the IMG is male or female. Information indicating the gender of the subject H (gender information) is another example of attribute information. Therefore, the gender detection unit 22 is another example of an attribute information acquisition unit.

[0041] (Attribute classification section 23) The attribute classification unit 23 classifies (detects) the attributes (hereinafter simply referred to as attributes) of the subject H based on the attribute information. In this specification, the attribute means an "attribute according to the blood vessel state." Specifically, the attribute classification unit 23 determines to which of a plurality of preset attribute-related patterns the attribute belongs. Hereinafter, the total number of patterns is represented as N, where N is any integer equal to or greater than 2. Each pattern is also referred to as pattern k, where k is an integer satisfying 1≦k≦N.

[0042] As an example, the attribute classification unit 23 may classify the attributes based on the vascular age and gender of the subject H. For example, the attribute classification unit 23 may classify a plurality of subjects H according to the vascular age generation and gender. In this case, for example, the attribute classification unit 23 may classify the attributes based on the vascular age generation and gender of the subject H. Subject HA characteristics: Vascular age: male in his 20s; Subject HB characteristics: Female with vascular age in her 30s; As such, the attributes of two different subjects (subjects HA and HB) can be classified into different patterns.

[0043] However, the method of attribute classification by attribute classification unit 23 is not limited to the above example. For example, attribute classification unit 23 may classify attributes based on the waveform feature amount calculated by pulse wave parameter calculation unit 20 (one example of the analysis result of the pulse wave by pulse wave parameter calculation unit 20).

[0044] In this example, it can be said that the pulse wave parameter calculation unit 20 is used as an attribute information acquisition unit. In other words, information indicating waveform feature amounts (waveform feature amount information) can also be used as attribute information. In this way, it should be noted that the attribute information is not limited to vascular age information and gender information.

[0045] As an example, the inventors of the present application (hereinafter simply the inventors) calculated the amplitude ratio "b / a" of the accelerated pulse wave waveform as a waveform feature using the pulse wave parameter calculation unit 20. Then, the inventors classified 10 subjects based on the waveform feature using the attribute classification unit 23. Figure 4 is a table showing an example of the classification result. In the table of Figure 4, the amplitude ratios of each subject are arranged in descending order.

[0046] In the example of FIG. 4, the attribute classification unit 23 classifies each object into two types by comparing the amplitude ratio of the object with a predetermined threshold. In the example of FIG. 4, the threshold is set to "-0.600". Specifically, when the amplitude ratio of a certain object is equal to or greater than the threshold, the attribute classification unit 23 classifies the attribute of the object as "attribute 1". On the other hand, when the amplitude ratio of a certain object is less than the threshold, the attribute classification unit 23 classifies the attribute of the object as "attribute 2". Attributes 1 and 2 are examples of patterns 1 and 2, respectively. Attribute k is also referred to as the "kth attribute".

[0047] Next, the inventors created estimation models (for convenience, referred to as attribute-specific models) according to attributes 1 and 2, respectively, by a method described below. The estimation models are computational models for calculating (estimating) blood pressure.

[0048] In addition, the inventors created a conventional estimation model for comparison with the attribute-specific model. The conventional estimation model differs from the attribute-specific model in that it is not created for each attribute (it is common to all attributes). For this reason, the conventional estimation model is also called a common model.

[0049] The inventors then performed a performance comparison between the common model and the attribute-specific models. Figure 5 shows an example of the performance comparison result. Figure 5(a) is a graph showing an example of the relationship between the true blood pressure value and the predicted blood pressure value in the common model. In contrast, Figure 5(b) is a graph showing an example of the relationship between the true blood pressure value and the predicted blood pressure value in the attribute-specific model.

[0050] Note that the true blood pressure value in FIG. 5 is the blood pressure (i.e., BPm) measured by a cuff sphygmomanometer (blood pressure acquisition unit 2). BPm in FIG. 5 is an example of an actual measurement value (true value) of blood pressure. In contrast, the predicted blood pressure value in FIG. 5 is the blood pressure (i.e., BPe) calculated by the model evaluation unit 40 (described later) using the measurement model. BPe in FIG. 5 is an example of a predicted value of blood pressure.

[0051] The inventors calculated the standard deviation between the true blood pressure value and the predicted blood pressure value (i.e., the standard deviation of the error) for each of the common model and the attribute-specific model using the model evaluation unit 40. As a result, the standard deviation for the common model was 12.23 mmHg. In contrast, the standard deviation for the attribute-specific model was 9.25 mmHg.

[0052] The inventors also calculated the mean square error (MSE) between the true blood pressure value and the predicted blood pressure value for each of the common model and the attribute-specific model using the model evaluation unit 40. As a result, the MSE for the common model was 149.64 mmHg. In contrast, the MSE for the attribute-specific model was 122.04 mmHg.

[0053] As described above, it has been confirmed that the attribute-specific model can reduce errors compared to the common model. In this way, the attribute-specific model can improve the blood pressure measurement accuracy compared to the common model.

[0054] (supplement) In the example of FIG. 4, N is not limited to 2. When the total number of subjects to be classified is NT, N may be any natural number N satisfying 2≦N≦NT-1. In the example of FIG. 4, NT=10, so 2≦N≦9. In this case, by setting N-1 thresholds from the first threshold to the N-1th threshold for the amplitude ratio, NT subjects can be classified into N patterns (first attribute to N-th attribute). As an example, the magnitude relationship of each threshold may be set in the order of threshold numbers as "first threshold>second threshold>...>N-1th threshold".

[0055] (Model Creation Section 30) The model creation section 30 creates an estimation model. Specifically, the model creation section 30 creates the estimation model by using (i) the blood pressure (BPm) of the subject H acquired by the blood pressure acquisition section 2 and (ii) the pulse wave parameters calculated by the pulse wave parameter calculation section 20 as training (learning) data. The pulse wave parameters may be at least one of the PTT and the waveform feature amount.

[0056] 1, the model creation unit 30 has a first model creation unit 300-1, a second model creation unit 300-2, ..., and an N-th model creation unit 300-N. The k-th model creation unit 300-k creates an estimation model according to the k-th attribute (pattern k). k is an integer satisfying 1≦k≦N. In this way, the model creation unit 30 can create an estimation model according to each attribute.

[0057] In this specification, for convenience, the first model creation unit 300-1 to the N-th model creation unit 300-N are also collectively referred to as the model creation unit 30. The description of the model creation unit 30 applies to any k-th model creation unit 300-k. For the same purpose, in this specification, the first model evaluation predicted blood pressure calculation unit 410-1 to the N-th model evaluation predicted blood pressure calculation unit 410-N described later are also collectively referred to as the evaluation predicted blood pressure calculation unit 41. In addition, the first model evaluation index calculation unit 420-1 to the N-th model evaluation index calculation unit 420-N described later are also collectively referred to as the model evaluation index calculation unit 42. Similarly, the first model selection unit 600-1 to the N-th model selection unit 600-N described later are also collectively referred to as the model selection unit 60.

[0058] (An example of how to create an estimation model) The velocity v at which the pulse wave propagates through the blood vessels is given by the Moens-Korteweg equation, i.e.,

[0059]

number

[0060] In equation (1), E is the Young's modulus of blood vessel, a is the blood vessel wall pressure, R is the blood vessel diameter, and ρ is the blood density.

