Ultrasound diagnostic device and control method for ultrasound diagnostic device

Abstract

The present invention provides an ultrasound diagnostic device and a control method for the ultrasound diagnostic device that can quickly and easily correct the segmentation results of the cardiac lumen. [Solution] The ultrasound diagnostic device includes an image acquisition unit (33) that acquires multiple frames of ultrasound images of the subject's heart, a contour extraction unit (24) that performs segmentation on the multiple frames of ultrasound images to extract the contour of the lumen of the heart for each frame, a cross-sectional area calculation unit (25) that calculates the cross-sectional area of ​​the lumen of the heart for each frame based on the extracted contour, a cross-sectional area change curve generation unit (26) that generates a cross-sectional area change curve representing the actual temporal change of the cross-sectional area based on the calculated cross-sectional area, and an error detection unit (28) that detects errors in the segmentation performed by comparing the cross-sectional area change curve with a predicted cross-sectional area change prediction curve.

Description

Technical Field

[0001] The present invention relates to an ultrasonic diagnostic apparatus for capturing ultrasonic images of a subject's heart and a control method for the ultrasonic diagnostic apparatus.

Background Art

[0002] Conventionally, an ultrasonic image representing a tomographic plane of a subject's heart has been captured using a so-called ultrasonic diagnostic apparatus, and using this ultrasonic image, the ejection fraction of the heart cavity such as a so-called left ventricular ejection fraction (LVEF: Left Ventricular Ejection Fraction) has been calculated. When calculating the ejection fraction of the heart cavity, ultrasonic images representing the states of the heart at the end of systole and the end of diastole in a so-called apical two-chamber cross-section and ultrasonic images representing the states of the heart at the end of systole and the end of diastole in a so-called apical four-chamber cross-section are often used.

[0003] [[ID=I5]] Also, when calculating the ejection fraction of the heart cavity, for example, by performing so-called segmentation on each ultrasonic image, the contour of the heart cavity is automatically extracted, and the ejection fraction is often calculated using the area surrounded by the contour. In such a series of processes, a contour different from the original contour of the heart cavity may be extracted as the contour of the heart cavity automatically extracted due to the presence of so-called artifacts reflected in the ultrasonic image. In order to easily correct such segmentation errors, for example An ultrasonic diagnostic apparatus as disclosed in Patent Document 1 has been developed. The ultrasonic diagnostic apparatus of Patent Document 1 has a plurality of correction modes corresponding to a plurality of correction methods for the extracted contour of the structure in the heart, and automatically corrects the contour using the correction mode selected by the user.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

[0005] However, in the technology described in Patent Document 1, the user of the ultrasound diagnostic device needs to examine multiple frames of ultrasound images in order to find ultrasound images in which segmentation errors have occurred, which can result in a considerable amount of time being required to correct the segmentation results.

[0006] This invention was made to solve the aforementioned problems of the past, and aims to provide an ultrasound diagnostic device and a control method for the ultrasound diagnostic device that can quickly and easily correct the segmentation results of the cardiac lumen. [Means for solving the problem]

[0007] The above objective can be achieved with the following configuration. [1] An image acquisition unit that acquires multiple frames of ultrasound images that are sequentially and continuously captured in time and of the subject's heart, A contour extraction unit that performs segmentation on multiple frames of ultrasound images to extract the contour of the cardiac lumen for each frame, A cross-sectional area calculation unit calculates the cross-sectional area of ​​the cardiac lumen for each frame based on the contour extracted by the contour extraction unit, A cross-sectional area change curve generation unit generates a cross-sectional area change curve that represents the actual change in cross-sectional area over time, based on the cross-sectional area calculated by the cross-sectional area calculation unit. An error detection unit detects segmentation errors performed in the contour extraction unit by comparing the cross-sectional area change curve generated by the cross-sectional area change curve generation unit with a predicted cross-sectional area change prediction curve. An ultrasound diagnostic device equipped with the following features. [2] The ultrasound diagnostic apparatus described in [1], wherein the error detection unit notifies the user of the error detection result. [3] The ultrasound diagnostic apparatus according to [1] or [2], wherein the error detection unit detects as an error that the cross-sectional area change curve and the cross-sectional area change prediction curve are at least partially separated from each other. [4] An ultrasound diagnostic apparatus according to any one of [1] to [3], comprising a prediction curve generation unit that generates a cross-sectional area change prediction curve based on a cross-sectional area change curve generated by a cross-sectional area change curve generation unit. [5] An ultrasound diagnostic apparatus according to any one of [1] to [4], comprising an error correction unit for correcting segmentation errors performed in the contour extraction unit. [6] The error correction section is: Multiple control points that can be manipulated by the user are set for the contour extracted by the contour extraction unit. The ultrasound diagnostic apparatus according to [5], which corrects segmentation errors by deforming the contour based on the movement of at least one of multiple control points by the user. [7] The error correction section is: In the image area specified by the user, the number of inflection points in the contours extracted by the contour extraction unit is calculated. In the image region, the edges of the cardiac lumen are detected and the number of inflection points at the edges is calculated. The ultrasonic diagnostic apparatus described in [6], which sets the number of control points by comparing the number of inflection points in the contour with the number of inflection points in the edge. [8] Acquire multiple ultrasound images of the subject's heart that are continuous in time series. Segmentation is performed on multiple frames of ultrasound images to extract the contour of the cardiac lumen for each frame. Based on the extracted contours, the cross-sectional area of ​​the cardiac cavity is calculated for each frame. Based on the calculated cross-sectional area, a cross-sectional area change curve is generated that represents the actual change in the cross-sectional area over time. Segmentation errors are detected by comparing the generated cross-sectional area change curve with the predicted cross-sectional area change curve. A method for controlling an ultrasound diagnostic device. [Effects of the Invention]

