Ultrasonic diagnostic apparatus and control method of ultrasonic diagnostic apparatus

Abstract

The present application provides an ultrasonic diagnostic apparatus and a control method of the ultrasonic diagnostic apparatus capable of easily and quickly calculating cardiac output. The ultrasonic diagnostic apparatus includes an image classification section (25) that classifies an input image as an ultrasonic image that has captured which one of a first cross section and a second cross section, a first measurement section (26) that measures a cross-sectional area of a left ventricular outflow tract from the input image in a case where the input image is classified as the ultrasonic image that has captured the first cross section, a second measurement section (27) that measures a velocity-time integral value of a left ventricular outflow tract blood flow from the input image in a case where the input image is classified as the ultrasonic image that has captured the second cross section, and a cardiac output calculation section (28) that calculates the cardiac output using the cross-sectional area and the velocity-time integral value.

Description

Technical Field

[0001] This invention relates to an ultrasound diagnostic device for imaging the heart of a subject and a control method for the ultrasound diagnostic device. Background Technology

[0002] For a long time, so-called ultrasound diagnostic devices have been used to capture ultrasound images of cross-sections of the heart of the subject and to calculate so-called cardiac output by analyzing the captured ultrasound images. The calculation of cardiac output is usually performed by the following steps: (1) measuring the diameter of the left ventricular outflow tract in a first ultrasound image of a so-called parasternal left ventricular long axis section representing the mid-systolic phase of the heart and calculating the cross-sectional area of ​​the left ventricular outflow tract; (2) using a second ultrasound image representing a so-called apical five-chamber section or a so-called apical three-chamber section, calculating the velocity-time integral value of the blood flow in the left ventricular outflow tract by so-called pulse Doppler; (3) calculating the so-called stroke volume by multiplying the cross-sectional area of ​​the left ventricular outflow tract by the velocity-time integral value of the blood flow in the left ventricular outflow tract; and (4) calculating cardiac output by multiplying the stroke volume by the heart rate.

[0003] Therefore, calculating cardiac output requires multiple steps, and various techniques have been developed to facilitate the measurement of stroke volume and cardiac output. For example, Patent Document 1 discloses a method that inputs a Doppler waveform image into a learned model in so-called machine learning, causing the learned model to automatically output the velocity-time integral value of blood flow.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2024-129281

[0005] However, even when using the technology disclosed in Patent Document 1, users still need to perform multiple steps, which can sometimes take a lot of time to calculate cardiac output. Summary of the Invention

[0006] The present invention was made to eliminate these previous problems, and its purpose is to provide an ultrasound diagnostic device and a control method for the ultrasound diagnostic device that can easily and quickly calculate cardiac output.

[0007] The above objectives can be achieved through the following structure.

[0008] [1] An ultrasound diagnostic device that measures cardiac output based on multiple frames of ultrasound images of different first and second cross-sections of the heart of a subject, the ultrasound diagnostic device comprising:

[0009] The image classification unit classifies the input image to determine whether it is an ultrasound image of the first section or the second section that was captured.

[0010] The first measurement unit measures the cross-sectional area of ​​the left ventricular outflow tract based on the input image, in cases where the image classification unit classifies the input image as an ultrasound image of the first cross section.

[0011] The second measurement unit, when the image classification unit classifies the input image as an ultrasound image of the second cross-section, measures the velocity-time integral value of the blood flow in the left ventricular outflow tract based on the input image; and

[0012] The cardiac output calculation unit calculates cardiac output using the cross-sectional area measured by the first measurement unit and the velocity-time integral value measured by the second measurement unit.

[0013] [2] According to the ultrasonic diagnostic device described in [1], wherein,

[0014] The image classification unit classifies the input image by using the parasternal left ventricular long axis section as the first section and the apical five-chamber section or the apical three-chamber section as the second section.

[0015] [3] According to the ultrasonic diagnostic device described in [2], wherein,

[0016] The first measurement unit includes:

[0017] The left ventricular outflow tract diameter measurement unit measures the diameter of the left ventricular outflow tract based on the input image.

[0018] The mid-shrinkage frame detection unit detects mid-shrinkage frames based on the input image; and

[0019] The cross-sectional area calculation unit calculates the cross-sectional area of ​​the left ventricular outflow tract based on the diameter of the left ventricular outflow tract measured by the left ventricular outflow tract diameter measurement unit in the mid-systolic frame detected by the mid-systolic frame detection unit.

[0020] [4] According to the ultrasonic diagnostic device described in [3], wherein,

[0021] The left ventricular outflow tract diameter measuring unit measures the left ventricular outflow tract diameter at a predetermined distance from the aortic valve insertion point toward the heart.

[0022] [5] The ultrasound diagnostic device according to [3] or [4], wherein,

[0023] The mid-systolic frame detection unit detects the frame with the largest angle between a pair of aortic valves as the mid-systolic frame.

[0024] [6] The ultrasound diagnostic apparatus according to any one of [3] to [5], wherein,

[0025] The first measurement unit includes a frame reliability calculation unit that calculates the reliability of frames.

[0026] The left ventricular outflow tract diameter measuring unit measures the left ventricular outflow tract diameter based on frames whose reliability, calculated by the frame reliability calculation unit, is above a predetermined reliability threshold.

[0027] [7] The ultrasonic diagnostic device according to [6], wherein,

[0028] The frame reliability calculation unit calculates reliability based on the confidence level of the classification in the image classification unit, or based on the time series change of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit.

[0029] [8] According to the ultrasonic diagnostic device described in [2], wherein,

[0030] The second measurement unit includes:

[0031] The contraction period recognition unit identifies the contraction period based on the input image;

[0032] The gate setting unit sets a Doppler gate in the input image that is identified as a contraction period by the contraction period recognition unit;

[0033] The Doppler waveform acquisition unit acquires the Doppler waveform in the Doppler gate set by the gate setting unit; and

[0034] The velocity-time integral value calculation unit calculates the velocity-time integral value based on the Doppler waveform acquired by the Doppler waveform acquisition unit.

[0035] [9] The ultrasonic diagnostic apparatus according to [8], wherein,

[0036] The systolic phase identification unit identifies the systolic phase based on any one of the following: mitral valve closure, presence or absence of aortic valve, left ventricular area, and depth of a reference point set in the input image.

[0037]

[10] The ultrasonic diagnostic apparatus according to [8] or [9], wherein,

[0038] The first measurement unit includes a left ventricular outflow tract diameter measurement unit that measures the left ventricular outflow tract diameter based on an input image.

[0039] The gate setting unit sets the Doppler gate in the input image based on the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit in the first measuring unit.

[0040]

[11] According to the ultrasonic diagnostic device described in [3], wherein,

[0041] The cardiac output calculation unit calculates the heart rate based on the number of plotting points of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit between two consecutive mid-systolic frames detected by the mid-systolic frame detection unit.

[0042] The cardiac output is calculated based on the cross-sectional area measured by the first measuring unit, the velocity-time integral value measured by the second measuring unit, and the calculated heart rate.

[0043]

[12] The ultrasonic diagnostic device according to [3] comprises:

[0044] The error detection unit detects errors in the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit.

[0045]

[13] The ultrasonic diagnostic apparatus according to

[12] , wherein,

[0046] The error detection unit detects errors in the left ventricular outflow tract diameter by comparing the time-series change curve of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit with a predetermined curve, or by comparing the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit with a predetermined set value.

[0047]

[14] The ultrasound diagnostic apparatus according to

[13] comprises:

[0048] The correction unit corrects the left ventricular outflow tract inner diameter that was previously detected as erroneous by the error detection unit, based on the cross-sectional area of ​​the left ventricular outflow tract calculated by the cross-sectional area calculation unit.

[0049]

[15] The ultrasonic diagnostic apparatus according to [8] comprises:

[0050] The error detection unit detects errors in the speed-time integral value calculated by the speed-time integral value calculation unit.