[0061] It is known that the Young's modulus E of blood vessels changes exponentially with respect to blood pressure P. Therefore, if the Young's modulus of blood vessels at P=0 is E0, E is expressed as follows:

[0062]

number

[0063] where γ is a vessel-dependent constant.

[0064] The length of the blood vessel path, L, is

[0065]

number

[0066] where T is the pulse transit time (PTT) and L is the vascular path length.

[0067] Therefore, from equations (1) to (3),

[0068]

number

[0069] is derived.

[0070] As shown in equation (4), when L is constant, T has a correlation with P. For this reason, model creation section 30 may create at least one estimation model of P using the PTT (an example of a pulse wave parameter) calculated by pulse wave parameter calculation section 20.

[0071] In the following description, for simplicity, a case where only PTT is used as a pulse wave parameter is illustrated. However, as described above, an estimation model can be created using only waveform features instead of PTT. Alternatively, an estimation model can be created using both PTT and waveform features.

[0072] First, the model creation unit 30 creates an estimation model M1 with a complexity level of 1. In this specification, "complexity" refers to the number of explanatory variables in the estimation model (e.g., the number of pulse wave parameters used in the estimation model). In the following example, the estimation model M1 uses one PTT as an explanatory variable.

[0073] In the following description, one PTT calculated by pulse wave parameter calculation unit 20 is represented as PTT1. PTT1 is the PTT between any two skin regions. Model creation unit 30 performs regression analysis using the least squares method on PTT1 and BPm. Model creation unit 30 creates an estimated model M1 as a result of the regression analysis. Each PTT calculated by pulse wave parameter calculation unit 20 and BPm acquired by blood pressure acquisition unit 2 are both examples of training (learning) data.

[0074] As an example, the estimation model M1 is BP1 = α1 × PTT1 + α2 … (5) Consider the case of a linear model (a computational model expressed by a linear function) represented by the following equation (5): In equation (5), BP1 is a predicted blood pressure, and α1 and α2 are constants. In this case, the model creation unit 30 calculates α1 and α2 by performing regression analysis (i.e., creates an estimation model M1).

[0075] For convenience, the 1326 PTTs in the example of Fig. 2 will be referred to as PTT1-1 to PTT1-1326. The model creation unit 30 uses PTT1-1 to PTT1-1326 to create the same number of estimation models M1 as the PTTs. For convenience, these estimation models M1 will be referred to as M1-1 to M1-1326.

[0076] Next, model creation unit 30 creates an estimation model M2 with a complexity level of 2. In estimation model M2, two PTTs are used as explanatory variables. In the following description, the two different PTTs calculated by pulse wave parameter calculation unit 20 are represented as PTT1 and PTT2.

[0077] The model creation unit 30 performs a regression analysis using the least squares method on (i) PTT1 and PTT2, and (ii) BPm. The model creation unit 30 creates an estimated model M2 as a result of the regression analysis.

[0078] As an example, the estimated model M2 is BP2=β1×PTT1+β2×PTT2+β3 …(6) In equation (6), BP2 is a predicted blood pressure, and β1 to β3 are constants. In this case, the model creation unit 30 calculates β1 to β3 by performing regression analysis.

[0079] In the example of FIG. 2, the model creation unit 30 uses PTT1-1 to PTT1-1326 to create an estimation model M2 that is larger than the number of PTTs. In this example, the number of combinations of PTT1 and PTT2 is 878475 (i.e., 1326 Therefore, the model creation unit 30 calculates 878475 estimated models M2.

[0080] As described above, the estimation model M2 can be created using both the PTT and the waveform feature amount. In this case, for example, the estimation model M2 may be created using the PTT as the first explanatory variable and the waveform feature amount as the second explanatory variable.

[0081] Similarly, the model creation unit 30 creates an estimated model M3 with a complexity level of 3, an estimated model M4 with a complexity level of 4, ..., an estimated model Mz with a complexity level of z. z indicates the upper limit of the complexity level. z may be appropriately set by the manufacturer of the model setting device 100. The model creation unit 30 supplies each of the created estimated models to the model evaluation unit 40 (more specifically, the evaluation predicted blood pressure calculation unit 41).

[0082] As described above, the first model creation unit 300-1 creates an estimation model (hereinafter, the first model) according to attribute 1. The same applies to the second model creation unit 300-2 to the N-th model creation unit 300-N. That is, the k-th model creation unit 300-k creates an estimation model (hereinafter, the k-th model) according to attribute k.

[0083] In this manner, the model creation unit 30 creates the first model to the Nth model. Hereinafter, the first model to the Nth model are collectively referred to as an estimated model group. In the example of Fig. 1, the model creation unit 30 supplies the created estimated model group to the model evaluation unit 40 and the model storage unit 55, respectively.

[0084] (Model Evaluation Section 40) The model evaluation unit 40 evaluates each estimation model created in the model creation unit 30, and outputs the evaluation result. Specifically, the model evaluation unit 40 outputs a PI described below as the evaluation result. In the example of Fig. 1, the model evaluation unit 40 directly acquires the estimation model group from the model creation unit 30. However, the model evaluation unit 40 may acquire the estimation model group stored in advance in the model storage unit 55.

[0085] The model evaluation unit 40 has an evaluation predicted blood pressure calculation unit 41 and a model evaluation index calculation unit 42. The evaluation predicted blood pressure calculation unit 41 has a first model evaluation predicted blood pressure calculation unit 410-1, a second model evaluation predicted blood pressure calculation unit 410-2, ..., and an N-th model evaluation predicted blood pressure calculation unit 410-N. The model evaluation index calculation unit 42 has a first model evaluation index calculation unit 420-1, a second model evaluation index calculation unit 420-2, ..., and an N-th model evaluation index calculation unit 420-N.

[0086] The k-th model evaluation predicted blood pressure calculation unit 410-k and the k-th model evaluation index calculation unit 420-k are functional units corresponding to the k-th model. The k-th model evaluation predicted blood pressure calculation unit 410-k and the k-th model evaluation index calculation unit 420-k are collectively referred to as the "k-th model evaluation unit."

[0087] The evaluation predicted blood pressure calculation unit 41 calculates a predicted blood pressure (hereinafter, BPe) in the estimation model created by the model creation unit 30. Specifically, the evaluation predicted blood pressure calculation unit 41 calculates BPe by applying (specifically, substituting) the pulse wave parameters calculated by the pulse wave parameter calculation unit 20 as test data to the estimation model.

[0088] The model evaluation index calculation unit 42 calculates an evaluation index (hereinafter, PI) of the above-mentioned estimation model. The model evaluation index calculation unit 42 may calculate the PI based on BPm and BPe. As an example, the model evaluation index calculation unit 42 calculates the MSE of BPm and BPe as the PI. The model evaluation index calculation unit 42 calculates the PI of each estimation model in order from the least complex estimation model. Then, the model evaluation index calculation unit 42 supplies the calculated PI to the model storage unit 55.

[0089] In addition, the PI (evaluation index) is not limited to MSE. The PI may be any as long as it can be calculated based on BPm and BSe (see also the second embodiment described later). As an example, the average of the errors between BPm and BSe (e.g., mean absolute error) may be used as the PI. Alternatively, the standard deviation of the errors between BPm and BSe may be used as the PI.

