Apparatus for measuring index of endothelial function of arterial vessel

Abstract

To provide an endothelial function index value measuring device of an arterial blood vessel capable of quickly acquiring an endothelial function index value capable of evaluating an endothelial function.SOLUTION: A cross-sectional image is generated from reflected ultrasonic signals obtained by the ultrasonic probe 24 on the skin 18, the biceps muscle cross-sectional area MCSA is calculated from the muscle image area GKN of the biceps muscle in the cross-sectional image, the biceps muscle mass MCSAI (= MCSA / BSA) is calculated from the biceps muscle cross-sectional area MCSA and the body surface area BSA, and the endothelial mass index MHI (= MCSAI / MB) is calculated from the biceps muscle mass MCSAI and the muscle brightness MB of the muscle image area GKN.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an endothelial function index measurement device for arterial blood vessels that can quickly obtain a new endothelial function index value for evaluating the endothelial function of a living body's arterial blood vessels from the living body's muscle mass and quality. [Background technology]

[0002] There is known an endothelial function testing device for arterial blood vessels that contacts an ultrasonic probe with the epidermis of a living body, generates a cross-sectional image showing a cross section of an arterial blood vessel located under the epidermis from an ultrasonic reflection signal from under the epidermis, measures a change in the diameter of the arterial blood vessel from the cross-sectional image, and tests the endothelial function of the arterial blood vessel based on the amount of change in the diameter. For example, the endothelial function testing devices for arterial blood vessels are described in Patent Documents 1 and 2.

[0003] The arterial endothelial function testing devices described in Patent Documents 1 and 2 synthesize ultrasound cross-sectional images of the brachial artery based on reflected waves detected by an ultrasound probe placed on the skin of the upper arm. These images are used to generate transverse (short-axis) or longitudinal (long-axis) cross-sectional images of the brachial artery, which are ultrasound cross-sectional images of the brachial artery under the skin. From these images, the diameter of the arterial blood vessel is calculated. To evaluate endothelial function, the percentage change in blood vessel diameter (%) FMD [= 100 × (dmax - d) / d] (where d is the resting vessel diameter and dmax is the maximum vessel diameter temporarily increased after ischemia-induced hyperemia) is calculated, representing flow-mediated vasodilation (FMD) after ischemia-reactive hyperemia. Flow-mediated vasodilation is believed to be due to nitric oxide (NO), which is generated in response to shear stress exerted on the arterial endothelium by blood flow. The FMD value, which reflects the degree of flow-mediated vasodilation, is therefore used as an endothelial function index to evaluate arterial endothelial function. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-124063 [Patent Document 2] Japanese Patent Application Laid-Open No. 2017-209454 Summary of the Invention [Problem to be solved by the invention]

[0005] In the endothelial function testing device described above, in order to obtain sufficient blood flow to apply shear stress to the endothelium of the arterial blood vessels when avascularization is released, it is necessary to hold the avascularization period for about 5 minutes, prior to the release of avascularization, to put the downstream side of the avascularized area of ​​the artery into an ischemic state.

[0006] Thus, in order to obtain the FMD value, taking into account the approximately 5-minute period for avascularization and the subsequent measurement time for the endothelial diameter of the arterial blood vessel after avascularization and release, a measurement time of at least approximately 7 minutes is required, which is inconvenient in that it places a heavy burden on the subject due to compression of the arm and having to maintain a resting state during that time.

[0007] The present invention has been made against the background of the above circumstances, and its purpose is to provide an arterial endothelial function index value measuring device that can quickly obtain an endothelial function index value that evaluates the endothelial function of the arterial blood vessels of a living body.

[0008] The inventors of the present invention have conducted various studies based on the above circumstances, and have found that the cross-sectional area of ​​the biceps brachii muscle (MCSA) near the brachial artery, which is the area of ​​the biceps brachii muscle between the epidermis and the brachial artery, can be calculated by the ultrasound short-axis image obtained by the endothelial function test device. 2 ) to the body surface area BSA (m 2 ) to obtain the normalized biceps muscle mass MCSAI (mm 2 / m 2After obtaining the muscle brightness ratio (MCSAI) of the biceps brachii muscle image, the researchers found that the biceps brachii muscle brightness ratio (MCSAI / MB) obtained by dividing the biceps brachii muscle mass (MCSAI) by the muscle brightness (MB) (e.g., 256-level grayscale) of the biceps brachii muscle image, which indicates the quality (fat mass) of the biceps brachii muscle, correlates significantly with FMD. Based on this finding, the researchers found that the endothelial function index (MHI) showing a significant correlation with FMD can be rapidly obtained by calculating the biceps brachii muscle cross-sectional area (MCSA) adjacent to the brachial artery displayed on an ultrasound short-axis image, the body surface area (BSA) calculated from the previously measured height and weight of the subject, and the muscle brightness (MB) of the biceps brachii muscle image adjacent to the brachial artery displayed on an ultrasound short-axis image. The researchers also found that the endothelial function index (MHI) can be used to rapidly evaluate the endothelial function of the subject's arterial blood vessels. The present invention is based on this finding. The endothelial function index (MHI) can be called a muscle health index. [Means for solving the problem]

[0009] That is, the gist of the first invention is an apparatus for measuring an endothelial function index value of an arterial blood vessel, which (a) detects an ultrasonic reflection signal from an ultrasonic probe brought into contact with the epidermis of the upper arm of a living body, generates a cross-sectional image including the biceps brachii muscle and the brachial artery located under the epidermis from the ultrasonic reflection signal, and calculates an endothelial function index value (MHI) of the arterial blood vessel of the living body from the cross-sectional image, and includes: (b) a muscle image region generation unit that generates an image region of the biceps brachii muscle within the cross-sectional image; (d) a muscle cross-sectional area calculation unit that calculates the biceps brachii muscle cross-sectional area MCSA from the muscle image area; (d) a muscle mass calculation unit that calculates the biceps brachii muscle mass MCSAI by dividing the biceps brachii muscle cross-sectional area MCSA by the body surface area BSA of the living body; and (e) an endothelial function index value calculation unit that calculates the biceps brachii muscle mass brightness ratio by dividing the biceps brachii muscle mass MCSAI by the muscle brightness MB in the image area of ​​the biceps brachii muscle in the cross-sectional image, and outputs the ratio as the endothelial function index value MHI. [Effects of the Invention]