[0051]

[16] According to the ultrasonic diagnostic device described in

[15] , wherein,

[0052] The error detection unit detects errors in the speed-time integral value by comparing the Doppler curve obtained by tracking the Doppler waveform acquired by the Doppler waveform acquisition unit with a predetermined curve, or by comparing the speed-time integral value calculated by the speed-time integral value calculation unit with a predetermined set value.

[0053]

[17] The ultrasonic diagnostic apparatus according to

[16] comprises:

[0054] The correction unit corrects the speed-time integral value that was detected as erroneous by the error detection unit based on the speed-time integral value calculated by the speed-time integral value calculation unit in the past.

[0055]

[18] A control method for an ultrasound diagnostic device, comprising measuring cardiac output based on multiple frames of ultrasound images of different first and second cross-sections of the heart of a subject, the control method comprising:

[0056] The input image is classified based on whether it is an ultrasound image of the first section or the second section that was captured.

[0057] The cross-sectional area of ​​the left ventricular outflow tract is measured based on the input image, which is classified as an ultrasound image of the first section.

[0058] The velocity-time integral of blood flow in the left ventricular outflow tract was measured based on the input image, which was classified as an ultrasound image captured at the second cross-section; and

[0059] The cardiac output is calculated using the measured cross-sectional area and the measured velocity-time integral.

[0060] Invention Effects

[0061] In this invention, the ultrasound diagnostic apparatus includes: an image classification unit that classifies an input image as either a first cross-section or a second cross-section of an ultrasound image; a first measurement unit that, when the image classification unit classifies the input image as an ultrasound image of the first cross-section, measures the cross-sectional area of ​​the left ventricular outflow tract based on the input image; a second measurement unit that, when the image classification unit classifies the input image as an ultrasound image of the second cross-section, measures the velocity-time integral value of the blood flow in the left ventricular outflow tract based on the input image; and a cardiac output calculation unit that calculates cardiac output using the cross-sectional area measured by the first measurement unit and the velocity-time integral value measured by the second measurement unit, thus enabling easy and rapid calculation of cardiac output. Attached Figure Description

[0062] Figure 1 This is a block diagram illustrating the structure of the ultrasonic diagnostic device according to Embodiment 1 of the present invention.

[0063] Figure 2 This is a block diagram illustrating the internal structure of the transceiver circuit in Embodiment 1 of the present invention.

[0064] Figure 3 This is a block diagram showing the internal structure of the image generation unit in Embodiment 1 of the present invention.

[0065] Figure 4 This is a diagram showing an example of an ultrasound image representing the long axis section of the left ventricle beside the sternum.

[0066] Figure 5 This is a diagram showing an example of an ultrasound image representing a five-chamber cross-section of the apex of the heart.

[0067] Figure 6 This is a block diagram showing the internal structure of the first measuring unit in Embodiment 1 of the present invention.

[0068] Figure 7 This is a diagram showing an example of a measurement line set in an ultrasound image representing the long axis section of the left ventricle beside the sternum.

[0069] Figure 8 This is a block diagram showing the internal structure of the second measuring unit in Embodiment 1 of the present invention.

[0070] Figure 9 This is a diagram showing an example of a Doppler gate set up in an ultrasound image representing a five-chamber cross-section of the apex of the heart.

[0071] Figure 10 This is a flowchart illustrating the operation of the ultrasonic diagnostic device according to Embodiment 1 of the present invention.

[0072] Figure 11 This is a diagram illustrating an example of a series of processing steps displayed on a monitor.

[0073] Figure 12 This is a block diagram showing the internal structure of the first measuring unit in a modified embodiment 1 of the present invention.

[0074] Figure 13 This is a block diagram illustrating the structure of the ultrasonic diagnostic device according to Embodiment 2 of the present invention.

[0075] Figure 14 This is a graph showing the time-series variation of the left ventricular outflow tract diameter and an example of a predetermined curve compared to the time-series variation curve.

[0076] Symbol Explanation

[0077] 1-Ultrasonic probe, 2-Device body, 11-Voltage array, 12-Transceiver circuit, 21-Image generation unit, 22-Display control unit, 23-Display, 24-Image memory, 25-Image classification unit, 26, 26A-First measurement unit, 27-Second measurement unit, 28-Cardiometry output calculation unit, 29-Measurement value memory, 30, 30B-Device control unit, 31-Input device, 32-Image acquisition unit, 33, 33B-Processor, 41-Pulse generator, 42-Amplifier, 43-AD converter, 44-Beamformer 45-Signal processing unit, 46-DSC, 47-Image processing unit, 51-Left ventricular outflow tract diameter measurement unit, 52-Mid-systolic frame detection unit, 53-Cross-area calculation unit, 54-Frame reliability calculation unit, 61-Systolic period identification unit, 62-Gate setting unit, 63-Doppler waveform acquisition unit, 64-Velocity-time integral value calculation unit, 71-Error detection unit, 72-Correction unit, C1-Time series change curve, C2-Predetermined curve, ML1-Measurement line, G-Doppler gate, T-Left ventricular outflow tract, U1, U2-Ultrasound images. Detailed Implementation

[0078] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0079] The following description of the constituent elements is based on a representative embodiment of the present invention, but the present invention is not limited to this embodiment.

[0080] In addition, in this specification, the numerical range represented by “~” indicates the range including the values ​​recorded before and after “~” as the lower limit and upper limit.

[0081] In this specification, "same" or "identical" includes the range of error that is generally permissible in the technical field.

[0082] Implementation Method 1

[0083] Figure 1 The diagram shows the structure of an ultrasonic diagnostic apparatus according to Embodiment 1 of the present invention. The ultrasonic diagnostic apparatus includes an ultrasonic probe 1 and an apparatus body 2 that are connected to each other via so-called wired communication or so-called wireless communication.

[0084] The ultrasonic probe 1 has a transducer array 11 and a transceiver circuit 12 connected thereto.

[0085] The device main body 2 includes an image generation unit 21 connected to a transceiver circuit 12. A display control unit 22 and a display 23 are sequentially connected to the image generation unit 21. An image memory 24 and an image classification unit 25 are also connected to the image generation unit 21. A first measurement unit 26 and a second measurement unit 27 are connected to the image classification unit 25. A cardiac output calculation unit 28 is connected to the first measurement unit 26 and the second measurement unit 27. A measurement value memory 29 and the display control unit 22 are connected to the first measurement unit 26, the second measurement unit 27, and the cardiac output calculation unit 28. A device control unit 30 is connected to the transceiver circuit 12, the image generation unit 21, the display control unit 22, the image memory 24, the image classification unit 25, the first measurement unit 26, the second measurement unit 27, the cardiac output calculation unit 28, and the measurement value memory 29. An input device 31 is connected to the device control unit 30.

[0086] The image acquisition unit 32 is composed of the transceiver circuit 12 and the image generation unit 21. Furthermore, the processor 33 of the device main body 2 is composed of the image generation unit 21, the display control unit 22, the image classification unit 25, the first measurement unit 26, the second measurement unit 27, and the output value calculation unit 28.

[0087] The transducer array 11 of the ultrasonic probe 1 has multiple ultrasonic transducers arranged in one-dimensional or two-dimensional order. These ultrasonic transducers emit ultrasonic waves according to the drive signal supplied from the transceiver circuit 12, and receive ultrasonic echoes from the subject to output signals based on the ultrasonic echoes. Each ultrasonic transducer is constructed, for example, by forming electrodes at both ends of a piezoelectric element, which is composed of piezoelectric ceramics represented by PZT (Lead Zirconate Titanate), polymer piezoelectric elements represented by PVDF (Poly Vinylidene Di Fluoride), and piezoelectric single crystals represented by PMN-PT (Lead Magnesium Niobate-Lead Titanate solid solution).