[0090] Alternatively, a number of predetermined parameters (numerical values) may be calculated based on BPm and BPe, and the number of parameters may be ranked (e.g., the superiority or inferiority of the number of parameters may be determined). In this case, the number indicating the rank of each parameter may be used as PI.

[0091] As explained above, the first model evaluation predicted blood pressure calculation unit 410-1 calculates the predicted blood pressure in the first model (first model predicted blood pressure). The same applies to the second model evaluation predicted blood pressure calculation unit 410-2 to the N-th model evaluation predicted blood pressure calculation unit 410-N. That is, the k-th model evaluation predicted blood pressure calculation unit 410-k calculates the k-th model predicted blood pressure (hereinafter, BPek). In this manner, the evaluation predicted blood pressure calculation unit 41 calculates BPe1 to BPeN. Hereinafter, BPe1 to BPeN are also collectively referred to as a predicted blood pressure group. The evaluation predicted blood pressure calculation unit 41 supplies the calculated predicted blood pressure group to the model evaluation index calculation unit 42.

[0092] Next, the first model evaluation index calculation unit 420-1 calculates each PI (first model evaluation index set) in the first model. The same is true for the second model evaluation index calculation unit 420-2 to the N-th model evaluation index calculation unit 420-N. That is, the k-th model evaluation index calculation unit 420-k calculates the k-th model evaluation index set (hereinafter, PIk).

[0093] Hereinafter, PI1 to PIk are collectively referred to as a group of evaluation index sets. The model evaluation index calculation unit 42 supplies the calculated group of evaluation index sets to the model storage unit 55 in association with a group of estimation models.

[0094] (Setting of measurement model by model evaluation unit 40) The model evaluation unit 40 sets (selects) at least one measurement model based on the evaluation results (each PI) by the model evaluation unit 40 (more specifically, the model evaluation index calculation unit 42). Specifically, the model evaluation unit 40 selects at least one measurement model from at least one estimation model stored in the model storage unit 55. As an example, a case in which one measurement model is selected will be described with reference to FIG. 6. FIG. 6 is a diagram for explaining an example of a process for setting a measurement model in the model evaluation unit.

[0095] As shown in Fig. 6, when blood pressure estimation models with the smallest MSE (an example of PI) at each complexity are plotted against each other, the model evaluation unit 40 selects an estimation model with a minimum MSE as a measurement model. In the example of Fig. 6, a given estimation model with a complexity of 3 is selected as the measurement model (see the legend of the stars in Fig. 6).

[0096] Furthermore, in the blood pressure measurement device 1, it is preferable that the data (training data) for creating the blood pressure estimation model in the model creation unit 30 is different from the data (test data) for evaluating the blood pressure estimation model in the model evaluation unit 40. In this case, the model evaluation unit 40 can select a measurement model with excellent generalization performance that fits well to the test data without falling into overlearning.

[0097] The method of setting the measurement model is not limited to the above example. For example, the model evaluation unit 40 may extract, as a measurement model candidate, an "estimation model whose PI (e.g., MSE) is equal to or less than a predetermined threshold value" from at least one estimation model. Then, the model evaluation unit 40 may select at least one measurement model from the measurement model candidates. As an example, the model evaluation unit 40 may select, as the measurement model, an estimation model whose PI is the smallest from among the measurement model candidates. Alternatively, the model evaluation unit 40 may select, as the measurement model, an estimation model whose complexity is the smallest from among the measurement model candidates.

[0098] As described above, the model evaluation unit 40 selects at least one measurement model (measurement model in the first model) from at least one first model. The same applies to the second model to the Nth model. That is, the model evaluation unit 40 identifies at least one measurement model (measurement model in the kth model) from at least one kth model. The measurement model in the kth model is a calculation model for measuring blood pressure (P) in the blood pressure measurement unit 160 in the case of attribute k.

[0099] In addition, when only one k-th model is created in the k-th model creation unit 300-k, the model evaluation unit 40 may select the k-th model as the measurement model (measurement model in the k-th model). Hereinafter, the measurement models in the first model to the N-th model are collectively referred to as a measurement model group. The model evaluation unit 40 supplies the set measurement model group to the model storage unit 55.

[0100] (Model Storage Section 55) The model storage unit 55 may be a known storage device capable of storing each data. In the first embodiment, the model storage unit 55 stores (i) a group of estimation models created by the model creation unit 30. In addition, the model storage unit 55 preferably further stores (i) a group of evaluation index sets calculated by the model evaluation index calculation unit 42, and (ii) a group of measurement models set by the model evaluation unit 40.

[0101] (Measuring model setting method) Fig. 7 is a flowchart showing an example of the flow of processing by the model setting device 100. Fig. 7 shows an example of a method for setting a measurement model by the model setting device 100 (measurement model setting method).

[0102] First, the imaging unit 11 captures an image of the subject (S1). The face image acquisition unit 14 acquires a face image (IMG) from the captured image of the subject (S2). The face image division unit 15 divides the IMG into a plurality of partial regions (S3). The skin region extraction unit 16 extracts skin regions from the plurality of partial regions (S4). The pulse wave calculation unit 17 calculates a pulse wave (pulse wave signal) for each of the skin regions (S5). S1 to S5 are collectively referred to as a pulse wave acquisition step.

[0103] Next, pulse wave parameter calculation section 20 calculates pulse wave parameters based on the pulse wave. First, pulse wave parameter calculation section 20 calculates the PTT (pulse wave transit time) between each skin region (S6). Next, pulse wave parameter calculation section 20 calculates waveform feature amounts for each skin region (S7). S6 and S7 are collectively referred to as the pulse wave parameter calculation step.

[0104] Next, the gender detection unit 22 detects the gender of the subject H by analyzing the IMG (S8). The vascular age calculation unit 21 calculates the vascular age of the subject H based on the pulse wave (S9). S8 and S9 are also collectively referred to as an attribute information acquisition process.

[0105] Then, the attribute classification unit 23 classifies the attributes of the subject H based on the attribute information (S10, attribute classification step). In this example, the attribute classification unit 23 classifies the attributes of the subject H based on the blood vessel age information and gender information. For example, the attribute classification unit 23 classifies the attribute of the subject H into one attribute (attribute k) of attributes 1 to N.

[0106] Each process described below is performed for each attribute k. First, the blood pressure acquiring unit 2 acquires the blood pressure (BPm, second blood pressure) of the subject H (S11, second blood pressure acquiring step).

[0107] Next, the model creation unit 30 (more specifically, the k-th model creation unit 300-k) creates at least one estimation model (k-th model) for attribute k using the training data. Specifically, the model creation unit 30 creates each estimation model using the pulse wave parameters and BPm (S12, model creation step). Note that BPm used in S12 is the blood pressure measured by the blood pressure acquisition unit 2 simultaneously with the imaging of the subject image (S1). In other words, the second blood pressure measurement step has been executed once in advance, simultaneously with S1, prior to S11.

[0108] Next, the evaluation predicted blood pressure calculation unit 41 (more specifically, the k-th model evaluation predicted blood pressure calculation unit 410-k) calculates the predicted blood pressure (BPe, more specifically, BPek) in each estimation model using the test data. Specifically, the evaluation predicted blood pressure calculation unit 41 calculates BPe by applying the pulse wave parameters to each estimation model (S13).