[0008] The present invention provides an ultrasound diagnostic device comprising: an image acquisition unit that acquires multiple frames of ultrasound images of a subject's heart in a time-series sequence; a contour extraction unit that performs segmentation on the multiple frames of ultrasound images to extract the contour of the lumen of the heart for each frame; a cross-sectional area calculation unit that calculates the cross-sectional area of ​​the lumen of the heart for each frame based on the contour extracted by the contour extraction unit; a cross-sectional area change curve generation unit that generates a cross-sectional area change curve representing the actual temporal change of the cross-sectional area based on the cross-sectional area calculated by the cross-sectional area calculation unit; and an error detection unit that detects errors in the segmentation performed by the contour extraction unit by comparing the cross-sectional area change curve generated by the cross-sectional area change curve generation unit with a predicted cross-sectional area change prediction curve. Therefore, the segmentation results of the lumen of the heart can be corrected quickly and easily. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing the configuration of an ultrasound diagnostic device according to an embodiment of the present invention. [Figure 2] This is a block diagram showing the internal configuration of the transmitting and receiving circuit in an embodiment of the present invention. [Figure 3] This is a block diagram showing the internal configuration of the image generation unit in an embodiment of the present invention. [Figure 4] This diagram schematically shows an example of an ultrasound image representing a four-chamber cross-section of the cardiac apex. [Figure 5] This figure shows an example of a curve showing the change in the cross-sectional area of ​​the cardiac lumen in ultrasound images across multiple frames. [Figure 6] This figure shows examples of cross-sectional area change curves where segmentation errors occurred in the contour of the cardiac lumen in several ultrasound images across multiple frames. [Figure 7] This figure shows an example of a predicted cross-sectional area change curve calculated based on the cross-sectional area change curve. [Figure 8] This figure shows an example of multiple control points when segmentation extracts a contour containing more irregularities than the contour of the heart lumen detected by edge detection. [Figure 9]It is a diagram showing examples of a plurality of control points when a contour smoother than the contour of the inner cavity of the heart detected by edge detection processing is extracted by segmentation. [Figure 10] It is a flowchart showing the operation of an ultrasonic diagnostic apparatus according to an embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments of this invention will be described based on the accompanying drawings. The description of the constituent elements described below is made based on typical embodiments of the present invention, but the present invention is not limited to such embodiments. In this specification, a numerical range represented by "~" means a range including the numerical values described before and after "~" as the lower limit value and the upper limit value. In this specification, "identical" and "the same" include an error range generally acceptable in the technical field.

[0011] Embodiment FIG. 1 shows the configuration of an ultrasonic diagnostic apparatus according to an embodiment of the present invention. The ultrasonic diagnostic apparatus includes an ultrasonic probe 1 and a device main body 2 connected to each other by so-called wired communication or so-called wireless communication.

[0012] The ultrasonic probe 1 includes a vibrator array 11 and a transmission / reception circuit 12 connected thereto.