[0088] The image acquisition unit 32, which consists of the transceiver circuit 12 and the image generation unit 21, acquires multiple frames of ultrasound images as dynamic images of the heart of the subject by using the ultrasound probe 1 to transmit and receive ultrasound beams.

[0089] Under the control of the device control unit 30, the transceiver circuit 12 transmits ultrasonic waves from the transducer array 11 and generates an acoustic signal based on the received signal acquired by the transducer array 11. For example... Figure 2As shown, the transceiver circuit 12 has a pulse generator 41 connected to the oscillator array 11, an amplifier 42, an AD (Analog to Digital) converter 43, and a beamformer 44 connected in series from the oscillator array 11.

[0090] The pulse generator 41 includes, for example, multiple pulse generators, and adjusts the delay of each drive signal and supplies it to the multiple ultrasonic transducers in such a way that an ultrasonic beam is formed from ultrasonic waves emitted from the multiple ultrasonic transducers of the transducer array 11, according to a transmission delay mode selected based on a control signal from the device control unit 30. Thus, if a pulsed or continuous wave voltage is applied to the electrodes of the ultrasonic transducers of the transducer array 11, the piezoelectric element expands and contracts, generating pulsed or continuous wave ultrasonic waves from each ultrasonic transducer, and the composite wave of these ultrasonic waves forms an ultrasonic beam.

[0091] The transmitted ultrasonic beam is reflected by an object such as a part of the subject being examined and propagates toward the transducer array 11 of the ultrasonic probe 1. The ultrasonic echoes propagating toward the transducer array 11 are received by each ultrasonic transducer constituting the transducer array 11. At this time, each ultrasonic transducer constituting the transducer array 11 expands and contracts by receiving the propagating ultrasonic echoes to generate received signals as electrical signals, and outputs these received signals to the amplification unit 42.

[0092] The amplification unit 42 amplifies the signals input from each ultrasonic transducer constituting the transducer array 11 and sends the amplified signals to the AD converter 43. The AD converter 43 converts the signals sent from the amplification unit 42 into digital received data. The beamformer 44 performs a so-called reception focusing process by applying a delay to each received data received from the AD converter 43 and adding them together. Through this reception focusing process, an acoustic signal is obtained by performing phase-integrated addition of each received data converted by the AD converter 43, and the focus of the ultrasonic echo is reduced.

[0093] like Figure 3 As shown, the image generation unit 21 has a structure in which the signal processing unit 45, the DSC (Digital Scan Converter) 46 and the image processing unit 47 are connected in series.

[0094] The signal processing unit 45 uses the sound velocity value set by the device control unit 30 and, based on the depth of the ultrasonic wave reflection position, corrects the attenuation caused by distance on the sound signal received from the transceiver circuit 12, and then performs envelope detection processing to generate tomographic image information related to the tissue in the subject body, i.e., a B-mode image signal.

[0095] DSC46 converts (rasterizes) the B-mode image signal generated by the signal processing unit 45 into an image signal that conforms to the scanning method of a general television signal.

[0096] After performing various necessary image processing operations, such as grayscale processing, on the B-mode image signal input from the DSC46, the image processing unit 47 sends the B-mode image signal to the display control unit 22, the image memory 24, and the image classification unit 25. Hereinafter, the B-mode image signal for which image processing has been performed by the image processing unit 47 will be referred to as an ultrasound image.

[0097] In this invention, the image acquisition unit 32 acquires multiple frames of ultrasound images of different first and second cross-sections of the heart of the subject. For example, such as Figure 4 As schematically shown, the image acquisition unit 32 can acquire an ultrasound image representing a so-called parasternal left ventricular long-axis section as an ultrasound image U1 representing the first section. The parasternal left ventricular long-axis section includes the left ventricle, the left ventricular outflow tract T, the left atrium, and the right ventricle. Furthermore, for example, as... Figure 5 Schematic illustration shows that the image acquisition unit 32 can acquire an ultrasound image representing a so-called apical five-chamber cross-section as an ultrasound image U2 representing a second cross-section. The apical five-chamber cross-section is a cross-section of the heart including the left ventricle, the left ventricular outflow tract T, the left atrium, the right ventricle, and the right atrium. Although not shown, the image acquisition unit 32 can also acquire an ultrasound image representing a so-called apical three-chamber cross-section instead of the ultrasound image U2 representing the apical five-chamber cross-section. The apical three-chamber cross-section is a cross-section of the heart including the left ventricle, the left atrium, and the right ventricle. Both the apical five-chamber cross-section and the apical three-chamber cross-section include the left ventricular outflow tract T.

[0098] Image memory 24 is a memory that stores multiple frames of ultrasound images acquired by image acquisition unit 32. Users can use it, for example, to check the multiple frames of ultrasound images stored in image memory 24 after inspection. Image memory 24 can be, for example, a recording medium such as flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), FD (Flexible Disk), MO (Magneto-Optical disk), MT (Magnetic Tape), RAM (Random Access Memory), CD (Compact Disc), DVD (Digital Versatile Disc), SD card (Secure Digital card), or USB memory (Universal Serial Bus memory).

[0099] The image classification unit 25 classifies the input image from the multiple frames of ultrasound images acquired by the image acquisition unit 32 to determine which of the first and second cross-sections the ultrasound image captured. For example, the image classification unit 25 can classify the ultrasound image by pre-storing template data representing typical image patterns of the parasternal left ventricular long-axis section, the apical five-chamber section, and the apical three-chamber section, and using a so-called template matching method that searches ultrasound images using the template data to identify any one of the three cross-sections. The image classification unit 25 may have a learned model in machine learning, for example, that wants to learn multiple ultrasound images representing the parasternal left ventricular long-axis section, the apical five-chamber section, and the apical three-chamber section, and can also output a classification result by inputting ultrasound images into the learned model. The image classification unit 25 can use algorithms such as SIFT (Scale-Invariant Feature Transform) or SVM (Support Vector Machine).

[0100] If the image classification unit 25 classifies the input image as an ultrasound image U1 capturing the first cross-section, the first measurement unit 26 measures the cross-sectional area (CSA) of the left ventricular outflow tract T based on the classified input image. Figure 6 As shown, the first measuring unit 26 has a structure in which the left ventricular outflow tract diameter measuring unit 51, the mid-systolic frame detection unit 52, and the cross-sectional area calculation unit 53 are connected in series.

[0101] The left ventricular outflow tract diameter measuring unit 51 measures the diameter of the left ventricular outflow tract T, i.e., the left ventricular outflow tract diameter, based on the input image. For example, the left ventricular outflow tract diameter measuring unit 51 detects the aortic valve connected to the left ventricular outflow tract T by analyzing the input image, and positions it within a predetermined distance, for example, within 1 cm to 2 cm, from the detected aortic valve toward the heart, more specifically toward the left ventricle. Figure 7 The measurement line ML1 is shown. The left ventricular outflow tract diameter measuring unit 51 measures the length of the configured measurement line ML1 as the left ventricular outflow tract diameter. In addition, the left ventricular outflow tract diameter measuring unit 51 can detect the aortic valve from the input image using known algorithms such as template matching or machine learning.

[0102] The mid-systolic frame detection unit 52 detects mid-systolic frames representing the mid-systolic phase of the heart based on the input image. Guidelines published by the American Society of Echocardiography (ASE) and others define the phase with the greatest aortic valve opening as the mid-systolic phase of the heart. Therefore, the mid-systolic frame detection unit 52 can, for example, detect frames with the largest angle between a pair of aortic valves as mid-systolic frames. In this case, the mid-systolic frame detection unit 52 can, for example, use known algorithms such as template matching or machine learning methods to detect the aortic valves from multiple ultrasound images.