[0010] In the device for measuring endothelial function index values ​​for arterial blood vessels configured in this manner, a cross-sectional image is generated from ultrasonic reflection signals from below the epidermis obtained by contacting an ultrasound probe with the epidermis of the upper arm, a muscle image region of the biceps brachii located within the cross-sectional image is generated, the biceps brachii cross-sectional area MCSA is calculated from the muscle image region of the biceps brachii, the biceps brachii cross-sectional area MCSA is divided by the body surface area BSA to calculate a normalized biceps brachii muscle mass MCSAI, the biceps brachii muscle mass MCSAI is divided by the muscle brightness MB of the image region of the biceps brachii within the cross-sectional image to calculate a biceps brachii muscle mass brightness ratio, and this biceps brachii muscle mass brightness ratio is output as the endothelial function index value MHI. This allows for the rapid calculation of the endothelial function index MHI, which shows a significant correlation with FMD, by using the cross-sectional area MCSA of the biceps brachii muscle adjacent to the brachial artery displayed on the ultrasound short-axis image, the body surface area BSA calculated from the subject's previously measured height and weight, and the muscle brightness MB of the biceps brachii muscle image adjacent to the brachial artery displayed on the ultrasound short-axis image. This endothelial function index MHI can then be used to rapidly evaluate the endothelial function of the subject's arterial blood vessels.

[0011] The gist of the second invention is that in the first invention, the cross-sectional image is a short-axis image showing a cross section perpendicular to an artery under the skin of the living body, and the muscle image region generation unit generates the muscle image region appearing in the short-axis image from image information appearing in the actual short-axis image using a trained model that has previously trained and stored muscle image regions showing muscle tissue appearing in the short-axis image, and displays the muscle image region superimposed on the short-axis image. This clearly generates a muscle image region that is likely to be unclear when displayed in a short-axis image, allowing for accurate calculation of the endothelial function index (MHI).

[0012] The gist of the third invention is that in the second invention, the trained model is a machine learning model that has repeatedly performed supervised learning using training data in which a predetermined short-axis image is used as input data and the area of ​​muscle tissue identified in the short-axis image by a skilled operator is used as output data. This allows the area of ​​muscle tissue to be clearly identified in the short-axis image, thereby enabling the endothelial function index (MHI) to be accurately calculated.

[0013] The gist of a fourth invention is that, in the third invention, the machine learning model is a multi-layer neural network that has undergone deep learning, in which weighting coefficients between nodes belonging to adjacent layers are sequentially updated when the input data is input to nodes in an input layer and the output data indicating the muscle tissue area identified by an experienced operator in the short-axis image is repeatedly taught to nodes in an output layer. This allows the muscle tissue area to be more clearly identified in the short-axis image, thereby allowing the endothelial function index (MHI) to be accurately calculated.

[0014] The gist of the fifth invention is that, in the fourth invention, in the multilayer neural network, brightness information for each pixel constituting the predetermined short-axis image is input as the input data to an input layer, and output data indicating pixels representing the muscle tissue area is output from an output layer, thereby more clearly identifying the muscle tissue area in the short-axis image and accurately calculating the endothelial function index (MHI).

[0015] The gist of the sixth invention is that, in the fifth invention, image recognition is performed on the output data output from the output layer of the multilayer neural network for each pixel constituting the short-axis image, the texture of the region to which the pixel belongs is characterized from information on each pixel and its adjacent pixels, and a meaningful region division, whether it belongs to the muscle tissue region or other tissue, is estimated for each pixel based on the texture, thereby reducing the number of nodes constituting the output layer of the multilayer neural network and making the configuration of the multilayer neural network more compact.

[0016] The gist of a seventh aspect of the present invention is that in the first aspect of the present invention, the muscle cross-sectional area calculation unit calculates the biceps brachii cross-sectional area (MCSA) of the muscle image region generated by the muscle image region generation unit from the total number of pixels in the muscle image region, thereby accurately and quantitatively obtaining the area of ​​the muscle image region.

[0017] The gist of an eighth aspect of the present invention is the first aspect of the present invention, further comprising a muscle brightness calculation unit that calculates the average value of the brightness of each pixel in the muscle image region generated by the muscle image region generation unit as the muscle brightness MB of the image region of the biceps brachii in the cross-sectional image, thereby accurately and quantitatively obtaining the brightness of the muscle image region as a parameter indicating the proportion of fatty tissue and connective tissue in the muscle.

[0018] The gist of a ninth aspect of the present invention is that the first aspect of the present invention further includes a body surface area calculation unit that calculates the body surface area BSA of the living body based on the height H and weight W of the living body using a pre-stored Dupois equation (1). This makes it possible to easily obtain the body surface area BSA of the living body. BSA=H 0.725 ×W 0.425 ×0.007184 (1) [Brief explanation of the drawings]

[0019] [Figure 1] 1A and 1B are diagrams for explaining the configuration and function of an apparatus for measuring an endothelial function index value of an arterial blood vessel according to one embodiment of the present invention, and a perspective view showing an ultrasound probe unit. [Figure 2] 2 is a diagram illustrating XYZ orthogonal coordinate axes used in the present embodiment for positioning an ultrasound probe in the endothelial function index measuring device of FIG. 1. FIG. [Figure 3] 2 is an enlarged view schematically showing the layer membrane structure of a blood vessel that is the measurement target of the endothelial function index measuring device of FIG. 1. FIG. [Figure 4] 1. FIG. 4 is a diagram illustrating an example of an ultrasound image of an arterial blood vessel displayed on an image display device during the positioning operation of an ultrasound probe by the endothelial function index measuring device of FIG. [Figure 5] 10 is a photograph showing a muscle image region superimposed on a short axis image. [Figure 6] 2 is a functional block diagram illustrating the configuration of a muscle image region generating unit provided in the endothelial function index measuring device of FIG. 1. FIG. [Figure 7]2 is a flowchart illustrating a main part of the control operation of the electronic control device of FIG. 1, showing an endothelial function index measurement control routine. [Figure 8] This is a two-dimensional coordinate graph consisting of a horizontal axis X representing the endothelial function index MHI and a vertical axis Y representing (%) FMD, on which 155 data points showing the ratio R of biceps brachii muscle mass MCSAI to muscle brightness MB (endothelial function index MHI=MCSAI / MB) and (%) FMD for 155 people are plotted. DETAILED DESCRIPTION OF THE INVENTION

[0020] In the present invention, the endothelial function index measuring device preferably measures the endothelial function index of the brachial artery, which is an artery located under the epidermis of the upper arm of a living body, but is also similarly applicable to detecting the vascular endothelial function of other arteries that can be measured from the epidermal surface, such as the forearm and phalangeal arteries of a living body.