[0109] Next, the model evaluation index calculation unit 42 (more specifically, the k-th model evaluation index calculation unit 420-k) calculates an evaluation index (PI, more specifically, PIk) of the estimated model. Specifically, the model evaluation index calculation unit 42 calculates the mean square error (MSE) between BPe and BPm as PI (S14). S13 and S14 are also collectively referred to as a model evaluation step.

[0110] Then, the model evaluation unit 40 selects at least one measurement model (measurement model in the k-th model) from each estimation model based on the evaluation result (each PI) by the model evaluation unit 40. For example, the model evaluation unit 40 sets the model with the smallest MSE among the k-th models as the measurement model in the k-th model (S15, model setting step). By executing S11 to S15 for each of the classes 1 to N, the model setting device 100 can set the first intra-model measurement model to the Nth intra-model measurement model.

[0111] (Pulse wave signal quality evaluation unit 150) Next, we will describe the remaining functional units of the blood pressure measurement device 1. The pulse wave signal quality evaluation unit 150 evaluates the quality of the pulse wave signal of each skin area, which is used when the blood pressure (P) is measured by the blood pressure measurement device 1 (more specifically, the blood pressure measurement unit 160). As an example, the pulse wave signal quality evaluation unit 150 calculates the SNR of the pulse wave signal as an index of the quality of the pulse wave signal.

[0112] 8 is a graph showing an example of the power spectrum of a pulse wave signal (hereinafter simply referred to as the power spectrum). In this graph, the horizontal axis represents frequency and the vertical axis represents the power of the pulse wave signal. Pulse wave signal quality evaluation section 150 derives the power spectrum by performing frequency analysis on the pulse wave signal.

[0113] A pulse wave is a wave that is transmitted to an artery by the pumping action of the heart. Therefore, a pulse wave signal has a constant period that corresponds to the heartbeat. In many cases, when the subject H is at rest, a peak in the power spectrum is confirmed in a frequency band of about 1 Hz. The PR (Pulse Rate) in FIG. 8 is an example of such a peak.

[0114] Therefore, pulse wave signal quality evaluation section 150 may calculate signal components (Signal) and noise components (Noise) within a predetermined bandwidth. As an example, pulse wave signal quality evaluation section 150 may set a frequency band of ±0.05 Hz centered on PR as the signal band. Pulse wave signal quality evaluation section 150 then calculates the power sum of the power spectrum in that signal band as Signal. In contrast, pulse wave signal quality evaluation section 150 may set a frequency band of 0.75 to 4.0 Hz excluding the Signal band as the noise band. Pulse wave signal quality evaluation section 150 then calculates the power sum of the power spectrum in that noise band as Noise. Pulse wave signal quality evaluation section 150 then calculates the SNR as SNR=Signal / Noise.

[0115] However, there may be cases where a pulse wave having a certain signal quality (high-precision pulse wave) cannot be obtained from some skin regions. Examples of such skin regions include (i) skin regions that are partially covered by a covering, or (ii) skin regions that are cast in shadow. Therefore, from the viewpoint of improving the accuracy of blood pressure measurement, it is preferable to take into consideration the presence of such skin regions using the results of the quality evaluation of the pulse wave signal by pulse wave signal quality evaluation section 150.

[0116] As an example, the pulse wave signal quality evaluation unit 150 may classify each skin region into (i) a region where a pulse wave having a certain signal quality was acquired (hereinafter, quality-conforming region) and (ii) other regions (hereinafter, quality-unconforming region). A quality-unconformable region can also be expressed as a region where a pulse wave having a certain signal quality was not acquired. As an example, consider a case where "certain signal quality" can be expressed as "SNR>0.15". In this case, the pulse wave signal quality evaluation unit 150 identifies, among each skin region, a region where SNR>0.15 is the quality-conformable region.

[0117] (Model Selection Unit 60) The model selection unit 60 has a first model selection unit 600-1, a second model selection unit 600-2, ..., and an N-th model selection unit 600-N. The k-th model selection unit 600-k is a functional unit corresponding to the k-th model. The model selection unit 60 operates to measure blood pressure (P) in the blood pressure measurement unit 160 after the processing by the model setting device 100 is completed.

[0118] The model selection unit 60 reads out a group of measurement models set in advance by the model setting device 100 (model evaluation unit 40) from the model storage unit 55. Then, the model selection unit 60 selects a measurement model from the group of measurement models according to the attribute of the subject H classified by the attribute classification unit 23. That is, the model selection unit 60 selects a measurement model according to the attribute k (pattern k). The measurement model selected by the model selection unit 60 is at least one measurement model (blood pressure measurement model) used for measuring blood pressure in the blood pressure measurement unit 160. In the following explanation, a case where there is one blood pressure measurement model will be illustrated as an example.

[0119] Specifically, the first model selection unit 600-1 selects one measurement model (first model for blood pressure measurement) from at least one measurement model (measurement model in the first model) according to the attribute 1. The same applies to the second model selection unit 600-2 to the N-th model selection unit 600-N. That is, the k-th model selection unit 600-k selects one measurement model (k-th model for blood pressure measurement) from at least one measurement model in the k-th model.

[0120] As described above, the model selection unit 60 selects a measurement model from at least one blood pressure estimation model based on the evaluation results (each PI) of at least one estimation model.

[0121] When only one measurement model in the k-th model is created in the model evaluation unit 40, the k-th model selection unit 600-k may select the one measurement model in the k-th model as the k-th model for blood pressure measurement. Therefore, it should be noted that the pulse wave signal quality evaluation unit 150 is not an essential component of the blood pressure measurement device 1.

[0122] However, in order to improve the accuracy of the measurement result (P) of the blood pressure measurement device 1, it is preferable to provide a pulse wave signal quality evaluation unit 150. For example, the model selection unit 60 may select a measurement model for blood pressure measurement based on the result of the quality evaluation of the pulse wave signal by the pulse wave signal quality evaluation unit 150. For example, the k-th model selection unit 600-k may select a model having the highest quality of the pulse wave signal from at least one measurement model in the k-th model as the k-th model for blood pressure measurement.

[0123] Furthermore, the model selection unit 60 may extract only models in which only the quality conformance region is used as candidates for blood pressure measurement models from the group of measurement models. This makes it possible to more reliably prevent a decrease in the measurement accuracy of the blood pressure measurement device 1.

[0124] (Blood pressure measurement unit 160 and blood pressure measurement result output unit 170) The blood pressure measurement unit 160 measures blood pressure (P) based on the pulse wave parameters using an estimated model (kth model) according to attribute k. More specifically, the blood pressure measurement unit 160 measures P using a blood pressure measurement model selected by the model selection unit 60. That is, the blood pressure measurement unit 160 calculates P by applying the pulse wave parameters calculated by the pulse wave parameter calculation unit 20 to the blood pressure measurement model. In this way, the blood pressure measurement unit 160 calculates P based on the pulse wave parameters by using the blood pressure measurement model.

[0125] As described above, the model selection unit 60 selects a blood pressure measurement model (the kth blood pressure measurement model) according to the attribute k. Therefore, the blood pressure measurement unit 160 can calculate P using a blood pressure measurement model suitable for the attribute of the subject H.

[0126] The blood pressure measurement result output section 170 acquires P measured by the blood pressure measurement section 160. Then, the blood pressure measurement result output section 170 outputs the P as a blood pressure measurement result. The blood pressure measurement result output section 170 may present P in any notification manner. As an example, the blood pressure measurement result output section 170 may be a display. In this case, the blood pressure measurement result output section 170 can visually present the blood pressure measurement result to the subject H by displaying a numerical value indicating P.