[0103] The cross-sectional area calculation unit 53 calculates the cross-sectional area of ​​the left ventricular outflow tract T based on the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit 51 in the mid-systolic frame detected by the mid-systolic frame detection unit 52. At this time, the cross-sectional area calculation unit 53 can, for example, calculate the cross-sectional area of ​​the left ventricular outflow tract T by treating the cross-sectional shape of the left ventricular outflow tract T as a circle.

[0104] When the image classification unit 25 classifies the input image as an ultrasound image U2 capturing the second cross-section, the second measurement unit 27 measures the velocity-time integral value of the blood flow in the left ventricular outflow tract T based on the input image. Figure 8 As shown, the second measuring unit 27 has a structure in which a contraction period identification unit 61, a gate setting unit 62, a Doppler waveform acquisition unit 63, and a velocity-time integral value calculation unit 64 are connected in series.

[0105] The systolic period recognition unit 61 identifies the systolic period of the heart based on the input image. During systole, the mitral valve closes to prevent backflow of blood into the left atrium. Therefore, the systolic period recognition unit 61 can, for example, analyze multiple frames of ultrasound images U2 and identify the period corresponding to the mitral valve closure in those frames as the systolic period. In this case, the systolic period recognition unit 61 can detect the mitral valve from the multiple frames of ultrasound images U2 using known algorithms such as template matching or machine learning.

[0106] Furthermore, while it is known that the aortic valve opens during the heart's systole, it is typically difficult to depict the open aortic valve in the ultrasound image U2 representing the second cross-section. Therefore, the systole recognition unit 61, for example, performs processing to detect the aortic valve from multiple frames of ultrasound images U2, and can identify the period corresponding to the ultrasound image U2 frames where the aortic valve is not detected as the heart's systole. In this case, the systole recognition unit 61 can, for example, use known algorithms such as template matching or machine learning methods to perform the processing to detect the aortic valve from multiple frames of ultrasound images U2.

[0107] Furthermore, it is known that the volume of the left ventricle decreases during the systole of the heart. Therefore, the systole identification unit 61 can, for example, measure the area of ​​the left ventricle in the multi-frame ultrasound images U2 and identify the period during which the area of ​​the left ventricle decreases in the time series as the systole of the heart. At this time, the systole identification unit 61 can, for example, detect the left ventricle from the multi-frame ultrasound images U2 using known algorithms such as template matching or machine learning. Moreover, the systole identification unit 61 can, for example, calculate the area of ​​the left ventricle by measuring the number of pixels in the image region of the left ventricle detected in the ultrasound images U2.

[0108] Furthermore, since the position of the aorta changes with the heartbeat, the depth of the anatomical structures including the aorta within a defined range varies in the multi-frame ultrasound image U2. The anatomical structures including the aorta within a defined range change in depth in the multi-frame ultrasound image U2 during cardiac systole. Therefore, the systole identification unit 61 can, for example, set at least one reference point near the aorta in the multi-frame ultrasound image U2 and identify the period during which the at least one reference point deepens as the cardiac systole. At this time, the systole identification unit 61 can, for example, set at least one reference point at the so-called aortic valve root or the center of the aorta.

[0109] Thus, the systolic period recognition unit 61 can recognize the systolic period of the heart based on any one of the following: closure of the mitral valve, depiction of the presence or absence of the aortic valve, area of ​​the left ventricle, and depth of at least one reference point set in the input image.

[0110] like Figure 9As shown, the gate setting unit 62 sets a Doppler gate G in the multi-frame ultrasound image U2, which is identified as the systolic phase by the systolic phase recognition unit 61, i.e., the input image. The gate setting unit 62 can detect the left ventricular outflow tract T from the input image using known algorithms such as template matching or machine learning, and configure the Doppler gate G on the detected left ventricular outflow tract T.

[0111] Furthermore, the gate setting unit 62 can also set a Doppler gate G in the input image based on the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit 51 in the first measuring unit 26. At this time, the gate setting unit 62 can, for example, arrange measurement lines ML1 at multiple positions of the left ventricular outflow tract T in the ultrasound image U2, and arrange the Doppler gate G at the position of the measurement line ML1 that can obtain the measurement value closest to the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit 51.

[0112] The Doppler waveform acquisition unit 63 acquires the Doppler waveform in the Doppler gate G set by the gate setting unit 62. At this time, the Doppler waveform acquisition unit 63 mixes a carrier signal at a reference frequency with an acoustic signal received from the transceiver circuit 12, performs quadrature detection on the acoustic signal and converts it into a complex signal, and removes the frequency components originating from the movement of the body's internal tissues from the complex signal. The Doppler waveform acquisition unit 63 also performs frequency analysis by performing Fourier transform on the complex signals of multiple sample points to determine the blood flow velocity, generates a spectral signal, and acquires a Doppler waveform image signal representing the Doppler waveform by aligning the spectral signal to the time axis and using brightness to represent the magnitude of each frequency component. The Doppler waveform image signal is called a so-called Doppler image, representing the time-varying waveform of blood flow velocity. The horizontal axis represents the time axis, the vertical axis represents the Doppler frequency shift (i.e., flow velocity), and the brightness of the waveform represents the power of each frequency component.

[0113] The velocity-time integral calculation unit 64 calculates the velocity-time integral value (VTI) based on the Doppler waveform acquired by the Doppler waveform acquisition unit 63. The velocity-time integral calculation unit 64 determines a heartbeat period within the Doppler waveform and calculates the velocity-time integral value by integrating the flow velocity value in the Doppler waveform over time within that heartbeat period. For example, the velocity-time integral calculation unit 64 determines repeating units of the Doppler waveform by analyzing the Doppler waveform and can define the period to which the determined repeating unit belongs as a heartbeat period.

[0114] Furthermore, the processing of calculating the velocity-time integral value by the Doppler waveform acquisition unit 63 and the velocity-time integral value calculation unit 64 is performed after the transceiver circuit 12 and the device control unit 30 send a drive signal to the transducer array 11 in a pulsed manner to transmit ultrasonic waves.

[0115] The cardiac output calculation unit 28 calculates cardiac output using the cross-sectional area of ​​the left ventricular outflow tract T measured by the first measurement unit 26 and the velocity-time integral value of the blood flow in the left ventricular outflow tract measured by the second measurement unit 27. More specifically, the cardiac output calculation unit 28 calculates the so-called stroke volume by multiplying the cross-sectional area of ​​the left ventricular outflow tract T by the velocity-time integral value, and calculates cardiac output by multiplying the calculated stroke volume by the heart rate. For example, the cardiac output calculation unit 28 can determine the repetitive waveform in the time-varying waveform of blood flow velocity acquired by the Doppler waveform acquisition unit 63 of the second measurement unit 27, and calculate the reciprocal of the period of the determined repetitive waveform as the heart rate.

[0116] Furthermore, the cardiac output calculation unit 28 can also calculate the heart rate based on the number of plotting points of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit 51 between two consecutive mid-systolic frames detected by the mid-systolic frame detection unit 52 of the first measurement unit 26.

[0117] The measurement value memory 29 stores data such as the inner diameter of the left ventricular outflow tract and the cross-sectional area of ​​the left ventricular outflow tract T measured by the first measurement unit 26, the Doppler waveform and the velocity-time integral value of the blood flow in the left ventricular outflow tract measured by the second measurement unit 27, and the stroke volume and cardiac output calculated by the cardiac output calculation unit 28. The measurement value memory 29 can be a recording medium such as flash memory, HDD, SSD, FD, MO disk, MT, RAM, CD, DVD, SD card, or USB memory.

[0118] If the user places the ultrasound probe 1 on the surface of the subject's body to capture ultrasound images U1 and U2, the device control unit 30 controls each part to automatically perform the subsequent processing in the image classification unit 25, the first measurement unit 26, the second measurement unit 27, and the cardiac output calculation unit 28. In this way, the user can automatically obtain the cardiac output measurement value simply by placing the ultrasound probe 1 on the surface of the subject's body. Therefore, the user does not need to perform cumbersome operations to measure cardiac output, and can easily and quickly measure cardiac output.