[0021] The ultrasonic probe provided in the endothelial function index measuring device in the present invention is preferably an H-shaped probe having, on a single plane, two parallel rows of first and second short-axis ultrasonic array probes and a long-axis ultrasonic array probe connecting the first and second short-axis ultrasonic array probes at their longitudinal centers. However, the present invention can also be similarly applied and effective to endothelial function testing devices equipped with other probes, such as those having a single short-axis ultrasonic array probe or those having two parallel rows of first and second short-axis ultrasonic array probes without a long-axis ultrasonic array probe.

[0022] In the present invention, the trained model is preferably a machine learning model based on a machine learning algorithm. The machine learning algorithm may be a nearest neighbor algorithm, a naive Bayes algorithm, a support vector machine, or the like. Furthermore, deep learning is used, which uses a neural network to generate features and connection weighting coefficients for learning. Furthermore, the training data is training data, and is composed of pairs of input data and output data.

[0023] Preferred embodiments of the present invention will be described in detail below with reference to the drawings. Note that in the following embodiments, the drawings are appropriately simplified or modified, and the dimensional ratios and shapes of the various parts are not necessarily drawn accurately. [Example]

[0024] FIG. 1 is a perspective view illustrating the overall configuration of an endothelial function index measuring device 22 that uses an ultrasound probe unit 12 supported by a sensor support 10 to non-invasively examine the endothelial function of blood vessels (arterial blood vessels) 20 located from above the skin 18 (strictly speaking, the epidermis) to below the skin 18 in the upper arm 16 of a living body 14.

[0025] The ultrasonic probe unit 12 functions as an ultrasonic sensor for measuring biological information related to the blood vessel 20, i.e., the endothelial function index MHI of the blood vessel 20, and is equipped with an H-shaped ultrasonic probe 24 having a pair of parallel first and second short-axis ultrasonic array probes 24a and 24b, and a long-axis ultrasonic array probe 24c connecting the longitudinal center portions of the first and second short-axis ultrasonic array probes 24a and 24b, on a single plane, i.e., a flat probe surface 25. The ultrasonic probe unit 12 also includes a multi-axis drive mechanism (positioning mechanism) 26 that positions the ultrasonic probe 24 in the X, Y, and Z directions and positions the rotation angles around the X and Z axes, and a drive device 27 including an XYZ motor that drives the multi-axis drive mechanism 26 in the X, Y, and Z directions, respectively. The first short-axis ultrasonic array probe 24a, the second short-axis ultrasonic array probe 24b, and the long-axis ultrasonic array probe 24c are, for example, as shown in FIG. 2 described later, made of a number of ultrasonic transducers a1 to a n are linearly arranged to form a longitudinal shape.

[0026] Fig. 2 is a diagram illustrating the XYZ orthogonal coordinate axes used in this embodiment for positioning the ultrasonic probe 24. With respect to the XYZ orthogonal coordinate axes shown in Fig. 2, the ultrasonic probe 24 is translated in the X-axis direction by, for example, the multi-axis drive mechanism 26. It is also rotated around the X-axis and the Z-axis.

[0027] FIG. 3 is an enlarged view showing a schematic diagram of the wall structure of a blood vessel 20 in the upper arm 16. The blood vessel 20 has a three-layer structure consisting of a tunica intima (endothelium) L1, a tunica media L2, and a tunica adventitia L3. Since ultrasound waves are generally reflected at areas with different acoustic impedances, an image of the blood vessel 20 taken using ultrasound shows the interface between the blood in the vessel lumen and the tunica intima L1 and the interface between the tunica media L2 and the tunica adventitia L3 in white, and the vessel lumen in black. Furthermore, between the blood vessel 20 and the skin 18, the muscle tissue KN of the biceps brachii is displayed as a black and white patchwork.

[0028] 1, the endothelial function index measuring device 22 includes an electronic control device 28 configured from a so-called microcomputer, an image display device 30, an ultrasound drive control circuit 32, and a three-axis drive motor control circuit 34. In measuring a blood vessel condition using the endothelial function index measuring device 22, when a drive signal is supplied from the ultrasound drive control circuit 32 by the electronic control device 28, beam-shaped ultrasound waves are sequentially emitted from the first short-axis ultrasonic array probe 24a, the second short-axis ultrasonic array probe 24b, and the long-axis ultrasonic array probe 24c of the ultrasound probe 24 in the ultrasound probe unit 12 by well-known beamforming drive. Then, the first short-axis ultrasonic array probe 24a, the second short-axis ultrasonic array probe 24b, and the long-axis ultrasonic array probe 24c detect reflected ultrasonic signals, and the detected reflected ultrasonic signals are processed in the electronic control device 28 to generate a first short-axis image, a second short-axis image, and a long-axis image, which are ultrasonic cross-sectional images of the blood vessels 20 under the skin 18, and these images are displayed in the first short-axis image display area G1, the second short-axis image display area G2, and the long-axis image display area G3 of the image display device 30, respectively.

[0029] In Figure 1, the electronic control device 28 functionally comprises an ultrasound drive control unit 72, a detection processing unit 74, an ultrasound image generation unit 76, a three-axis drive motor control unit 78, a muscle image area generation unit 82, a display control unit 88, and an endothelial function index value measurement unit 90 (a muscle cross-sectional area calculation unit 92, a body surface area calculation unit 93, a muscle brightness calculation unit 94, a muscle mass calculation unit 95, and an endothelial function index value calculation unit 96).