[0127] The blood pressure measurement result output unit 170 may display at least a part of the various attribute information (e.g., vascular age information) together with the numerical value indicating P. In this case, it is possible to indicate to the subject H that the blood pressure measurement is being performed according to the attributes of the subject H.

[0128] (Blood pressure measurement method) An example of a method for measuring blood pressure by the blood pressure measurement device 1 (blood pressure measurement method) will be described below. Each process of the blood pressure measurement method is executed after all the processes in Fig. 7 are completed. Therefore, in the following example, it is assumed that the estimation model group, evaluation index set group, and measurement model group derived by the model setting device 100 are stored in the model storage unit 55 in advance, prior to the start of each process of the blood pressure measurement method.

[0129] First, in the blood pressure measurement method, the same processes as S1 to S10 in Fig. 7 are executed. That is, in the blood pressure measurement method, like the model setting method, a pulse wave acquiring step, a pulse wave parameter calculating step, an attribute information acquiring step, and an attribute classifying step are executed.

[0130] Thereafter, as described above, the blood pressure measurement unit 160 calculates P based on the pulse wave parameters using the kth model (first blood pressure measurement step). More specifically, prior to the first blood pressure measurement step, the model selection unit 60 selects a measurement model according to the attribute k (more strictly, the kth model for blood pressure measurement) (model selection step). Next, in the first blood pressure measurement step, the blood pressure measurement unit 160 calculates P using the measurement model.

[0131] (effect) Unlike a contact type blood pressure measuring device (e.g., blood pressure measuring device 1), a non-contact type blood pressure measuring device cannot directly acquire the blood pressure of the subject H as a physical quantity, unlike a contact type blood pressure measuring device (e.g., blood pressure acquiring unit 2 as a cuff sphygmomanometer). For this reason, a non-contact type blood pressure measuring device needs to be provided with a model for deriving the blood pressure (P) from biological information (e.g., pulse wave parameters) of the subject H.

[0132] However, as shown in FIG. 9, it is known that the correlation between blood pressure and PWV (Pulse Wave Velocity) may differ significantly depending on factors such as the age and sex of the subject H. FIG. 9 is a graph showing an example of the relationship between blood pressure and PWV for each of subjects having different attributes. As an example, FIG. 9 shows an example of the relationship between blood pressure and PWV for each of "a man with a certain vascular age (vascular age A)", "a man with another vascular age (vascular age B)", and "a woman with yet another vascular age (vascular age C)". It should be noted that PWV has a negative correlation with PTT. For this reason, PWV in FIG. 9 may be read as PTT.

[0133] For this reason, one model suitable for measuring the blood pressure of a certain subject (e.g., a male with vascular age A) is not necessarily suitable for measuring the blood pressure of another subject (e.g., a male with vascular age B or a female with vascular age C). Therefore, when the above-mentioned common model is used, the individual differences (e.g., age and sex) of the subject H cannot be taken into consideration, and therefore the measurement accuracy of P cannot be sufficiently improved.

[0134] In consideration of the above, the inventors have come up with a new configuration in which "P is measured using an individual measurement model set according to the attributes of the subject H." More essentially, the inventors have come up with a new configuration in which "at least one type of estimation model is created according to each attribute of the subject H." With these configurations, it becomes possible to perform blood pressure measurement taking into account the individual differences of the subject H, thereby improving the measurement accuracy of P compared to the conventional method (see also FIG. 5).

[0135] Incidentally, when creating at least one type of estimation model corresponding to each attribute of the subject H, it is also possible to classify the attributes based on the actual age of the subject H (hereinafter simply referred to as actual age) (see Patent Document 1). However, it is also considered that the actual age is not necessarily sufficient as an index showing the vascular condition of the subject H. For example, depending on the lifestyle habits (e.g., eating habits and exercise habits) of the subject H, the vascular age may be high even if the actual age is low (or vice versa).

[0136] Considering this, in order to improve the measurement accuracy of P, it is preferable to use vascular age instead of actual age as an index indicating the vascular condition of the subject H. Therefore, for example, in the example of Fig. 7, attribute classification based on vascular age is performed. By performing attribute classification in this manner, an estimation model that is more in line with the actual vascular condition of the subject H can be created compared to the case where attribute classification based on actual age is performed.

[0137] Moreover, in the model setting device 100, the attribute information can be acquired by, for example, analyzing IMG. This makes it unnecessary for the user of the blood pressure measurement device 1 to manually select a measurement model according to the subject H. It also makes it unnecessary to ask the subject H questions about the attribute information and to directly acquire a response about the attribute information from the subject. In this way, the blood pressure measurement device 1 makes it possible to measure P more easily and accurately than ever before.

[0138] (Supplementary Note 1) As shown in Patent Document 3, it is known that the apparent age of subject H (hereinafter referred to as apparent age) has a strong correlation with the vascular age. For example, as the blood vessels age, the enlargement of age spots, the increase in the number of pores, and the sagging of the cheeks become more noticeable on the face. Therefore, as the vascular age increases, the apparent age also tends to increase.

[0139] Therefore, in the first embodiment, attribute classification may be performed based on the apparent age instead of the vascular age. In this case, the attribute information acquisition unit (e.g., the vascular age calculation unit 21) may calculate the apparent age instead of the vascular age. A known method (e.g., see Patent Document 4) may be used to calculate the apparent age. For example, the attribute information acquisition unit calculates the apparent age by analyzing IMG. In this way, information indicating the apparent age (apparent age information) may be used as the attribute information.

[0140] The vascular age and the apparent age are collectively referred to as the vascular-related age. Information indicating the vascular-related age is also referred to as the vascular-related age information. Both the vascular age information and the apparent age information are examples of the vascular-related age information.

[0141] By using blood vessel-related age information as attribute information, the actual blood vessel condition of the subject H can be taken into consideration. Similarly, when waveform feature amount information is used as attribute information, the actual blood vessel condition of the subject H can also be taken into consideration. In addition, it is preferable that the attribute information includes gender information. In this case, the gender difference of the subject H can be further taken into consideration.

[0142] (Supplementary Note 2) As described above, the essential concept of the blood pressure measurement device according to one embodiment of the present disclosure is to "classify the subject H based on attribute information, and use an estimation model suitable for the subject H according to the classification result." For this reason, it should be noted that in the blood pressure measurement device, "selection of a measurement model based on the evaluation result of each estimation model" is not an essential process. In other words, it is also possible to remove the selection unit from the blood pressure measurement device according to one embodiment of the present disclosure.

[0143] For the above reasons, the evaluation index sets and the measurement models do not necessarily need to be stored in advance in the model storage unit. Therefore, it is possible to remove the model evaluation unit from the model setting device according to one aspect of the present disclosure.

[0144] [Modifications] (1) However, some drugs have a significant effect on the blood vessel condition of the subject H. For this reason, when the subject H takes the drug, swelling may occur in the face of the subject H due to the change in the blood vessel condition. Therefore, the attribute information acquisition unit may determine whether swelling has occurred in the face of the subject H by analyzing the IMG.