[0013] The main unit 2 of the device includes an image generation unit 21 connected to the transmitting / receiving circuit 12. In the main unit 2, the display control unit 22 and the monitor 23 are sequentially connected to the image generation unit 21. Furthermore, the contour extraction unit 24, the cross-sectional area calculation unit 25 and the cross-sectional area change curve generation unit 26 are sequentially connected to the image generation unit 21. The cross-sectional area change prediction curve generation unit 27 is connected to the cross-sectional area change curve generation unit 26. The error detection unit 28 is connected to both the cross-sectional area change curve generation unit 26 and the cross-sectional area change prediction curve generation unit 27. Furthermore, a memory 29 is connected to the contour extraction unit 24, and an error correction unit 30 is connected to the memory 29. The error correction unit 30 is connected to the memory 29. The memory 29 is further connected to the cross-sectional area calculation unit 25. In addition, the cross-sectional area change curve generation unit 26, the cross-sectional area change prediction curve generation unit 27, the error detection unit 28 and the error correction unit 30 are connected to the display control unit 22. Furthermore, the main control unit 31 is connected to the transmitting / receiving circuit 12, image generation unit 21, display control unit 22, contour extraction unit 24, cross-sectional area calculation unit 25, cross-sectional area change curve generation unit 26, cross-sectional area change prediction curve generation unit 27, error detection unit 28, memory 29, and error correction unit 30. An input device 32 is connected to the main control unit 31.

[0014] The image acquisition unit 33 is composed of a transmitting / receiving circuit 12 and an image generation unit 21. Furthermore, the processor 34 for the main unit 2 is composed of the image generation unit 21, display control unit 22, contour extraction unit 24, cross-sectional area calculation unit 25, cross-sectional area change curve generation unit 26, cross-sectional area change prediction curve generation unit 27, error detection unit 28, error correction unit 30, and main unit control unit 31.

[0015] The transducer array 11 of the ultrasonic probe 1 has a plurality of ultrasonic transducers arranged in one or two dimensions. Each of these ultrasonic transducers transmits ultrasound according to a drive signal supplied from the transmitting / receiving circuit 12, and also receives ultrasonic echoes from the subject and outputs a signal based on the ultrasonic echoes. Each ultrasonic transducer is constructed by forming electrodes at both ends of a piezoelectric body made of, for example, a piezoelectric ceramic represented by PZT (Lead Zirconate Titanate), a polymer piezoelectric element represented by PVDF (Poly Vinylidene Di Fluoride), or a piezoelectric single crystal represented by PMN-PT (Lead Magnesium Niobate-Lead Titanate).

[0016] The image acquisition unit 33, which consists of a transmitting / receiving circuit 12 and an image generation unit 21, acquires an ultrasound image of the subject by transmitting and receiving an ultrasound beam using an ultrasound probe 1.

[0017] The transmitting / receiving circuit 12 transmits ultrasonic waves from the transducer array 11 and generates a sound line signal based on the received signal acquired by the transducer array 11, under the control of the main unit control 31. As shown in Figure 2, the transmitting / receiving circuit 12 has a pulser 41 connected to the transducer array 11, and an amplifier 42, an AD (Analog to Digital) converter 43, and a beamformer 44 connected sequentially in series from the transducer array 11.

[0018] The pulser 41 includes, for example, multiple pulse generators, and based on a transmission delay pattern selected according to a control signal from the main unit control 31, it supplies each drive signal to the multiple ultrasonic transducers of the transducer array 11, adjusting the delay amount, so that the ultrasonic waves transmitted from the transducers form an ultrasonic beam. In this way, when a pulsed or continuous wave voltage is applied to the electrodes of the ultrasonic transducers of the transducer array 11, the piezoelectric material expands and contracts, generating pulsed or continuous wave ultrasonic waves from each ultrasonic transducer, and an ultrasonic beam is formed from the combined wave of these ultrasonic waves.

[0019] The transmitted ultrasonic beam is reflected from a target, such as a part of the subject, and propagates toward the transducer array 11 of the ultrasonic probe 1. The ultrasonic echo propagating toward the transducer array 11 is received by each ultrasonic transducer that makes up the transducer array 11. At this time, each ultrasonic transducer that makes up the transducer array 11 expands and contracts upon receiving the propagating ultrasonic echo, generating a received signal which is an electrical signal, and outputs these received signals to the amplification unit 42.

[0020] The amplification unit 42 amplifies the signals input from each ultrasonic transducer constituting the transducer array 11 and transmits the amplified signals to the AD conversion unit 43. The AD conversion unit 43 converts the signals transmitted from the amplification unit 42 into digital received data. The beamformer 44 performs so-called receive focus processing by adding each received data received from the AD conversion unit 43 with a corresponding delay. Through this receive focus processing, each received data converted by the AD conversion unit 43 is phase-corrected and added together, and a sound ray signal with a focused ultrasonic echo is obtained.

[0021] As shown in Figure 3, the image generation unit 21 has a configuration in which a signal processing unit 45, a DSC (Digital Scan Converter) 46, and an image processing unit 47 are connected in series in sequence.