[0030] FIG. 4 shows the positional relationship between the ultrasound probe 24 positioned at a predetermined measurement position and the blood vessel 20 when an ultrasound image of the blood vessel 20 is generated in measurement of the endothelial function index (MHI) by the endothelial function index measuring device 22, and also illustrates an example of an ultrasound image of the blood vessel 20 displayed on the image display device 30 in that positional relationship. In the image display device 30, as shown in FIG. 4(a), for example, the first short-axis image display region G1, the second short-axis image display region G2, and the long-axis image display region G3 have a common vertical axis indicating a constant depth dimension preset, for example, 40 mm, from the skin 18. Note that "ImA" and "ImB" in FIG. 4(a) each indicate a cross section of the blood vessel 20. The first short-axis image display region G1 and the second short-axis image display region G2 are ultrasound images displayed with a constant horizontal dimension preset, for example, 30 mm, so that the blood vessel 20 is located at the center in the width direction.

[0031] In the endothelial function index measuring device 22, when measuring the endothelial function index MHI of the blood vessel 20, in order to obtain a first short-axis image display region G1 and a second short-axis image display region G2 including the muscle tissue KN to be measured, the ultrasound probe 24 is automatically positioned at a measurement position directly above the blood vessel 20 by driving the multi-axis drive mechanism 26 to which a drive signal is supplied from a three-axis drive motor control circuit 34 by a three-axis drive motor control unit 78 provided in the electronic control device 28. The measurement position is a position where the first short-axis ultrasonic array probe 24a and the second short-axis ultrasonic array probe 24b are perpendicular to the blood vessel 20 and the long-axis ultrasonic array probe 24c is parallel to the blood vessel 20. Explaining this with reference to FIG. 4(b), the measurement position is a position where "a=b, c=d, e=f" in FIG. 4(b). That is, the measurement position is one where the distance from the first short-axis ultrasonic array probe 24a to the center of the blood vessel 20 is equal to the distance from the second short-axis ultrasonic array probe 24b to the center of the blood vessel 20, and the image of the blood vessel 20 is positioned at the center in the width direction in both the first short-axis image display region G1 and the second short-axis image display region G2. The first short-axis image display region G1 and the second short-axis image display region G2 have the same horizontal axis dimensions (c+d, e+f) and the same vertical axis dimensions (a+g, b+h).

[0032] When measuring the state of blood vessels using the endothelial function index measuring device 22, the sensor support 10 holds the ultrasound probe unit 12 in a fixed position so as to lightly contact the blood vessel 20 located above the skin 18 of the upper arm 16 of the living body 14 and directly below the skin 18 without deforming the blood vessel 20.

[0033] As shown in FIG. 1, the sensor support 10 includes, for example, a base 36 fixed to a pedestal (not shown), a unit fixture 38 to which the ultrasonic probe unit 12 is fixed so as to be able to be pressed toward the upper arm 16 by a pressing actuator 37, a first arm 40 whose base end is connected to the base 36 via a flexible fixture, and a flexible arm 44 having a second arm 42 whose tip end is connected to the unit fixture 38 via a flexible fixture and which is connected to the tip of the first arm 40 so as to be rotatable around one axis.

[0034] The multi-axis drive mechanism 26 is configured to include an X-axis rotation (yawing) mechanism fixed to the unit fixture 38 for positioning the rotation position of the ultrasonic probe 24 about the X axis by an X-axis rotation actuator, an X-axis translation mechanism for positioning the translation position of the ultrasonic probe 24 in the X-axis direction by an X-axis rotation actuator, and a Z-axis rotation mechanism for positioning the rotation position of the ultrasonic probe 24 about the Z axis by a Z-axis actuator. With this configuration, the multi-axis drive mechanism 26 positions the ultrasonic probe 24 by a drive device 27 that operates in accordance with commands from an electronic control device 28.

[0035] The ultrasonic wave drive control circuit 32 controls the emission of ultrasonic waves from the ultrasonic probe 24 to the blood vessel 20 in accordance with a command from an ultrasonic wave drive control unit 72 provided in the electronic control device 28. For example, in the first short-axis ultrasonic array probe 24a, a number of ultrasonic transducers a1 to a n Among them, a certain number of ultrasonic transducers from the end ultrasonic transducer a1, for example, 15 a1 to a 15 By performing beamforming driving in which the ultrasonic transducers are simultaneously driven at a frequency of about 10 MHz while providing a predetermined phase difference between each other, convergent ultrasonic beams are sequentially emitted toward the blood vessel 20 in the direction of the arrangement of the ultrasonic transducers. Then, the ultrasonic transducers are shifted one by one to scan the ultrasonic beam, and the reflected waves for each emission are received and input to the electronic control device 28. The reflected wave signals input to the electronic control device 28 are detected by the detection processing unit 74 and processed by the ultrasonic image generation unit 76 as information that can be used for image synthesis, as will be described in detail below.

[0036] When measuring the endothelial function index value MHI using the endothelial function index value measuring device 22, the biceps brachii muscle mass MCSAI (=MCSA / BSA) calculated from the biceps brachii muscle cross-sectional area MCSA of the muscle image area GKN located within the first short-axis image display area G1 and the second short-axis image display area G2 based on the position of the blood vessel 20 and the body surface area BSA of the living body 14, and the muscle brightness MB of the muscle image area GKN are obtained, and the ratio R (=MCSAI / MB) of the biceps brachii muscle mass MCSAI of the muscle image area GKN to the muscle brightness MB of the muscle image area GKN is calculated, and this ratio value R is output to the image display device 30 as the endothelial function index value MHI.

[0037] The biceps brachii cross-sectional area MCSA of the muscle image region GKN is a parameter corresponding to the muscle mass of the living body 14, and the muscle brightness MB of the muscle image region GKN is a parameter corresponding to the muscle quality. The muscle brightness MB of the muscle image region GKN shown in the ultrasound cross-sectional image is used as an evaluation value indicating the muscle quality. In the ultrasound cross-sectional image, an increase in non-contractile tissues such as fat and connective tissue within the muscle appears brighter and whitish. This is because the greater the difference in acoustic impedance between the contractile tissue and non-contractile tissue within the muscle, the larger the echo.