[0145] That is, the attribute information acquisition unit may analyze the IMG to acquire information (swelling information) indicating whether or not the subject H has facial swelling as attribute information. In this case, attribute classification can be performed based on the presence or absence of facial swelling of the subject H. In other words, attribute classification can be performed taking into consideration whether or not the subject H has taken the above-mentioned medicine. This type of attribute classification also achieves the same effects as the first embodiment.

[0146] (2) Of course, the above-mentioned various individual attribute information can be combined to perform attribute classification. As an example, the attribute information according to one embodiment of the present disclosure may include at least one of vascular age information, apparent age information, gender information, waveform feature amount information, and swelling information.

[0147] [Embodiment 2] In the first embodiment, the attributes of the subjects were classified using vascular age information calculated based on the pulse wave signal calculated by the pulse wave calculation unit 17 and gender information detected based on the face image acquired by the face image acquisition unit 14.

[0148] In embodiment 2, as an alternative or modification to embodiment 1, instead of vascular age information calculated based on a pulse wave signal, blood pressure is measured using a cuff sphygmomanometer or the like, and subjects are classified based on the average blood pressure (calculated as diastolic blood pressure + systolic blood pressure × 1 / 3) that can be calculated based on the result. A model is set for each classification, and blood pressure is measured.

[0149] 10 is a functional block diagram showing the configuration of the main parts of the blood pressure measurement device 1 of embodiment 2. The blood pressure acquisition unit 2, the vascular age calculation unit 21, and the attribute classification unit 23, which have functions different from those of embodiment 1, will be described. The blood pressure acquisition unit 2 outputs BPm to the model creation unit 30, the model evaluation unit 40, and also to the vascular age calculation unit .

[0150] In order to select a blood pressure estimation model optimal for the subject H, the vascular age calculation unit 21 calculates vascular age information of the subject H based on the blood pressure of the subject H acquired by the blood pressure acquisition unit 2 as attribute information of the subject H. The vascular age calculation unit 21 calculates the mean blood pressure from the blood pressure values ​​(systolic blood pressure (SBP) and diastolic blood pressure (DBP)) acquired by the blood pressure acquisition unit 2. The mean blood pressure can be calculated, for example, as DBP+1 / 3*(SBP-DBP). The mean blood pressure calculated by the vascular age calculation unit 21 is output to the attribute classification unit 23.

[0151] The attribute classification unit 23 classifies the subject H based on the mean blood pressure, which reflects vascular age information calculated from the cuff blood pressure value, in addition to the gender information detected by the gender detection unit 22. When the attributes are classified according to the mean blood pressure value (classification index), for example, as shown in FIG.

[0152] Incidentally, attribute classification of subject H does not need to be performed each time blood pressure is measured; it only needs to be performed at least once before measurement. However, by performing this classification periodically (e.g., once every few months to years), the model can be updated to reflect the latest vascular condition of subject H.

[0153] In addition, the blood pressure acquiring unit 2, the vascular age calculating unit 21, and the attribute classifying unit 23 do not necessarily have to be provided inside the blood pressure measuring device 1. For example, when blood pressure measurement is performed using a camera built into a smartphone as the blood pressure measuring device 1, the blood pressure value measured by a contact type blood pressure monitor (e.g., a cuff type blood pressure monitor) separate from the blood pressure measuring device main body may be manually input as shown in FIG.

[0154] The blood pressure measurement unit 160 associates information capable of identifying the subject H, such as the name, ID, and face image (face recognition), with the attributes of the subject H and at least one estimation model corresponding to the attributes, and stores these in the blood pressure measurement unit 160 (FIGS. 13(a) and 13(b)), thereby eliminating the need to select a model every time blood pressure is measured, and measurement can be performed even in a state where the model selection unit 60 is not communicatively connected. Moreover, the model selection unit 60 does not need to be inside the blood pressure measurement device 1, and the blood pressure measurement device 1 and the model selection unit 60 outside the blood pressure measurement device 1 may be communicatively connected.

[0155] Fig. 14 is a flowchart showing an example of the flow of processing in the blood pressure measurement device 1. In Fig. 14, an example of a method for measuring blood pressure by the blood pressure measurement device 1 is shown.

[0156] 14, in the blood pressure measurement method by the blood pressure measurement device 1, the blood pressure measurement device 1 determines whether an estimated model of the blood pressure measurement method and the person being measured has been selected (S20A), and if an estimated model of the blood pressure measurement method and the person being measured has not been selected (S20A: No), performs the following S21 to S23. If an estimated model of the blood pressure measurement method and the person being measured has been selected (S20A: Yes), the blood pressure measurement device 1 determines whether to update the selected model (S20B), and if the selected model is to be updated even if a model has already been selected (S20B: Yes), performs the following S21 to S23. If the selected model is not to be updated (S20B: No), the blood pressure measurement device 1 does not perform the following S21 to S23, and performs the processing from S24.

[0157] First, the blood pressure acquisition unit 2 acquires the blood pressure of the subject (S21). Next, the vascular age calculation unit 21 calculates a classification index (mean blood pressure) based on the blood pressure acquired by the blood pressure acquisition unit (S22). Next, the attribute classification unit determines which attribute the subject is classified into based on the classification index calculated by the vascular age calculation unit 21, and the model selection unit 60 selects an estimated model of the corresponding attribute as the optimal model for the subject (S23).

[0158] The following steps S24 to S30 are performed every time a predicted blood pressure is calculated. First, the imaging section 11 captures an image of the person to be measured (S24). Next, the face image acquisition section 14 acquires a face image of the person to be measured from the image of the person to be measured captured by the imaging section 11 (S25).

[0159] Next, facial image division unit 15 sets (divides) areas for calculating a pulse wave from the facial image extracted by facial image acquisition unit 14 (S26). Next, skin area extraction unit 16 extracts areas where the skin is not hidden from among the areas set by facial image division unit 15 as skin areas, and pulse wave calculation unit 17 calculates a pulse wave for each of the skin areas extracted by skin area extraction unit 16 (S27).

[0160] Next, pulse wave parameter calculation section 20 calculates pulse wave waveform feature amounts as pulse wave parameters from the pulse waves of each skin region calculated by pulse wave acquisition section 10 (S28). Note that in S28, pulse wave parameter calculation section 20 may calculate the pulse wave transit time as the pulse wave parameter if there are multiple regions for which pulse waves have been calculated.

[0161] Next, blood pressure measurement unit 160 calculates a predicted blood pressure from the pulse wave parameters calculated by pulse wave parameter calculation unit 20 and the blood pressure estimation model selected in S21 to S23 that matches the attributes of the person being measured (S29). Next, blood pressure measurement result output unit 170 outputs the result of the predicted blood pressure calculated by blood pressure measurement unit 160 (S30).

[0162] (effect) In the second embodiment, for example, the attributes of subject H are classified based on the mean blood pressure, which reflects information regarding the degree of arteriosclerosis in small peripheral blood vessels (reference: http: / / medica.sanyonews.jp / article / 4523), and blood pressure is measured using a model created for each attribute classification.

[0163] Therefore, it is possible to improve the degree of conformity of each subject to the estimation model, and to predict blood pressure values ​​with high accuracy. In this embodiment, the model selection unit 60 selects the blood pressure estimation model most suitable for the subject according to the mean blood pressure value calculated from the blood pressure value acquired by the cuff sphygmomanometer and the gender, but this is not limiting, and the subject may be further classified according to the actual age, weight, etc.