[0022] The signal processing unit 45 receives the sound line signal from the transmitting / receiving circuit 12, and after correcting for attenuation due to distance according to the depth of the ultrasonic reflection position using the sound velocity value set by the main unit control unit 31, it performs envelope detection processing to generate a B-mode image signal, which is tomographic image information of the tissue within the subject.

[0023] The DSC46 converts the B-mode image signal generated by the signal processing unit 45 into an image signal that follows the scanning method of a normal television signal (raster conversion). The image processing unit 47 performs various necessary image processing, such as gradation processing, on the B-mode image signal input from the DSC 46, and then sends the B-mode image signal to the display control unit 22 and the contour extraction unit 24. Hereafter, the B-mode image signal processed by the image processing unit 47 will be referred to as the ultrasonic image.

[0024] The ultrasound diagnostic apparatus according to the embodiment of the present invention is used to calculate the ejection fraction of the cardiac lumen of a subject, such as the so-called left ventricular ejection fraction (LVEF). To calculate the ejection fraction of the cardiac lumen, the image acquisition unit 33 acquires multiple frames of ultrasound images U representing the so-called apical four-chamber section 4C of the heart H, as shown in Figure 4, and multiple frames of ultrasound images U representing the so-called apical two-chamber section of the heart H, although these are not shown.

[0025] The display control unit 22, under the control of the main unit control unit 31, performs predetermined processing on the first ultrasound image U1 and the second ultrasound image U2, etc., generated by the image generation unit 21 and displays them on the monitor 23.

[0026] The monitor 23 displays the first ultrasound image U1 and the second ultrasound image U2, etc., under the control of the display control unit 14, and has a display device such as an LCD (Liquid Crystal Display) or an organic EL display (Organic Electroluminescence Display).

[0027] The main unit control unit 31 controls each part of the main unit 2 and the transmitting / receiving circuit 12 of the ultrasonic probe 1 based on a control program or the like that is stored in advance.

[0028] The input device 32 is for the user to perform input operations and consists of devices such as a keyboard, mouse, trackball, touchpad, and touch sensor placed on top of the monitor 23.

[0029] The contour extraction unit 24 performs segmentation on multiple frames of ultrasound images U acquired by the image acquisition unit 33, and extracts the contour B of the lumen A of the heart H for each frame, for example, as shown in Figure 4. In this figure, the left ventricle is shown as an example of the lumen A of the heart H.

[0030] The contour extraction unit 24 can extract the contour B of the lumen A of heart H from the ultrasound image U by segmentation using a learning model in so-called machine learning, which has learned the relationship between a large number of ultrasound images including the lumen A of heart H and the contour B of the lumen A. Alternatively, the contour extraction unit 24 can also extract the contour B of the lumen A of heart H from the ultrasound image U using a so-called template matching method, which involves pre-storing multiple template image data representing the lumen A of heart H and searching the ultrasound image U using these multiple template image data.

[0031] The contour extraction unit 24 sends the information of the contour B of the lumen A of the heart H, which has been extracted in this manner, to the cross-sectional area calculation unit 25. The contour extraction unit 24 also stores the extracted information of the contour B of the lumen A of the heart H, together with the ultrasound image U from which the contour B was extracted, in the memory 29.

[0032] The cross-sectional area calculation unit 25 calculates the cross-sectional area of ​​the lumen A of the heart H for each frame based on the contour B extracted by the contour extraction unit 24. The cross-sectional area calculation unit 25 states that the cross-sectional area of ​​the lumen A of the heart H calculated here refers to the area of ​​the lumen A of the heart H on the ultrasound image U. The cross-sectional area calculation unit 25 can calculate the cross-sectional area of ​​the lumen A of the heart H by known methods, such as calculating the total number of pixels within the extracted contour B.

[0033] The cross-sectional area change curve generation unit 26 generates a cross-sectional area change curve that represents the actual temporal change of the cross-sectional area of ​​the lumen A of the heart H, based on the cross-sectional area of ​​the lumen A of the heart H calculated by the cross-sectional area calculation unit 25. For example, as schematically shown by the cross-sectional area change curve C1 in Figure 5, the cross-sectional area of ​​the lumen A of the heart H changes periodically over time due to the beating of the heart H.

[0034] If segmentation is performed correctly by the contour extraction unit 24 for all acquired frames, that is, if the original contour B of the lumen A of the heart H is extracted correctly, then a cross-sectional area change curve C1 that correctly represents the changes according to the beating of the heart H is obtained, for example, as schematically shown in Figure 5. On the other hand, if an error occurs in the segmentation by the contour extraction unit 24 due to the presence of so-called artifacts in the ultrasound image U, and a contour B different from the original contour B of the lumen A of the heart H is extracted, then a partially distorted cross-sectional area change curve C2 may be obtained, for example, as schematically shown in Figure 6. In the example in Figure 6, it is shown that the portion of the cross-sectional area change curve C2 corresponding to the maximum value in the cross-sectional area change curve C1 in Figure 5 is distorted.