[0119] Under the control of the device control unit 30, the display control unit 22 performs prescribed processing on the ultrasound images U1 and U2 acquired by the image acquisition unit 32, the measurement line ML1 set by the first measurement unit 26 and the obtained measurement value, the Doppler gate G set by the second measurement unit 27 and the obtained measurement value, and the cardiac output measurement value obtained by the cardiac output calculation unit 28, and displays them on the display 23.

[0120] Under the control of the display control unit 22, the display 23 displays ultrasound images U1 and U2, measurement line ML1, Doppler gate G, cross-sectional area of ​​the left ventricular outflow tract T, velocity-time integral value, and cardiac output, etc., and has a display device such as LCD (Liquid Crystal Display) or Organic Electroluminescence Display.

[0121] The input device 31 is a device for users to perform input operations, such as a keyboard, mouse, trackball, touchpad, and touch sensor superimposed on the display 23.

[0122] In this embodiment, each process in processor 33 is executed by any computer. Furthermore, any computer can execute these processes via a processor as hardware, a program as software, or a combination thereof. In this case, the processor is configured to cooperate with the program to execute the various processes in this embodiment and can function as a unit or means in this embodiment. Moreover, the execution order of the processor-based processes is not limited to the order described and can be appropriately varied. The arbitrary computer can be a general-purpose computer, a special-purpose computer, a workstation, or other system capable of executing the processes.

[0123] The processor 33 may also be composed of one or more hardware components, and the type of hardware is not limited. For example, the processor 33 may be composed of programmable logic devices such as CPUs (Central Processing Units), MPUs (Micro Processing Units), FPGAs (Field Programmable Gate Arrays), dedicated circuits for performing specific processes such as ASICs (Application Specific Integrated Circuits), GPUs (Graphics Processing Units), or NPUs (Neural Processing Units). Furthermore, the type of hardware can be a combination of different types of hardware. When multiple hardware components are configured to execute one or more processes of a certain processor, these multiple hardware components may reside in physically separate devices or in the same device. Moreover, in any embodiment, the order in which the processor 33 executes each process is not limited to the above order and can be appropriately varied. Additionally, the hardware may be composed of circuits, such as those combining semiconductor elements.

[0124] Alternatively, a program can be software such as firmware or microcode. Furthermore, a program can be, for example, a group of program modules, each of whose functions can be implemented by a processor configured to execute those functions. A program can also be program code or multiple code segments stored on one or more non-transitory computer-readable media (e.g., storage media or other storage devices). A program can be segmented and stored on multiple non-transitory computer-readable media existing in physically separate devices. Program code or code segments can represent any combination of procedures, functions, subroutines, routines, subroutines, modules, software packages, classes, commands, data structures, or program statements. Program code or code segments can be connected to other code segments or hardware circuitry by sending and receiving information, data, arguments, parameters, or the contents of memory.

[0125] Next, refer to Figure 10 The flowchart shown illustrates the operation of the ultrasound diagnostic apparatus according to Embodiment 1. Here, an example is given of acquiring an ultrasound image representing the long axis section of the left ventricle beside the sternum as ultrasound image U1 representing the first section, and acquiring an ultrasound image representing the five chambers of the apex of the heart as ultrasound image U2 representing the second section.

[0126] In step S1, the user places the ultrasonic probe 1 on the surface of the subject, and the image acquisition unit 32 acquires multiple frames of ultrasonic images. At this time, under the control of the device control unit 30, ultrasonic waves are transmitted and received from multiple transducers of the transducer array 11 according to the drive signal from the pulse generator 41 of the transducer circuit 12 of the ultrasonic probe 1. The ultrasonic echo from the subject is received by multiple transducers of the transducer array 11, and the received signal, as an analog signal, is output to the amplification unit 42 for amplification. The signal is then converted to analog signal by the AD conversion unit 43 to acquire the received data.

[0127] The received data is focused by beamformer 44, and the resulting acoustic signal is sent to image generation unit 21 of device body 2, whereby image generation unit 21 generates an ultrasonic image. At this time, signal processing unit 45 of image generation unit 21 performs attenuation correction and envelope detection processing on the acoustic signal based on the depth of the ultrasonic wave reflection position, and DSC 46 converts it into an image signal conforming to the scanning mode of general television signals. Image processing unit 47 then performs various required image processing such as grayscale processing.

[0128] In step S2, the image classification unit 25 classifies whether the multiple ultrasonic images U1 and U2 acquired in step S1 were captured as ultrasonic image U1 of the first cross-section or ultrasonic image U2 of the second cross-section. The image classification unit 25 can classify the multiple ultrasonic images U1 and U2, for example, by template matching or by using a machine learning-based learned model.

[0129] In step S3, the device control unit 30 determines whether the multiple ultrasonic images acquired in step S1 were classified as ultrasonic images U1 of the first cross-section in step S2. If it is determined in step S3 that the multiple ultrasonic images were classified as ultrasonic images U1 of the first cross-section, then proceed to step S4.

[0130] In step S4, the first measuring unit 26 measures the cross-sectional area of ​​the left ventricular outflow tract T based on the ultrasound images U1 of the first section captured in step S2. The first measuring unit 26 measures the cross-sectional area of ​​the left ventricular outflow tract T in multiple frames of ultrasound images U1, such as... Figure 7 As shown, a measuring line ML1 is positioned on the left ventricular outflow tract T, and the length of the positioned measuring line ML1 is measured as the inner diameter of the left ventricular outflow tract. At this time, the first measuring unit 26 can, for example, detect the aortic valve using template matching or machine learning methods, and automatically position the measuring line ML1 within a predetermined range of 1 cm to 2 cm from the detected aortic valve toward the left ventricle. The first measuring unit 26, for example, detects the frame with the largest angle between a pair of aortic valves as the mid-systolic frame, and calculates the cross-sectional area of ​​the left ventricular outflow tract T based on the inner diameter of the left ventricular outflow tract in the mid-systolic frame.

[0131] In step S5, the device control unit 30 determines whether the multiple ultrasonic images acquired in step S1 have been classified as ultrasonic images U2 representing the second cross-section in step S2. If, in step S5, it is determined that the multiple ultrasonic images have not been classified as ultrasonic images U2 representing the second cross-section, the process returns to step S1 and acquires multiple ultrasonic images again to obtain ultrasonic images U2 representing the second cross-section. In this case, the user adjusts the position and orientation of the ultrasonic probe 1 to acquire ultrasonic images U2 representing the second cross-section. In the subsequent step S2, the image classification unit 25 classifies the multiple ultrasonic images.

[0132] In step S3, if the multiple frames of ultrasonic images acquired in step S1 are not classified as ultrasonic images U1 of the first cross section, proceed to step S6.

[0133] In step S6, the device control unit 30 determines whether the multiple ultrasonic images acquired in step S1 were classified as ultrasonic images U2 captured in the second cross-section in step S2. If, in step S6, it is determined that the multiple ultrasonic images were not classified as ultrasonic images U2 captured in the second cross-section, then the multiple ultrasonic images do not correspond to either the first or second cross-section, so the process returns to step S1 to acquire the multiple ultrasonic images again. Then, the processing after step S2 is performed. If, in step S6, it is determined that the multiple ultrasonic images were classified as ultrasonic images U2 captured in the second cross-section, then the process proceeds to step S7.