[0038] The ultrasound image generation unit 76 generates a first short-axis image representing a cross section of the blood vessel 20 based on the ultrasound reflection signals detected by the first short-axis ultrasound array probe 24a, generates a second short-axis image representing a cross section of the blood vessel 20 based on the ultrasound reflection signals detected by the second short-axis ultrasound array probe 24b, and generates a long-axis image representing a longitudinal section of the blood vessel 20 based on the ultrasound reflection signals detected by the long-axis ultrasound array probe 24c.

[0039] The display control unit 88 displays the first short axis cross-sectional image, the second short axis cross-sectional image, and the long axis cross-sectional image generated by the ultrasound image generation unit 76 in the first short axis image display area G1, the second short axis image display area G2, and the long axis image display area G3, respectively, on the image display device 30.

[0040] Generally, short-axis cross-sectional images obtained using ultrasound tend to have a lot of speckle noise, making the muscle tissue KN unclear, as shown in Fig. 5 showing the actual first short-axis cross-sectional image or second short-axis cross-sectional image. For this reason, the endothelial function index measuring device 22 of this embodiment includes a muscle image region generating unit 82 that generates a muscle image region GKN indicating the muscle tissue KN located in the first short-axis image display region G1 and the second short-axis image display region G2 from image information of the actual first short-axis image and second short-axis image display region G2, using a learned model LM that has previously learned and stored the muscle tissue KN appearing in the short-axis image.

[0041] As shown in Fig. 6, the muscle image region generation unit 82 includes a trained model unit 84 that stores a trained model LM that has been previously trained and stored for the muscle tissue KN appearing in the first short-axis image and the second short-axis image, and a muscle image region synthesis unit 86 that uses the trained model LM to generate a muscle image region GKN representing the muscle tissue KN located in the actual first short-axis image and the second short-axis image, respectively, from image information appearing in the actual first short-axis image and the second short-axis image. The display control unit 88 displays the muscle image region GKN representing the muscle tissue KN over the first short-axis image and the second short-axis image, respectively, in the first short-axis image display region G1 and the second short-axis image display region G2 of the image display device 30, for example, as shown in Fig. 5. In the case of Fig. 5, the muscle image region GKN is displayed surrounded by a curved image LKN that indicates the outer edge of the muscle tissue KN. The muscle image region GKN may also be an image displayed with diagonal lines or color that indicates the entire region of the muscle tissue KN.

[0042] The trained model LM is a machine learning model MLM obtained by repeatedly performing supervised learning using training data in which the image information of the entire first short-axis image and / or second short-axis image displayed in the first short-axis image display area G1 and the second short-axis image display area G2 is used as input data, and the area of ​​muscle tissue KN identified by an experienced operator drawing with a mouse or pen in the short-axis image displayed on an input display device (not shown) is used as output data. The image display device 30 may be used as the input display device.

[0043] Preferably, the machine learning model MLM is a multi-layer neural network that has undergone deep learning in which weight coefficients between nodes belonging to adjacent layers are sequentially updated when input data is input to nodes in the input layer and output data indicating the area of ​​muscle tissue KN identified by an experienced operator in the short-axis image is repeatedly taught to nodes in the output layer.

[0044] In the above multilayer neural network, preferably, brightness information Pin(x,y) for each pixel P constituting a predetermined short-axis image is input as input data to the input layer, and output data Pout(x,y,t) indicating the pixel P displaying the area of ​​muscle tissue KN is output from the output layer. The t of the output data is a texture t indicating whether the pixel P indicated by the output data Pout(x,y) is within the area of ​​muscle tissue KN (if t=1, it is within the area of ​​muscle tissue KN, and if t=0, it is a feature indicating other tissue).

[0045] That is, in the output data output from the output layer of the multilayer neural network, image recognition is performed on a pixel P basis that constitutes the short-axis image, and the texture t of the region (tissue) to which the pixel P belongs is characterized from the information on each pixel P and its adjacent pixels P, and a meaningful region division, whether it belongs to muscle tissue KN or other tissue, is estimated for each pixel P(x, y) based on the texture t.

[0046] Returning to FIG. 1 , the endothelial function index measurement unit 90 includes a muscle cross-sectional area calculation unit 92, a body surface area calculation unit 93, a muscle brightness calculation unit 94, a muscle mass calculation unit 95, and an endothelial function index calculation unit 96. The muscle cross-sectional area calculation unit 92 calculates the biceps brachii muscle cross-sectional area MCSA (mm ) of the muscle image region GKN, which is synthesized by the muscle image region synthesis unit 86 of the muscle image region generation unit 82 and shows the muscle tissue KN shown in the first short-axis image and / or the second short-axis image, based on the total number of pixels P in the muscle image region GKN. 2 ) is calculated, and the biceps brachii cross-sectional area MCSA is displayed in the character display area G4 of the image display device 30.

[0047] The muscle brightness calculation unit 94 calculates the average brightness of each pixel P in the muscle image region GKN, which is synthesized by the muscle image region synthesis unit 86 of the muscle image region generation unit 82 and shows the muscle tissue KN in the first short-axis image and / or the second short-axis image, and sets this as the muscle brightness MB of the muscle image region GKN. For example, the brightness of each pixel P is expressed using an 8-bit grayscale with 256 gradations, with white being 255, and the muscle brightness calculation unit 94 calculates the muscle brightness MB, which is the average brightness of all pixels P in the muscle image region GKN. This muscle brightness MB corresponds to fat contained in the biceps and represents muscle quality.

[0048] The body surface area calculation unit 93 reads the actual height H and weight W of each living body 14 to be measured, and calculates the body surface area BSA (m ) of each living body 14 to be measured based on the actual height H and weight W from the DuPont formula (1) stored in advance. 2 ) is calculated. BSA=H 0.725 ×W 0.425 ×0.007184 (1)

[0049] The muscle mass calculation unit 95 calculates the biceps brachii cross-sectional area MCSA (mm 2 ) is calculated by the body surface area calculation unit 93. 2 ) is divided by the biceps muscle mass MCSAI (mm 2 / m 2 ) is calculated.