[0035] The cross-sectional area change prediction curve generation unit 27 generates a cross-sectional area change prediction curve D1, for example, as shown in Figure 7, based on the cross-sectional area change curve C1 or C2 generated by the cross-sectional area change curve generation unit 26. The cross-sectional area change prediction curve D1 is a curve that represents the time-series change in the cross-sectional area of ​​the lumen A of the heart H, predicted assuming that segmentation is performed normally in all of the ultrasound images U of multiple frames. The cross-sectional area change prediction curve generation unit 27 can generate a curve as the cross-sectional area change prediction curve D1 by approximating the cross-sectional area change curve C1 or C2 with a so-called Bézier curve using a so-called least squares method or the like. In the example in Figure 7, a cross-sectional area change prediction curve D1 is shown that approximates a partially distorted cross-sectional area change curve C2 with a Bézier curve. In this example, the cross-sectional area change curve C2 and the cross-sectional area change prediction curve D1 do not intersect each other in the portion containing the maximum value of the cross-sectional area change prediction curve D1, and are separated from each other.

[0036] The error detection unit 28 detects segmentation errors performed by the contour extraction unit 24 by comparing the cross-sectional area change curve C1 or C2 generated by the cross-sectional area change curve generation unit 26 with the predicted cross-sectional area change prediction curve D1. For example, the error detection unit 28 can detect a segmentation error if the cross-sectional area change curve C1 or C2 and the cross-sectional area change prediction curve D1 are at least partially separated from each other. For example, in the example in Figure 7, the cross-sectional area change curve C2 and the cross-sectional area change prediction curve D1 are separated from each other in the portion containing the maximum value of the cross-sectional area change prediction curve D1, and this is detected as a segmentation error.

[0037] Thus, since the error detection unit 28 automatically detects segmentation errors by comparing the cross-sectional area change curve C1 or C2 with the cross-sectional area change prediction curve D1, the user of the ultrasound diagnostic device does not need to check multiple frames of ultrasound images U to find segmentation errors, and segmentation errors can be found quickly and easily.

[0038] The error detection unit 28 can notify the user of the error detection result by displaying a message such as "An error has occurred in segmentation" on the monitor 23 along with the cross-sectional area change curve C2 and the cross-sectional area change prediction curve D1 as shown in Figure 7. The user can then confirm the notification from the error detection unit 28 and check only the necessary frames of ultrasound images U to correct the segmentation error.

[0039] The memory 29 stores, under the control of the main unit control 31, the contour B of the lumen A of the heart H extracted by the contour extraction unit 24 and the information of the contour deformed by the error correction unit 30, which will be described later.

[0040] For memory 29, for example, recording media such as flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), FD (Flexible Disk), MO disk (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) can be used.

[0041] The error correction unit 30 corrects the segmentation errors performed by the contour extraction unit 24 by deforming the contour B of the lumen A of the heart H extracted by the contour extraction unit 24 based on user input operations via the input device 32. The error correction unit 30 can deform contour B in an image region R on the ultrasound image U specified by the user via the input device 32, for example, as shown in Figure 8, by placing a plurality of user-movable control points P on the ultrasound image U, and can deform contour B in response to the user's movement of the plurality of control points P. In this case, the error correction unit 30 can, for example, approximate contour B in the image region R with a Bézier curve, and then place a plurality of control points of the approximated Bézier curve as a plurality of control points P.

[0042] In this process, the error correction unit 30 can, for example as shown in Figure 8, calculate the number of inflection points in the contour B extracted by the contour extraction unit 24 in the image region R specified by the user, detect the edge F of the lumen A of the heart H in the image region R and calculate the number of inflection points at that edge F, and set the number of control points P by comparing the number of inflection points in contour B with the number of inflection points at edge F. Here, the error correction unit 30 can, for example, create multiple brightness profiles along the scan line in the ultrasound image U, and in each brightness profile, detect locations where the brightness value is above a certain brightness threshold as edges F.

[0043] The error correction unit 30 can set a predetermined first number of control points P if, for example, the number of inflection points at the detected edge F is less than the number of inflection points at contour B, as shown in Figure 8. Furthermore, the error correction unit 30 can set a predetermined second number of control points P that is greater than the predetermined first number if, for example, the number of inflection points at the detected edge F is greater than the number of inflection points at contour B, as shown in Figure 9.

[0044] The error correction unit 30 can also, for example, set the number of control points P according to the difference between the number of inflection points on the detected edge F and the number of inflection points on the contour B.