[0134] In step S7, the second measurement unit 27 calculates the velocity-time integral value of the blood flow in the left ventricular outflow tract based on the ultrasound image U2 of the second cross-section captured in step S2. At this time, the second measurement unit 27 identifies the systolic phase of the heart, for example, based on any one of the following in the multi-frame ultrasound image U2: mitral valve closure, presence or absence of the aortic valve, left ventricular area, and depth of at least one set reference point. Figure 9 As shown, a Doppler gate G is set on the left ventricular outflow tract T in the ultrasound image U2 of the frame corresponding to the systolic phase. The second measurement unit 27 can automatically set the Doppler gate G on the detected left ventricular outflow tract T in the ultrasound image U2 of the frame corresponding to the systolic phase, for example. The second measurement unit 27 further acquires the Doppler waveform in the Doppler gate G and calculates the velocity-time integral value based on the Doppler waveform.

[0135] In step S8, the device control unit 30 determines whether the multiple ultrasonic images acquired in step S1 have been classified as ultrasonic images U1 representing the first cross-section in step S2. If, in step S8, it is determined that the multiple ultrasonic images have not been classified as ultrasonic images U1 representing the first cross-section, the system returns to step S1 to acquire ultrasonic images U1 representing the first cross-section again. In this case, the user adjusts the position and orientation of the ultrasonic probe 1 to acquire ultrasonic images U1 representing the first cross-section. In the subsequent step S2, the image classification unit 25 classifies the multiple ultrasonic images.

[0136] If in step S5 it is determined that multiple frames of ultrasonic images have been classified as ultrasonic images U2 of the second section, or in step S8 it is determined that multiple frames of ultrasonic images have been classified as ultrasonic images U1 of the first section, then proceed to step S9.

[0137] In step S9, the cardiac output calculation unit 28 calculates cardiac output using the cross-sectional area of ​​the left ventricular outflow tract T measured in step S4 and the velocity-time integral value of the blood flow in the left ventricular outflow tract measured in step S7. At this time, the cardiac output calculation unit 28 calculates the so-called stroke volume by multiplying the cross-sectional area of ​​the left ventricular outflow tract T by the velocity-time integral value, and calculates cardiac output by multiplying the calculated stroke volume by the heart rate.

[0138] Steps S1 through S9 are performed automatically when the user places the ultrasound probe 1 on the surface of the subject. Therefore, the user does not need to perform cumbersome operations to measure cardiac output, and can easily and quickly measure cardiac output.

[0139] If step S9 is completed in this way, then the process according to... Figure 10 The flowchart illustrates the operation of the ultrasonic diagnostic device.

[0140] Therefore, in the ultrasound diagnostic apparatus according to Embodiment 1 of the present invention, the image classification unit 25 classifies the input image as either a first cross-section or a second cross-section ultrasound image. When the input image is classified as an ultrasound image U1 that captures the first cross-section, the first measurement unit 26 measures the cross-sectional area of ​​the left ventricular outflow tract T based on the input image. When the input image is classified as an ultrasound image U2 that captures the second cross-section, the second measurement unit 27 measures the velocity-time integral value of the blood flow in the left ventricular outflow tract based on the input image. The cardiac output calculation unit 28 uses the cross-sectional area measured by the first measurement unit 26 and the velocity-time integral value measured by the second measurement unit 27 to calculate the cardiac output. Therefore, the cardiac output can be calculated easily and quickly.

[0141] Furthermore, the case where the transceiver circuit 12 is configured in the ultrasonic probe 1 has been described, but the transceiver circuit 12 can also be configured in the main body 2 of the device.

[0142] Furthermore, the case where the image generation unit 21 is arranged in the main body 2 of the device has been described, but the image generation unit 21 may also be arranged in the ultrasonic probe 1.

[0143] The main body 2 of the device can be a fixed type, a portable type that is easy to carry, or a handheld type, such as a smartphone or tablet. Thus, there is no particular limitation on the type of device that constitutes the main body 2 of the device.

[0144] The display control unit 22 can, for example, display data during a series of processes performed by the ultrasound diagnostic device, from acquiring multiple frames of ultrasound images to calculating cardiac output. Figure 11 A series of processing steps are displayed on the monitor 23. At this time, the display control unit 22 can highlight the currently executing processing step in the series. Figure 11 In the example shown, the outline of the processing step "depicting the second section" is displayed in a different way than the outlines of other processing steps. As an emphasis on the processing step, the display control unit 22 can, for example, change the display color, outline display method, and transmittance of the processing step relative to the display of other processing steps.

[0145] When the position and orientation of the ultrasound probe 1 are stable over time, the diameter of the left ventricular outflow tract can be accurately measured. Therefore, the first measuring unit 26 can, for example, perform measurements only when the position and orientation of the ultrasound probe 1 are stable. In this case, the device body 2 can be equipped with... Figure 12 The first measuring unit 26A is shown. In the first measuring unit 26A, in... Figure 6In the first measurement unit 26 shown, a frame reliability calculation unit 54 is further inserted between the mid-contraction frame detection unit 52 and the cross-sectional area calculation unit 53.

[0146] The frame reliability calculation unit 54 calculates the frame reliability based on the confidence level of the classification in the image classification unit 25, or based on the time series change of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit 51.

[0147] When classifying multiple ultrasound images, the image classification unit 25 can calculate the classification confidence, namely, the similarity between the multiple frames of ultrasound images and the ultrasound image U1 corresponding to the first section, and the similarity between the multiple frames of ultrasound images and the ultrasound image U2 corresponding to the second section. For example, the image classification unit 25 can calculate the similarity between a typical image pattern representing the ultrasound image U1 corresponding to the first section and each of the multiple frames of ultrasound images, and the similarity between a typical image pattern representing the ultrasound image U2 corresponding to the second section and each of the multiple frames of ultrasound images, as the classification confidence. Furthermore, the image classification unit 25 can calculate the so-called intermediate data calculated by the learned model in machine learning when classifying ultrasound images as the classification confidence.

[0148] The frame reliability calculation unit 54 can calculate the confidence level of the classification in the image classification unit 25 as the reliability of the frame. The first measurement unit 26 can only measure frames whose reliability reaches or exceeds a predetermined reliability threshold.

[0149] Furthermore, with the position and orientation of the ultrasound probe 1 fixed, the left ventricular outflow tract diameter (LVTD) measured by the LVTD diameter measuring unit 51 changes periodically with each heartbeat. Therefore, the frame reliability calculation unit 54 can, for example, calculate the reliability of frames whose LVTD diameter changes periodically over time as 1, and calculate the reliability of frames whose LVTD diameter does not change periodically over time as 0. In this case, the first measurement unit 26 can set a predetermined reliability threshold to 1 and only measure frames with a reliability of 1.

[0150] The display control unit 22 can display the frame reliability calculated by the frame reliability calculation unit 54 as a numerical value on the display 23. Furthermore, when intermediate data from machine learning is used as the reliability calculation, the frame reliability calculation unit 54 can calculate the reliability per pixel in the ultrasonic image. In this case, the display control unit 22 can display the reliability calculated per pixel in the ultrasonic image as a so-called thermal image overlay on the ultrasonic image. By checking the displayed reliability, the user can ensure that the inspection is proceeding appropriately and continue the inspection. Furthermore, if the reliability is determined to be low, the user can, for example, adjust the position or orientation of the ultrasonic probe 1, or manually adjust the measurement line ML1 via the input device 31.

[0151] It has been explained that the mid-systolic frame detection unit 52 of the first measurement unit 26 detects the ultrasound image U1 of the frame with the largest angle between a pair of aortic valves as the mid-systolic frame ultrasound image U1. However, sometimes the aortic valve cannot be detected due to reasons such as changes in the position or posture of the ultrasound probe 1 causing the ultrasound image U1 to become unclear. As a result, the mid-systolic frame cannot be detected either. In this case, the mid-systolic frame detection unit 52 can, for example, select the frame with the largest left ventricular outflow tract diameter among the multiple left ventricular outflow tract diameters measured by the left ventricular outflow tract diameter measurement unit 51 from multiple ultrasound images U1 as the frame used by the cross-sectional area calculation unit 53 to calculate the cross-sectional area of ​​the left ventricular outflow tract T.