[0050] The endothelial function index calculation unit 96 calculates the ratio R (=MCSAI / MB) between the biceps muscle mass MCSAI calculated by the muscle mass calculation unit 95 and the muscle brightness MB of the muscle image area GKN calculated by the muscle brightness calculation unit 94, and displays the ratio R as the endothelial function index MHI (Muscle Health Index) in the character display area G4 of the image display device 30.

[0051] Fig. 7 is a flowchart illustrating the main control operations of the electronic control device 28, which are repeatedly executed each time a measurement start operation is performed. Fig. 7 shows an endothelial function index measurement routine that calculates an endothelial function index MHI, which is the ratio R (=MCSAI / MB) of the biceps brachii muscle mass MCSAI to the muscle brightness MB, based on the biceps brachii muscle mass MCSAI and the muscle brightness MB based on the muscle image area GKN, and outputs the endothelial function index MHI.

[0052] 7 (hereinafter, "step" will be omitted), it is determined whether or not a start operation, including a command operation to start measurement of the endothelial function index value MHI, an operation to start ultrasound image display control, and an operation to input the height H and weight W of the subject, has been performed, based on, for example, a touch operation of any input switch on the screen of the image display device 30. If the determination in S1 is negative, the system is put into standby mode. If the determination in S1 is positive, in S2 corresponding to the ultrasound image generating unit 76, a first short-axis image representing a cross section of the blood vessel 20 is generated based on the ultrasound reflection signals detected by the first short-axis ultrasound array probe 24a, a second short-axis image representing a cross section of the blood vessel 20 is generated based on the ultrasound reflection signals detected by the second short-axis ultrasound array probe 24b, and a long-axis image representing a longitudinal section of the blood vessel 20 is generated based on the ultrasound reflection signals detected by the long-axis ultrasound array probe 24c, and these are displayed in the first short-axis image display region G1, the second short-axis image display region G2, and the long-axis image display region G3, respectively.

[0053] Next, in S3, it is determined whether an operation to activate control for generating and displaying a muscle image region GKN has been performed, for example, based on whether the screen of either the first short-axis image display region G1 or the second short-axis image display region G2 displaying the short-axis image of the image display device 30 has been touched. If the determination in S3 is negative, S1 and subsequent steps are repeated. If the determination is positive, in S4 corresponding to the muscle image region generating unit 82, the trained model LM is used to generate muscle image regions GKN representing muscle tissue KN located in the first short-axis image display region G1 and / or the second short-axis image display region G2, respectively, from the image information of the first short-axis image and the second short-axis image displayed in the actual first short-axis image display region G1 and the second short-axis image display region G2. Then, in S5 corresponding to the display control unit 88, the muscle image region GKN is clearly displayed superimposed on the first short axis tomographic image and the second short axis tomographic image in the first short axis image display region G1 and the second short axis tomographic image display region G2 where the first short axis tomographic image and the second short axis tomographic image are displayed, respectively. Figure 5 shows this state.

[0054] Next, in S6 corresponding to the muscle brightness calculation unit 94, the muscle brightness MB of the muscle image area GKN, which is the average value of the brightness of each pixel P in the muscle image area GKN showing the muscle tissue KN shown in the first short-axis image and / or the second short-axis image, is calculated, and the muscle brightness MB is displayed as the muscle brightness MB in the character display area G4 of the image display device 30.

[0055] Next, in S7 corresponding to the body surface area calculation unit 93, the actual height H and weight W of each living body 14 to be measured, which have been input in advance, are read, and the body surface area BSA (m 2 ) is calculated.

[0056] Next, in S8 corresponding to the muscle cross-sectional area calculation unit 92 and the muscle mass calculation unit 95, the biceps brachii cross-sectional area MCSA (mm 2) is calculated. Next, the body surface area BSA (m 2 The biceps brachii muscle mass MCSAI (= MCSA / BSA) (mm) was calculated by dividing the biceps brachii muscle cross-sectional area MCSA by the muscle cross-sectional area MCSA. 2 / m 2 The biceps brachii cross-sectional area MCSA and the biceps brachii muscle mass MCSAI are displayed in the character display area G4 of the image display device 30.

[0057] Then, in S9 corresponding to the endothelial function index calculation unit 96, the ratio R (=MCSAI / MB) between the biceps muscle mass MCSAI of the muscle image area GKN calculated by S8 (muscle mass calculation unit 95) and the muscle brightness MB of the muscle image area GKN calculated by S6 (muscle brightness calculation unit 94) is calculated, and the ratio R is displayed in the character display area G4 of the image display device 30 as the endothelial function index MHI.

[0058] The following describes the method and results of an experiment conducted by the present inventors to demonstrate the basis for the endothelial function index MHI representing the endothelial function of the blood vessels 20 of the living body 14. (Experimental Method) Using a vascular endothelial function index measurement device (UNEXEF 18VG) manufactured by Unex Co., Ltd., the (%) FMD was calculated for each of 155 adult male and female patients (average age 68.2±10.8 years). Furthermore, the muscle image region generation unit 82, muscle cross-sectional area calculation unit 92, body surface area calculation unit 93, muscle brightness calculation unit 94, muscle mass calculation unit 95, and endothelial function index calculation unit 96 were used to generate a muscle image region GKN of the biceps brachii muscle, and the biceps brachii muscle cross-sectional area MCSA (mm 2 ), body surface area BSA (m 2 ), biceps muscle mass MCSAI (=MCSA / BSA) (mm 2 / m 2 ), muscle brightness (muscle quality) MB, which is the average brightness of all pixels P within the muscle image area GKN of the biceps brachii, and the ratio R of the biceps brachii muscle mass MCSAI to the muscle brightness MB of the muscle image area GKN (endothelial function index value MHI = MCSAI / MB) were calculated, and a simple correlation was calculated between the endothelial function index value MHI and (%)FMD.