[0045] The information of contour B after correction by the error correction unit 30 is stored in memory 29. This corrected information of contour B is read from memory 29 and sent to the cross-sectional area calculation unit 25. The cross-sectional area calculation unit 25 recalculates the cross-sectional area of ​​the lumen A of heart H based on the corrected contour B. The cross-sectional area change curve generation unit 26 updates the cross-sectional area change curve C2 by using the recalculated value of the cross-sectional area of ​​the lumen A of heart H in place of the cross-sectional area of ​​the lumen A of heart H before recalculation. The cross-sectional area change prediction curve generation unit 27 regenerates the cross-sectional area change prediction curve D1 based on the updated cross-sectional area change curve C2. The error detection unit 28 performs a process to detect segmentation errors by comparing the updated cross-sectional area change curve C2 and the regenerated cross-sectional area change prediction curve D1.

[0046] The user deforms the contour B of the lumen A of the heart H via the input device 32, referring to the updated cross-sectional area change curve C2, the regenerated cross-sectional area change prediction curve D1, and the detection results of segmentation errors based on these, for example, until it is determined that the segmentation errors have been appropriately corrected.

[0047] The processor 34, which includes an image generation unit 21, a display control unit 22, a contour extraction unit 24, a cross-sectional area calculation unit 25, a cross-sectional area change curve generation unit 26, a cross-sectional area change prediction curve generation unit 27, an error detection unit 28, an error correction unit 30, and a main unit control unit 31, may be composed of one or more hardware components, and the type of hardware is not limited. For example, the processor can be composed of hardware such as a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a programmable logic device such as an FPGA (Field Programmable Gate Array), a dedicated circuit for executing specific processing such as an ASIC (Application Specific Integrated Circuit), a GPU (Graphic Processing Unit), or an NPU (Neural Processing Unit). The processor also has various units or means that execute the various processes in this embodiment. Furthermore, the type of hardware may 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. Furthermore, in any embodiment, the order of each process performed by the processor is not limited to the order described above and may be changed as appropriate. The hardware is composed of electrical circuits and the like, which are made up of circuit elements such as semiconductor elements.

[0048] Furthermore, this embodiment may be implemented by hardware, software, firmware, microcode, or a combination thereof. The software, firmware, and microcode are composed of a program. The program may also be, for example, a group of program modules, each of which may be implemented by a processor configured to perform its respective function. The program may be program code or multiple code segments stored on one or more non-temporary computer-readable media (e.g., storage media or other storage). The program may be divided and stored on multiple non-temporary computer-readable media located on devices that are physically separated from each other. The program code or code segment may represent any combination of procedures, functions, subprograms, routines, subroutines, modules, software packages, classes, or instructions, data structures, or program statements. The program code or code segment may be connected to other code segments or hardware circuits by sending and receiving information, data, arguments, parameters, or memory contents.

[0049] Next, the operation of the ultrasound diagnostic apparatus according to the embodiment will be described with reference to the flowchart shown in Figure 10.

[0050] In step S1, the image acquisition unit 33 generates an ultrasound image U of the lumen A of the heart H. At this time, under the control of the main unit control unit 31, the transmission and reception of ultrasound is started from multiple transducers of the transducer array 11 according to the drive signal from the pulser 41 of the transmitting and receiving circuit 12 of the ultrasound probe 1. The ultrasound echo from within the subject is received by multiple transducers of the transducer array 11, the received signal which is an analog signal is output to the amplification unit 42 for amplification, and the received data is acquired by AD conversion in the AD conversion unit 43.

[0051] The beamformer 44 performs reception focus processing on this received data, and the resulting sound ray signal is sent to the image generation unit 21 of the main body 2 of the device, where the image generation unit 21 generates an ultrasound image U representing the lumen A of the subject's heart H. At this time, the signal processing unit 45 of the image generation unit 21 performs attenuation correction according to the depth of the ultrasound reflection position and envelope detection processing on the sound ray signal, and the DSC 46 converts it into an image signal following the scanning method of a normal television signal, and the image processing unit 47 performs various necessary image processing such as gradation processing. The first ultrasound image U1 generated in step S1 in this way is displayed on the monitor 23 via the display control unit 22 and sent to the contour extraction unit 24.

[0052] In step S2, the contour extraction unit 24 extracts the contour B of the lumen A of the heart H from the ultrasound image U acquired in step S1. The contour extraction unit 24 can extract the contour B of the lumen A of the heart H by, for example, using a trained machine learning model that has learned the relationship between a large number of ultrasound images U and the contour B of the lumen A of the heart H shown therein.