[0152] Furthermore, in cases where mid-systolic frames cannot be detected, the mid-systolic frame detection unit 52 can, for example, detect frames representing the end-diastolic phase of the heart and select the frame representing the end-diastolic phase as the frame used by the cross-sectional area calculation unit 53 to calculate the cross-sectional area of ​​the left ventricular outflow tract T. The mid-systolic frame detection unit 52 can, for example, identify the left ventricular region in multiple ultrasound images U1 using methods such as template matching or learned models in machine learning, and calculate the area of ​​the identified left ventricular region, detecting the frame representing the end-diastolic phase as the phase where the area of ​​the left ventricular region is the largest in the time series. In cases where mid-systolic frames cannot be detected, the mid-systolic frame detection unit 52 can, for example, display a message on the display 23 and allow the user via the input device 31 to select which frame—the frame that measures the largest left ventricular outflow tract diameter or the frame representing the end-diastolic phase—to be used by the cross-sectional area calculation unit 53 to calculate the cross-sectional area of ​​the left ventricular outflow tract T.

[0153] Furthermore, generally, based on the anatomical structure of the heart, it is easier to depict an ultrasound image representing the long axis section of the left ventricle parasternally than an ultrasound image representing the five-chamber section at the apex of the heart. Moreover, it is more accurate to measure the left ventricular outflow tract diameter using an ultrasound image representing the long axis section of the left ventricle parasternally than using an ultrasound image representing the five-chamber section at the apex of the heart. Therefore, the device control unit 30, for example, can control various parts of the device body 2 to measure the velocity-time integral value using an ultrasound image representing the five-chamber section at the apex of the heart after measuring the left ventricular outflow tract diameter using an ultrasound image representing the long axis section of the left ventricle parasternally.

[0154] Furthermore, the display control unit 22 can, for example, display messages such as "Please take a picture of the parasternal left ventricular long axis section" urging the user to take a picture of the parasternal left ventricular long axis section on the display 23 until the image classification unit 25 classifies and displays an ultrasound image representing the parasternal left ventricular long axis section. Then, after the image classification unit 25 classifies and displays an ultrasound image representing the parasternal left ventricular long axis section, it displays messages such as "Please take a picture of the apical five-chamber section" urging the user to take a picture of the apical five-chamber section on the display 23.

[0155] Furthermore, in medical settings, when doctors and other users confirm that the ultrasound image of the left ventricle along its long axis as the first section indicates that the patient's cardiac function is normal, the depiction of the second section may sometimes be omitted. Therefore, the device control unit 30 may, for example, be able to halt a series of processes for calculating stroke volume or cardiac output based on a command from the user via the input device 31.

[0156] Implementation Method 2

[0157] The measurement values ​​obtained by the first measuring unit 26 and the second measuring unit 27 may sometimes deviate from the normal values ​​for various reasons, such as changes in the position or orientation of the ultrasonic probe 1. In such cases, the ultrasonic diagnostic device can detect the error in the measurement value.

[0158] Figure 13 The structure of the ultrasonic diagnostic apparatus of Embodiment 2 is shown. The ultrasonic diagnostic apparatus of Embodiment 2... Figure 1The ultrasound diagnostic apparatus of Embodiment 1 shown includes an apparatus body 2B instead of the apparatus body 2. In the apparatus body 2 of Embodiment 1, the apparatus body 2B further includes an error detection unit 71 and a correction unit 72, and includes an apparatus control unit 30B instead of the apparatus control unit 30. In the apparatus body 2B, the error detection unit 71 is connected to the first measuring unit 26 and the second measuring unit 27. The correction unit 72, the display control unit 22, and the apparatus control unit 30B are connected to the error detection unit 71. The correction unit 72 is connected to the cardiac output calculation unit 28 and the apparatus control unit 30B. Furthermore, the processor 33B for the apparatus body 2B is composed of the image generation unit 21, the display control unit 22, the image classification unit 25, the first measuring unit 26, the second measuring unit 27, the cardiac output calculation unit 28, the apparatus control unit 30B, the error detection unit 71, and the correction unit 72.

[0159] Error detection unit 71 detects errors related to the measurement values ​​obtained by first measurement unit 26 and second measurement unit 27 respectively.

[0160] Error detection unit 71, for example, detects errors in the left ventricular outflow tract diameter measured by left ventricular outflow tract diameter measuring unit 51 of first measuring unit 26. Ideally, the left ventricular outflow tract diameter changes periodically and regularly with heartbeats, but sometimes the measured value becomes irregular due to factors such as changes in the position or posture of the ultrasound probe 1 during measurement. Therefore, as... Figure 14 As shown, the error detection unit 71 generates a time-series variation curve C1 of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit 51, and by comparing the generated time-series variation curve C1 with, for example, a predetermined curve C2 (a so-called B-spline curve), errors concerning the left ventricular outflow tract diameter can be detected. Figure 14 In the case where there are inconsistencies between the time series curve C1 and the predetermined curve C2, the error detection unit 71 can determine that there is an error in the diameter of the left ventricular outflow tract, just like the peak portion of the predetermined curve C2.

[0161] Furthermore, the ideal value for the left ventricular outflow tract diameter is generally known to be 2 cm. Therefore, the error detection unit 71 can, for example, detect errors in the left ventricular outflow tract diameter by comparing the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit 51 with a predetermined set value. As a specific example, the error detection unit 71 compares the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit 51 with a predetermined measurement range including 2 cm. If the diameter is outside the predetermined measurement range, it can be determined that there is an error in the left ventricular outflow tract diameter.

[0162] The error detection unit 71 can also detect errors in the velocity-time integral value of the left ventricular outflow tract blood flow calculated by the velocity-time integral value calculation unit 64 of the second measurement unit 27. Ideally, the Doppler waveform changes in a predetermined time series, but sometimes the measured value does not change as ideally as due to factors such as changes in the position or posture of the ultrasound probe 1 during measurement. In such cases, a normal value cannot always be obtained as the velocity-time integral value. Therefore, the error detection unit 71 can detect errors in the velocity-time integral value by comparing the Doppler waveform acquired by the Doppler waveform acquisition unit 63 of the second measurement unit 27 with a predetermined curve representing the ideal Doppler waveform of the left ventricular outflow tract blood flow. If there are inconsistencies between the Doppler waveform and the predetermined curve, the error detection unit 71 can determine that there is an error in the velocity-time integral value.

[0163] Furthermore, it is generally known that the ideal value for the velocity-time integral of blood flow in the left ventricular outflow tract is 18–22 cm in adults. Therefore, the error detection unit 71 can, for example, detect errors in the velocity-time integral value by comparing the velocity-time integral value calculated by the velocity-time integral value calculation unit 64 with a predetermined set value. As a specific example, the error detection unit 71 compares the velocity-time integral value calculated by the velocity-time integral value calculation unit 64 with a predetermined measurement range of 18–22 cm, and if the value is outside the predetermined measurement range, it can determine that there is an error in the velocity-time integral value.

[0164] The correction unit 72 corrects the measurement values ​​of the first measurement unit 26 and the second measurement unit 27, which were detected as erroneous by the error detection unit 71. The correction unit 72 can correct the left ventricular outflow tract diameter detected as erroneous by the error detection unit 71 based on the cross-sectional area of ​​the left ventricular outflow tract T calculated by the cross-sectional area calculation unit 53 in the past. For example, the correction unit 72 inputs the ultrasound image U1 into a first learned model in machine learning that has pre-learned landmarks such as the aortic valve or mitral valve in the ultrasound image U1 taken at the first cross-section, outputs the confidence score of each pixel of the landmark as a heatmap, and then inputs the heatmap of the landmark and the cross-sectional area of ​​the left ventricular outflow tract T calculated by the first measurement unit 26 into a second learned model in machine learning that has pre-learned the relationship between the heatmap of the landmark and the cross-sectional area of ​​the left ventricular outflow tract T, thereby outputting the corrected cross-sectional area of ​​the left ventricular outflow tract T.