[0059] (Experimental results) Figure 8 shows a two-dimensional coordinate system consisting of a horizontal axis X, which represents the ratio R (endothelial function index MHI) of the biceps brachii muscle mass MCSAI in the muscle image region GKN to the muscle brightness MB in the muscle image region GKN, and a vertical axis Y, which represents the (%) FMD value. It shows 155 plots of the endothelial function index MHI (=MCSAI / MB) and (%) FMD for 155 adults (78 men and 77 women), and the regression line y calculated from these plots, as shown in equation (2). The p-value for the regression line y in Figure 8 was less than 0.01. Since this p-value is less than 0.05, it is deemed to satisfy the 5% significance level. y=3.14+0.60x 〔R 2 =0.053, p<0.01) ···(2)

[0060] As described above, the endothelial function index MHI has a certain relationship with (%)FMD, which is used as an index of endothelial function. Therefore, it was found that endothelial function can be evaluated using the endothelial function index MHI, which can be measured in a short time. According to the experiments of the present inventors, the relationship between (%)FMD and the biceps brachii cross-sectional area MCSA (mm 2 ) and between (%)FMD and muscle brightness (muscle quality) MB in the muscle image area GKN.

[0061] As described above, the endothelial function index measuring device 22 of this embodiment generates a cross-sectional image from ultrasonic reflection signals from under the epidermis obtained by contacting the ultrasonic probe 24 with the skin 18 of the living body 14, generates a muscle image region GKN of the biceps brachii muscle located within the cross-sectional image, calculates the biceps brachii cross-sectional area MCSA from the muscle image region GKN, calculates the biceps brachii muscle mass MCSAI (=MCSA / BSA) from the biceps brachii cross-sectional area MCSA and the body surface area BSA, and calculates the biceps brachii muscle mass brightness ratio from the biceps brachii muscle mass MCSAI and the muscle brightness MB of the muscle image region GKN, and outputs it as the endothelial function index value MHI (=MCSAI / MB). In this way, the ultrasonic probe 24 is brought into contact with the skin 18 of the living body 14 to generate a cross-sectional image from the ultrasonic reflection signal, a muscle image area GKN of the muscles located within the cross-sectional image is generated, the biceps brachii muscle mass MCSAI is calculated from the biceps brachii muscle cross-sectional area MCSA and body surface area BSA of the muscle image area GKN, and the endothelial function index value MHI is calculated from the biceps brachii muscle mass MCSAI and muscle brightness MB.With short operation and calculation times, the endothelial function index value MHI can be quickly obtained in about 15 seconds, reducing the burden on the person being measured.

[0062] As shown in Figure 8, the endothelial function index (MHI) does not correlate very well with the (%)FMD value. This is likely due to risk factors that are common to the mechanisms underlying atherosclerosis, including but not limited to those underlying atherosclerosis. For example, it is known that the onset of sarcopenia, which occurs due to a decrease in skeletal muscle mass and muscle strength with aging, includes risk factors common to the mechanisms underlying atherosclerosis, such as oxidative stress and chronic inflammation. Therefore, particularly in elderly adults with cardiovascular risk, repeated and continuous measurement of the endothelial function index (MHI), which can be measured in a relatively short period of time, can contribute to predicting not only the tendency toward sarcopenia but also the tendency toward a decline in vascular endothelial function.

[0063] Furthermore, according to the endothelial function index measuring device 22 of this embodiment, the cross-sectional image is a short-axis image showing a cross section perpendicular to an artery beneath the skin 18 of the living body 14, and the muscle image region generating unit 82 generates the muscle image region GKN appearing in the short-axis image from image information appearing in the actual short-axis image using a trained model LM that has previously learned and stored a muscle image region GKN showing muscle tissue KN appearing in the short-axis image, and displays the muscle image region GKN superimposed on the short-axis image. This clearly generates the muscle image region GKN, which tends to be blurred when displayed in the first short-axis image display region G1 and / or the second short-axis image display region G2, and therefore accurately calculates the ratio R of the biceps brachii muscle mass MCSAI to the muscle brightness MB of the muscle image region GKN, i.e., the endothelial function index MHI.

[0064] According to the endothelial function index measurement device 22 of this embodiment, the learned model LM is a machine learning model MLM that has repeatedly undergone supervised learning using learning data in which a predetermined short-axis image is used as input data and the area of ​​muscle tissue KN identified by a skilled operator in the short-axis image is used as output data. This allows the area of ​​muscle tissue KN to be clearly identified in the first short-axis image display region G1 and the second short-axis image display region G2. This clearly identifies the muscle image region GKN in the short-axis image, allowing the ratio R of the biceps brachii muscle mass MCSAI to the muscle brightness MB in the muscle image region GKN, i.e., the endothelial function index MHI, to be accurately calculated.

[0065] In the endothelial function index measuring device 22 of this embodiment, the machine learning model MLM is a multi-layer neural network that has undergone deep learning, in which weighting coefficients between nodes belonging to adjacent layers are sequentially updated when input data is input to nodes in the input layer and output data indicating the area of ​​muscle tissue KN identified by an experienced operator in the short-axis image is repeatedly taught to nodes in the output layer. This allows the area of ​​muscle tissue KN to be more clearly identified in the short-axis image, and therefore the ratio R of the biceps brachii muscle mass MCSAI to the muscle brightness MB in the muscle image area GKN, i.e., the endothelial function index MHI, can be accurately calculated.

[0066] According to the endothelial function index measuring device 22 of this embodiment, in the multilayer neural network, brightness information for each pixel P constituting a predetermined short-axis image is input as input data to the input layer, and output data indicating the pixel P displaying the area of ​​muscle tissue KN is output from the output layer. This allows the area of ​​muscle tissue KN to be more clearly identified in the short-axis image, and therefore the ratio R of the biceps brachii muscle mass MCSAI to the muscle brightness MB in the muscle image area GKN, i.e., the endothelial function index MHI, can be accurately calculated.

[0067] According to the endothelial function index measuring device 22 of this embodiment, image recognition is performed on the output data output from the output layer of the multilayer neural network for each pixel P constituting the short-axis image, the texture t of the region to which the pixel P belongs is characterized from information on each pixel P and its adjacent pixels P, and a meaningful region division, whether it belongs to the region of muscle tissue KN or other tissue, is estimated for each pixel P based on the texture t. This reduces the number of nodes constituting the output layer of the multilayer neural network, making the configuration of the multilayer neural network more compact.