[0053] The contour extraction unit 24 stores the information of the extracted contour B together with the ultrasound image U from which contour B was extracted in the memory 29, linking them to each other.

[0054] In step S3, the cross-sectional area calculation unit 25 calculates the cross-sectional area of ​​the lumen A of the heart H based on the contour B of the lumen A of the heart H extracted in step S2. The cross-sectional area calculation unit 25 can calculate the cross-sectional area of ​​the lumen A of the heart H by, for example, calculating the total number of pixels in the region enclosed by contour B in the ultrasound image U.

[0055] In step S4, the cross-sectional area change curve generation unit 26 generates a cross-sectional area change curve C1 or C2 as shown in Figures 5 and 6 by plotting the cross-sectional area values ​​of the lumen A of the heart H calculated in step S3 along the time axis.

[0056] In step S5, the main unit control unit 31 determines whether or not to terminate the acquisition of the ultrasound image U. The main unit control unit 31 can determine to terminate the acquisition of the ultrasound image U if, for example, the user inputs an instruction to terminate the acquisition of the ultrasound image U via the input device 32, and to continue acquiring the ultrasound image U if the user does not give any specific instruction via the input device 32.

[0057] If the main control unit 31 determines in step S5 to continue acquiring the ultrasound image U, the process returns to step S1, and a new frame of the ultrasound image U is acquired. In step S2, the contour B of the lumen A of the heart H shown in the ultrasound image U is extracted, and based on this contour B, the cross-sectional area of ​​the lumen A of the heart H is calculated in step S3. Furthermore, in step S4, the cross-sectional area of ​​the lumen A of the heart H, newly calculated in the current step S4, is plotted along the time axis on the cross-sectional area change curve C1 or C2 generated in the previous step S4, and the cross-sectional area change curve C1 or C2 is updated. By repeating the process from steps S1 to S5 in this manner, a cross-sectional area change curve C1 or C2 is obtained, which shows that the value of the cross-sectional area is actually changing periodically, as shown in Figures 5 and 6. If it is determined in step S5 to terminate the acquisition of the ultrasound image U, the process proceeds to step S6.

[0058] In step S6, the cross-sectional area change prediction curve generation unit 27 generates a cross-sectional area change prediction curve D1 based on the cross-sectional area change curve C1 or C2 generated in the final step S4, assuming that segmentation was successfully performed in all of the multiple frames of ultrasound images U, as shown in Figure 7, for example. The cross-sectional area change prediction curve generation unit 27 can generate a curve as the cross-sectional area change prediction curve D1 by approximating the cross-sectional area change curve C1 or C2 with a Bézier curve using a method such as the least squares method.

[0059] In step S7, the error detection unit 28 performs a process to detect a segmentation error performed in step S2 in any of the multiple frames by comparing the cross-sectional area change curve C1 or C2 generated in the final step S4 with the cross-sectional area change prediction curve D1 generated in step S6. The error detection unit 28 can detect this as a segmentation error if, for example, as shown in Figure 7, a part of the waveform is distorted, as in the cross-sectional area change curve C2, causing a part of the cross-sectional area change curve C2 and a part of the cross-sectional area change prediction curve D1 to be separated from each other.

[0060] Thus, since the error detection unit 28 automatically detects segmentation errors by comparing the cross-sectional area change curve C1 or C2 with the cross-sectional area change prediction curve D1, the user of the ultrasound diagnostic device does not need to check multiple frames of ultrasound images U to find segmentation errors, and segmentation errors can be found quickly and easily.

[0061] In step S8, the main unit control 31 refers to the result of the error detection process in step S7 and determines whether or not a segmentation error has been detected. If it is determined that a segmentation error has been detected, the process proceeds to step S9.

[0062] In step S9, the error detection unit 28 notifies the user that it detected a segmentation error in step S7. The error detection unit 28 can notify the user by displaying a message such as "A segmentation error has been detected" along with the cross-sectional area change curve C2 and the cross-sectional area change prediction curve D1 shown in Figure 7 on the monitor 23.

[0063] In step S10, the error correction unit 30 corrects the segmentation error by deforming the contour B of the lumen A of the heart H in the ultrasound image U of the frame in which the segmentation error was detected, based on the user's input via the input device 32. At this time, the user checks the error notification result in step S9, selects the ultrasound image U of the frame in which the segmentation error was detected from among multiple ultrasound images U, and inputs an instruction to deform the contour B extracted in the ultrasound image U of that frame.