[0165] Furthermore, the correction unit 72 can also correct the erroneous velocity-time integral value detected by the error detection unit 71 based on the velocity-time integral value of the left ventricular outflow tract blood flow calculated by the velocity-time integral value calculation unit 64 in the past. For example, the correction unit 72 inputs the ultrasound image U2 into the first learned model in machine learning that has pre-learned the landmarks such as the aortic valve or mitral valve in the ultrasound image U2 captured in the second section, outputs the confidence level of each pixel of the landmark as a heatmap, and then inputs the heatmap of the landmark, the velocity-time integral value of the left ventricular outflow tract blood flow obtained by the second measurement unit 27, and the Doppler waveform into the second learned model in machine learning that has pre-learned the relationship between the heatmap of the landmark and the velocity-time integral value of the left ventricular outflow tract blood flow and the Doppler waveform, thereby outputting the corrected velocity-time integral value of the left ventricular outflow tract blood flow.

[0166] As described above, in the ultrasonic diagnostic apparatus according to Embodiment 2, the calibration unit 72 automatically corrects errors in the measured values, so the user can easily obtain appropriate calibration values.

[0167] Furthermore, in Embodiment 2, the case where the erroneous measurement value detected by the error detection unit 71 is automatically corrected by the correction unit 72 has been described. However, it is also possible, for example, to display the error on the display 23 without correcting the error. In this case, by confirming the error, the user can take measures such as adjusting the position and orientation of the ultrasonic probe 1 to avoid erroneous measurement values.

Claims

1. An ultrasound diagnostic device that measures cardiac output based on multiple frames of ultrasound images of different first and second cross-sections of the heart of a subject, wherein the ultrasound diagnostic device comprises: The image classification unit classifies the input image to determine which of the first and second cross sections the ultrasound image captured. The first measurement unit measures the cross-sectional area of ​​the left ventricular outflow tract based on the input image when the image classification unit classifies the input image as an ultrasound image of the first cross section. The second measurement unit, when the image classification unit classifies the input image as an ultrasound image of the second cross-section, measures the velocity-time integral value of the blood flow in the left ventricular outflow tract based on the input image; and The cardiac output calculation unit calculates the cardiac output using the cross-sectional area measured by the first measurement unit and the velocity-time integral value measured by the second measurement unit.

2. The ultrasonic diagnostic device according to claim 1, wherein, The image classification unit classifies the input image by using the parasternal left ventricular long axis section as the first section and the apical five-chamber section or the apical three-chamber section as the second section.

3. The ultrasonic diagnostic device according to claim 2, wherein, The first measuring unit includes: The left ventricular outflow tract diameter measuring unit measures the left ventricular outflow tract diameter based on the input image. The mid-shrinkage frame detection unit detects mid-shrinkage frames based on the input image; and The cross-sectional area calculation unit calculates the cross-sectional area of ​​the left ventricular outflow tract based on the inner diameter of the left ventricular outflow tract measured by the inner diameter measurement unit in the mid-systolic frame detected by the mid-systolic frame detection unit.

4. The ultrasonic diagnostic device according to claim 3, wherein, The left ventricular outflow tract diameter measuring unit measures the left ventricular outflow tract diameter at a predetermined distance from the aortic valve insertion point toward the heart.

5. The ultrasonic diagnostic device according to claim 3 or 4, wherein, The mid-systolic frame detection unit detects the frame with the largest angle between a pair of aortic valves as the mid-systolic frame.

6. The ultrasonic diagnostic device according to claim 5, wherein, The first measurement unit includes a frame reliability calculation unit for calculating the reliability of frames. The left ventricular outflow tract diameter measuring unit measures the left ventricular outflow tract diameter based on frames whose reliability, calculated by the frame reliability calculation unit, is above a predetermined reliability threshold.

7. The ultrasonic diagnostic device according to claim 6, wherein, The frame reliability calculation unit calculates the reliability based on the confidence level of the classification in the image classification unit, or based on the time series change of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit.

8. The ultrasonic diagnostic device according to claim 2, wherein, The second measuring unit includes: The contraction period recognition unit identifies the contraction period based on the input image; A gate setting unit sets a Doppler gate in the input image that is identified as the contraction period by the contraction period recognition unit; The Doppler waveform acquisition unit acquires the Doppler waveform in the Doppler gate set by the gate setting unit; and The velocity-time integral value calculation unit calculates the velocity-time integral value based on the Doppler waveform acquired by the Doppler waveform acquisition unit.

9. The ultrasonic diagnostic apparatus according to claim 8, wherein, The systolic period recognition unit identifies the systolic period based on any one of the following: mitral valve closure, presence or absence of aortic valve, left ventricular area, and depth of a reference point set in the input image.

10. The ultrasonic diagnostic apparatus according to claim 8 or 9, wherein, The first measuring unit includes a left ventricular outflow tract diameter measuring unit that measures the left ventricular outflow tract diameter based on the input image. The gate setting unit sets the Doppler gate in the input image based on the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit in the first measuring unit.

11. The ultrasonic diagnostic device according to claim 3, wherein, The cardiac output calculation unit calculates the heart rate based on the number of plotting points between two consecutive mid-systolic frames detected by the mid-systolic frame detection unit, and the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measurement unit. The cardiac output is calculated based on the cross-sectional area measured by the first measuring unit, the velocity-time integral value measured by the second measuring unit, and the calculated heart rate.

12. The ultrasonic diagnostic apparatus according to claim 3, comprising: The error detection unit detects errors in the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit.

13. The ultrasonic diagnostic apparatus according to claim 12, wherein, The error detection unit detects errors in the left ventricular outflow tract diameter by comparing the time-series variation curve of the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit with a predetermined curve, or by comparing the left ventricular outflow tract diameter measured by the left ventricular outflow tract diameter measuring unit with a predetermined set value.

14. The ultrasonic diagnostic apparatus according to claim 13, comprising: The correction unit corrects the inner diameter of the left ventricular outflow tract that was detected as erroneous by the error detection unit, based on the cross-sectional area of ​​the left ventricular outflow tract that was previously calculated by the cross-sectional area calculation unit.

15. The ultrasonic diagnostic apparatus according to claim 8, comprising: The error detection unit detects errors in the speed-time integral value calculated by the speed-time integral value calculation unit.

16. The ultrasonic diagnostic apparatus according to claim 15, wherein, The error detection unit detects errors in the velocity-time integral value by comparing a Doppler curve obtained by tracking the Doppler waveform acquired by the Doppler waveform acquisition unit with a predetermined curve, or by comparing the velocity-time integral value calculated by the velocity-time integral value calculation unit with a predetermined set value.

17. The ultrasonic diagnostic apparatus according to claim 16, comprising: The correction unit corrects the speed-time integral value that was detected as erroneous by the error detection unit based on the speed-time integral value calculated by the speed-time integral value calculation unit in the past.

18. A control method for an ultrasound diagnostic device, comprising measuring cardiac output based on multiple frames of ultrasound images of a subject's heart, each captured in a different first and second cross-section, the control method comprising: The input image is classified based on whether it is an ultrasound image of the first section or the second section that was captured. The cross-sectional area of ​​the left ventricular outflow tract is measured based on the input image classified as having captured an ultrasound image of the first section. The velocity-time integral value of blood flow in the left ventricular outflow tract is measured based on the input image classified as having captured an ultrasound image of the second section. and The cardiac output is calculated using the measured cross-sectional area and the measured velocity-time integral value.

Images