[0068] According to the endothelial function index measuring device 22 of this embodiment, the muscle cross-sectional area calculation unit 92 calculates the biceps brachii cross-sectional area MCSA of the muscle image region GKN from the total number of pixels P within the muscle image region GKN generated by the muscle image region generation unit 82. This allows the biceps brachii cross-sectional area MCSA to be obtained accurately and quantitatively.

[0069] The endothelial function index measuring device 22 of this embodiment includes a muscle brightness calculation unit 94 that calculates the average brightness of each pixel P in the muscle image region GKN generated by the muscle image region generation unit 82 as the muscle brightness MB of the image region of the biceps brachii muscle in the cross-sectional image. This allows accurate and quantitative acquisition of the muscle brightness MB of the muscle image region GKN as a parameter indicating the proportion of fatty tissue and connective tissue in the muscle.

[0070] The endothelial function index measuring device 22 of this embodiment includes a body surface area calculation unit 93 that calculates the body surface area BSA of the living body 14 based on the height H and weight W of the living body 14 using the DuPont formula (1) stored in advance. This makes it possible to easily obtain the body surface area BSA of the living body 14.

[0071] Although the preferred embodiment of the present invention has been described in detail above with reference to the drawings, the present invention is not limited to this and may be implemented with various modifications within the scope of the spirit of the present invention.

[0072] For example, the endothelial function index value measuring device 22 may measure the expansion rate (rate of change) (%) of the vascular lumen diameter, which indicates the flow-mediated dilation (FMD) after ischemia-reactive hyperemia, in order to evaluate the endothelial function of the blood vessel 20. MAX -d) / d], for example, the endothelial function testing device described in Patent Documents 1 and 2. In this case, there is an advantage that the hardware configuration is common. Note that "d" in the above formula indicates the vascular lumen diameter at rest (base diameter, resting diameter), and "d MAX " is the maximum value of the vascular lumen diameter after release from 5 minutes of ischemia. [Explanation of symbols]

[0073] 14: Living body, 18: Skin (epidermis), 20: Blood vessel (arterial blood vessel), 22: Endothelial function index measurement device, 24: Ultrasound probe, 82: Muscle image area generation unit, 92: Muscle cross-sectional area calculation unit, 93: Body surface area calculation unit, 94: Muscle brightness calculation unit, 95: Muscle mass calculation unit, 96: Endothelial function index calculation unit, KN: Muscle tissue, GKN: Muscle image area, LM: Trained model, MLM: Machine learning model, P: Pixel, t: Texture, MCSA: Biceps brachii cross-sectional area, H: Height, W: Weight, BSA: Body surface area, MCSAI: Biceps brachii muscle mass, MB: Muscle brightness, MHI: Endothelial function index, R: Ratio of biceps brachii muscle mass MCSAI to muscle brightness MB (MHI = MCSAI / MB)

Claims

1. 1. An apparatus for measuring an endothelial function index of an arterial blood vessel, which detects an ultrasonic reflection signal from an ultrasonic probe brought into contact with the epidermis of an upper arm of a living body, generates a cross-sectional image including the biceps brachii muscle and the brachial artery located under the epidermis from the ultrasonic reflection signal, and calculates an endothelial function index (MHI) of the arterial blood vessel of the living body from the cross-sectional image, a muscle image region generating unit that generates an image region of the biceps brachii muscle in the cross-sectional image; a muscle cross-sectional area calculation unit that calculates a biceps brachii cross-sectional area (MCSA) from the image region of the biceps brachii; a muscle mass calculation unit that calculates a biceps brachii muscle mass MCSAI by dividing the biceps brachii muscle cross-sectional area MCSA by a body surface area BSA of the living body; an endothelial function index value calculation unit that calculates a biceps brachii muscle mass brightness ratio by dividing the biceps brachii muscle mass MCSAI by a muscle brightness MB in an image region of the biceps brachii muscle in the cross-sectional image, and outputs the calculated ratio as the endothelial function index value MHI. An apparatus for measuring an endothelial function index value of an arterial blood vessel, characterized by:

2. the cross-sectional image is a short-axis image showing a cross section perpendicular to an artery under the skin of the living body, The muscle image region generating unit generates the muscle image region appearing in the short-axis image from image information appearing in the actual short-axis image using a trained model that has been trained and stored in advance for the muscle image region showing the muscle tissue appearing in the short-axis image, and displays the muscle image region superimposed on the short-axis image.

2. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 1.

3. The trained model is a machine learning model that has repeatedly performed supervised learning using training data in which a predetermined short-axis image is used as input data and the area of ​​the muscle tissue identified by an experienced operator in the short-axis image is used as output data.

3. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 2.

4. The machine learning model is a multi-layer neural network that has undergone deep learning, in which weight coefficients between nodes belonging to adjacent layers are sequentially updated when the input data is input to nodes in an input layer and the output data indicating the area of ​​muscle tissue identified by an experienced operator in the short-axis image is repeatedly taught to nodes in an output layer.

4. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 3.

5. In the multilayer neural network, brightness information for each pixel constituting the predetermined short-axis image is input to an input layer as the input data, and output data indicating pixels representing the muscle tissue area is output from an output layer.

5. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 4.

6. In the output data output from the output layer of the multilayer neural network, image recognition is performed on a pixel-by-pixel basis that constitutes the short-axis image, and the texture of the region to which the pixel belongs is characterized from information on each pixel and its neighboring pixels, and a meaningful region division, whether it belongs to the muscle tissue region or other tissue, is estimated for each pixel based on the texture.

6. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 5.

7. The muscle cross-sectional area calculation unit calculates the biceps brachii cross-sectional area MCSA of the muscle image region from the total number of pixels in the muscle image region generated by the muscle image region generation unit.

2. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 1.

8. a muscle brightness calculation unit that calculates an average value of brightness of each pixel in the muscle image region generated by the muscle image region generation unit as the muscle brightness MB of the image region of the biceps brachii muscle in the cross-sectional image.

2. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 1.

9. a body surface area calculation unit that calculates the body surface area BSA of the living body based on the height H and weight W of the living body using a DuPont formula (1) stored in advance.

2. The apparatus for measuring an endothelial function index value of arterial blood vessels according to claim 1. BSA=H ０．７２５ ×W ０．４２５ ×0.007184 ・・・(1)

Images