[0164] [Modifications] In embodiment 2, a mean blood pressure reflecting information regarding arteriosclerosis of small peripheral blood vessels is calculated for each subject H from the systolic and diastolic blood pressures previously obtained using a cuff sphygmomanometer, and the attributes of subject H are classified based on the mean blood pressure, and an appropriate model is selected.

[0165] Here, a case will be described in which, in addition to the mean blood pressure, the pulse pressure, which can be calculated from the difference between the systolic blood pressure and the diastolic blood pressure, is used as a classification index for determining the attributes of a subject.

[0166] The overall configuration is the same as in the second embodiment, but a function is added in the vascular age calculation unit 21. The vascular age calculation unit 21 calculates the mean blood pressure and pulse pressure from the blood pressure values ​​(SBP and DBP) acquired by the blood pressure acquisition unit 2. The pulse pressure can be calculated by SBP-DBP. The vascular age calculation unit 21 outputs the calculated mean blood pressure and pulse pressure to the attribute classification unit 23.

[0167] Mean blood pressure reflects information about arteriosclerosis in small peripheral blood vessels, while pulse pressure reflects information about arteriosclerosis in larger blood vessels closer to the heart (reference: http: / / medica.sanyonews.jp / article / 4523). It is also known that arteriosclerosis occurs first in small peripheral blood vessels with age, and then progresses to larger blood vessels. Therefore, mean blood pressure rises first (arteriosclerosis in small peripheral blood vessels), and then pulse pressure gradually increases from around age 50 (arteriosclerosis in larger blood vessels closer to the heart).

[0168] Therefore, as shown in Figure 15, when attributes are classified into four categories according to the magnitude of mean blood pressure and pulse pressure, it can be seen that the subject data for (4) (high mean blood pressure and large pulse pressure) is one with more advanced arteriosclerosis than (3) (high mean blood pressure and small pulse pressure).

[0169] Classifying each subject based on these mean blood pressure and pulse pressure values ​​enables blood pressure estimation that takes into account individual differences in vascular conditions such as the progression of arteriosclerosis and the location of hardening. As people age, arteriosclerosis progresses (hardening of small blood vessels → hardening of large blood vessels), and even for people of the same age, there are individual differences in the progression of arteriosclerosis depending on lifestyle habits such as past exercise and dietary habits. However, classification based on information about the actual vascular condition makes it possible to estimate blood pressure using a model that is more suited to the subject than classification by age.

[0170] In the second embodiment, the mean blood pressure and pulse pressure are calculated based on the SBP and DBP obtained by a cuff sphygmomanometer and used as indices for attribute classification, but the obtained SBP and DBP themselves may also be used as classification indices.

[0171] [Modifications] In the second embodiment, the blood pressure values ​​for determining the attribute information of each subject are manually input into the smartphone by the user.

[0172] Here, when blood pressure is measured in a hospital, pharmacy, etc., or during a health checkup at a school or workplace, etc., the obtained blood pressure value is automatically sent to the cloud (more specifically, for example, a server on the cloud), the indices required to determine the subject's attributes are calculated on the cloud (more specifically, for example, a server on the cloud), and the subject's attributes and the corresponding estimation model are downloaded to each subject's hardware (smartphone, PC, etc.), and model selection and updating are performed (Figure 16).

[0173] If the blood pressure acquisition unit 2 is hardware capable of communicating with the cloud, the blood pressure value acquired by the blood pressure acquisition unit 2 is linked to information that can identify the subject, such as the subject's name, ID, and facial image (facial recognition), and sent to the cloud.

[0174] When the measurement results from health checkups at schools, companies, etc. are used to select or update a model for a subject, the system is linked to a database that compiles the measurements from the health checkups, and the model that is most suitable for the subject is selected based on the obtained measurements.

[0175] When the model selection unit 60 inside the blood pressure measurement device 1 is communicatively connected to a cloud (more specifically, for example, a server on the cloud), it may automatically update the model corresponding to the subject's attributes on a regular basis, or may download a model corresponding to the subject's attributes that are optimal for the subject from the cloud (more specifically, for example, a server on the cloud) when the subject wishes to update the model.

[0176] Many companies and schools require employees to undergo health checkups at least once a year, and by using the results of these checkups, it is possible to update the estimation model that is optimal for the individual at least once a year. This allows blood pressure to be predicted using an estimation model that periodically reflects the vascular condition of the individual, and therefore the accuracy can be maintained at a certain level without deterioration.

[0177] Even if you do not have a cuff blood pressure monitor at home, you can accurately measure your blood pressure on a smartphone, PC, etc. by using the results of measurements taken at a health checkup or hospital, etc. If a blood test is performed during a health checkup, an index such as blood viscosity may be used as an attribute classification index.

[0178] If, for example, a certain period of time has passed since the subject's latest classification index (mean blood pressure and / or pulse pressure) was calculated, the blood pressure measurement device 1 may notify the subject and prompt the subject to update the model. This allows a model to be selected based on the subject's current vascular condition and blood pressure to be measured using the selected model, ensuring a certain level of accuracy.

[0179] [Embodiment 3] (1) The subject H is not limited to a human being. The subject H may be any object to which the blood pressure measurement method according to one embodiment of the present disclosure can be applied. For example, the subject H may be an animal such as a dog or a cat.

[0180] (2) The ROI is not limited to the face. The ROI may be any body surface of a living body from which a pulse wave can be acquired. Other examples of the ROI include the neck, chest, and palms. However, it is preferable that the ROI is the face. When an IMG (face image) is used, the burden on the subject H during blood pressure measurement can be reduced. In other words, it becomes easier to measure the blood pressure of the subject H in a natural state (relaxed state).

[0181] (3) The blood pressure measurement unit 160 does not necessarily have to use only one blood pressure measurement model to calculate P. In other words, the model selection unit 60 may select multiple blood pressure measurement models. As an example, the blood pressure measurement unit 160 calculates multiple blood pressures using each of the multiple blood pressure measurement models. In this case, the blood pressure measurement unit 160 may calculate a representative value (e.g., average or median) of the multiple blood pressures and output the representative value as P.

[0182] (4) The attribute information does not necessarily have to be acquired based on image analysis. For example, a contact-type sensor can be used as the attribute information acquisition unit. The sensor may be configured so as to minimize the sense of restraint on the subject H. In this manner, the blood pressure measurement device 1 may be a contact-type blood pressure measurement device.

[0183] (5) The estimation model is not limited to a linear model. The model creation unit 30 may create a nonlinear model (a computational model expressed by a nonlinear function) by performing regression analysis. The model creation unit 30 may also create an estimation model using a method other than the least squares method. For example, the model creation unit 30 may create a linear model by Lasso regression with L1 regularization. By using Lasso regression, a linear model that takes into account the suppression of overlearning can be created.

[0184] (6) As mentioned above, the evaluation index (PI) of the estimation model may be a parameter calculated based on BPe and BPm. Therefore, the PI is not necessarily limited to a parameter related to error. For example, (i) the degree of freedom adjusted decision index and (ii) AIC (Akaike's Information Criteria) can be used as the PI.

[0185] (7) The evaluation index for the signal quality of the pulse wave is not limited to the SNR. For example, the pixel value of each skin region can be used as the evaluation index.

[0186] (8) The model storage unit 55 may be connected to the blood pressure measurement device 1 so as to be able to communicate with the blood pressure measurement device 1. For example, the model storage unit 55 may be a server device provided outside the blood pressure measurement device 1. In this way, the model storage unit 55 does not necessarily have to be provided inside the blood pressure measurement device 1. Similarly, the model storage unit 55 does not necessarily have to be provided inside the model setting device 100.

[0187] Also, the model storage unit 55 can be omitted. In this case, the model selection unit 60 may directly acquire each estimation model from the model creation unit 30. Also, the model selection unit 60 may directly acquire each PI from the model evaluation unit 40. However, in order to speed up the measurement of P by the blood pressure measurement device 1, it is preferable to provide the model storage unit 55.

[0188] (9) The model setting device 100 may be connected to the blood pressure measurement device 1 so as to be able to communicate with the blood pressure measurement device 1. For example, the model setting device 100 may be a server device provided outside the blood pressure measurement device 1. In this way, the model setting device 100 does not necessarily have to be provided inside the blood pressure measurement device 1. Therefore, the blood pressure measurement device 1 may be realized by a known information processing device (e.g., a smartphone, a tablet, or a personal computer).

[0189] [Embodiment 4] The control blocks of the blood pressure measurement device 1 (in particular, the model setting device 100, the model selection unit 60, the pulse wave signal quality evaluation unit 150, and the blood pressure measurement unit 160) may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software.

[0190] In the latter case, the blood pressure measurement device 1 includes a computer that executes instructions of a program, which is software that realizes each function. The computer includes, for example, at least one processor (control device) and at least one computer-readable recording medium that stores the program. The processor in the computer reads the program from the recording medium and executes it, thereby achieving the object of one embodiment of the present disclosure. The processor may be, for example, a CPU (Central Processing Unit). The recording medium may be a "non-transient tangible medium", such as a ROM (Read Only Memory) or the like, as well as a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like. The device may further include a RAM (Random Access Memory) in which the program is expanded. The program may be supplied to the computer via any transmission medium (such as a communication network or a broadcast wave) that can transmit the program. Note that one embodiment of the present disclosure may also be realized in the form of a data signal embedded in a carrier wave in which the program is embodied by electronic transmission.

[0191] [Additional Notes] One aspect of the present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Also, embodiments obtained by appropriately combining the technical means disclosed in different embodiments are included in the technical scope of one aspect of the present disclosure. Furthermore, new technical features can be formed by combining the technical means disclosed in each embodiment.

Claims

1. A blood pressure measurement device that measures a first blood pressure of a living body based on a pulse wave of the living body, a pulse wave acquiring unit for acquiring at least one pulse wave in a predetermined region on a body surface of the living body; a pulse wave parameter calculation unit that calculates at least one pulse wave parameter based on only the at least one pulse wave; a blood pressure acquisition unit for acquiring a second blood pressure of the living body; an attribute information acquiring unit that acquires attribute information related to a blood vessel state of the living body based on the second blood pressure and an image of the living body; an attribute classification unit that classifies attributes of the living body based on information indicating a vascular age of the living body and information indicating a sex of the living body, which are included in the attribute information; the blood pressure measurement device is communicably connected to a model storage unit in which at least one blood pressure estimation model for estimating the first blood pressure corresponding to the classification result of the attribute is stored in advance; The blood pressure measurement device further includes a first blood pressure measurement unit that calculates the first blood pressure based only on the at least one pulse wave parameter and a constant using the at least one blood pressure estimation model corresponding to the classification result of the attribute.

2. The model storage unit further stores in advance an evaluation result of each of the at least one blood pressure estimation model, 2. The blood pressure measurement device according to claim 1, wherein the first blood pressure measurement unit calculates the first blood pressure based on the at least one pulse wave parameter using the measurement model corresponding to a classification result of the attribute selected by a model selection unit that selects a measurement model for calculating the first blood pressure from the at least one blood pressure estimation model based on the evaluation result of each of the at least one blood pressure estimation model.

3. N is an integer equal to or greater than 2, the attribute classification unit classifies the attributes into N different patterns from a first attribute to an Nth attribute; k is an integer greater than or equal to 1 and less than or equal to N, The blood pressure measurement device according to claim 1 , wherein the model storage unit stores in advance at least one k-th model, which is the at least one blood pressure estimation model according to a k-th attribute.

4. 4. The blood pressure measurement device according to claim 1, wherein the living body is notified of the classification index included in the attribute information if a certain period of time has passed since the latest calculation of the classification index.

5. The blood pressure measurement device according to claim 1 , wherein the attribute information acquisition unit acquires the attribute information based on the value of the second blood pressure at least once before measuring the first blood pressure.

6. The blood pressure measurement device according to claim 1 , wherein the attribute information includes information indicating a mean blood pressure of the living body.

7. The blood pressure measurement device according to claim 1 , wherein the attribute information includes information indicating a pulse pressure of the living body.

8. The blood pressure measurement device according to claim 1 , wherein the attribute information acquisition unit acquires the attribute information via a cloud.

9. The blood pressure measurement device according to any one of claims 1 to 8, wherein the attribute information is calculated on a cloud.

10. The blood pressure measurement device according to claim 1 , wherein the attribute information acquisition unit acquires the attribute information by manually inputting the attribute information.

11. The blood pressure measurement device according to claim 1 , wherein the attribute information includes information regarding a degree of arteriosclerosis of blood vessels of the living body.

12. The blood pressure measurement device according to claim 1 , wherein the attribute information includes information indicating an apparent age of the living body.

13. The blood pressure measurement device according to claim 1 , wherein the attribute information includes information indicating a waveform feature value of the at least one pulse wave.

14. The blood pressure measurement device according to claim 1 , wherein the attribute information includes information indicating whether or not swelling is present in the predetermined area.

15. The blood pressure measurement device according to claim 1 , wherein the predetermined area is a face of the living body.

16. 1. A blood pressure measurement method using a blood pressure measurement device that measures a first blood pressure of a living body based on a pulse wave of the living body, comprising: a pulse wave acquiring step of acquiring at least one pulse wave in a predetermined region on a body surface of the living body; a pulse wave parameter calculation step of calculating at least one pulse wave parameter based on only the at least one pulse wave; a blood pressure acquiring step of acquiring a second blood pressure of the living body; an attribute information acquiring step of acquiring attribute information related to a blood vessel condition of the living body based on the second blood pressure and an image of the living body; and an attribute classification step of classifying attributes of the living body based on information indicating a vascular age of the living body and information indicating a sex of the living body, which are included in the attribute information, the blood pressure measurement device is communicably connected to a model storage unit in which at least one blood pressure estimation model for estimating the first blood pressure corresponding to the classification result of the attribute is stored in advance; The blood pressure measurement method includes a first blood pressure measurement step of calculating the first blood pressure based only on the at least one pulse wave parameter and a constant using the at least one blood pressure estimation model corresponding to a classification result of the attribute.

Citation Information

Patent Citations

  • Portable blood pressure gauge

    JP2000225097A

  • Noninvasive continuous blood pressure estimating device and noninvasive continuous blood pressure predicting device

    JP2000237153A

  • Method of evaluating blood vessel aging

    JP2002238867A

  • Age estimation system and age estimation method

    JP2009086901A

  • Biological state estimating device, program, and recording medium

    JP2009089829A