[0064] The error correction unit 30, as shown in Figures 8 and 9, for example, sets multiple control points P on the ultrasound image U that can be moved by the user, which are used to deform the extracted contour B in the image region R specified by the user, and can deform the contour B in response to the user's movement of the control points P. In addition, the error correction unit 30 can detect the edge F of the lumen A of the heart H in the image region R specified by the user, and set the number of multiple control points P by comparing the number of inflection points of the detected edge F with the number of inflection points of the contour B in the image region R.

[0065] The error correction unit 30 can, for example, place a predetermined first number of control points P on the ultrasound image U in order to deform contour B to be smooth when the number of inflection points of edge F is less than the number of inflection points of contour B, that is, when edge F is smoother than contour B, as shown in Figure 8. Furthermore, the error correction unit 30 can, for example, place a predetermined second number of control points P (more than the predetermined first number) on the ultrasound image U in order to deform contour B to include more irregularities when the number of inflection points of edge F is greater than the number of inflection points of contour B, that is, when edge F contains more irregularities than contour B, as shown in Figure 9.

[0066] When the error correction unit 30 deforms the contour B of the lumen A of the heart H, information about the deformed contour B is stored in the memory 29. Subsequently, the information about the deformed contour B is read from the memory 29 and sent to the cross-sectional area calculation unit 25. The cross-sectional area calculation unit 25 recalculates the cross-sectional area of ​​the lumen A of the heart H based on the deformed contour B. The cross-sectional area change curve generation unit 26 updates the cross-sectional area change curve C2 by re-plotting the cross-sectional area of ​​the lumen A of the heart H obtained by the recalculation, instead of the cross-sectional area calculated based on the contour B before deformation. The cross-sectional area change prediction curve generation unit 27 regenerates the cross-sectional area change prediction curve D1 based on the updated cross-sectional area change curve C2. The user can deform the contour B while checking the updated cross-sectional area change curve C2, the regenerated cross-sectional area change prediction curve D1, and the ultrasound image U in which the contour B is being corrected, which are displayed on the monitor 23, respectively.

[0067] Thus, the error correction unit 30 can automatically set the number of control points P by comparing the detected edge F with the contour B in the image region R, and can easily deform the extracted contour B to be closer to the original contour of the lumen A of the heart H based on user instructions. The cross-sectional area calculated from the contour B finally obtained after correction by the error correction unit 30 is used, for example, to calculate the ejection fraction of the lumen A of the heart H.

[0068] Once the process in step S10 is completed, the operation of the ultrasound diagnostic device according to the flowchart in Figure 10 is complete. Furthermore, if, for example, the cross-sectional area change curve C1 generated in the final step S4 does not contain distortion in its waveform, and all of the cross-sectional area change curve C1 and the cross-sectional area change prediction curve D1 overlap each other, and no segmentation error is detected in step S7, then it is determined in step S8 that no segmentation error was detected. In this case, the processing in steps S9 and S10 is skipped, and the operation of the ultrasound diagnostic device according to the flowchart in Figure 10 is completed. In this case, the cross-sectional areas corresponding to the maximum and minimum values ​​in the cross-sectional area change curve C1 are used, for example, to calculate the ejection fraction of the lumen A of the heart H.

[0069] As described above, according to the ultrasound diagnostic apparatus of the embodiment of the present invention, the contour extraction unit 24 performs segmentation on multiple frames of ultrasound images U to extract the contour B of the lumen A of the heart H for each frame, the cross-sectional area calculation unit 25 calculates the cross-sectional area of ​​the lumen A of the heart H for each frame based on the extracted contour B, the cross-sectional area change curve generation unit 26 generates a cross-sectional area change curve C1 or C2 that represents the actual temporal change of the cross-sectional area based on the calculated cross-sectional area, and the error detection unit 28 detects errors in the segmentation performed by the contour extraction unit 24 by comparing the cross-sectional area change curve C1 or C2 with a predicted cross-sectional area change prediction curve D1, so that the segmentation result of the lumen A of the heart H can be corrected quickly and easily.

[0070] Although it is explained that the transmitting / receiving circuit 12 is provided in the ultrasonic probe 1, the transmitting / receiving circuit 12 may also be provided in the main body of the device 2. Furthermore, although it is explained that the image generation unit 21 is provided in the main body 2 of the device, the image generation unit 21 may also be provided in the ultrasonic probe 1.

[0071] The main unit 2 of the device may be a stationary type, a portable type that is easy to carry, or a handheld type, for example, composed of a smartphone or tablet computer. Thus, the type of equipment that makes up the main unit 2 is not particularly limited.

[0072] Generally, when calculating the ejection fraction of the lumen A of the heart H, ultrasound images U corresponding to the end-diastolic and end-systolic phases of the heart H are often used. In ultrasound images U corresponding to the end-diastolic and end-systolic phases of the heart H, the cross-sectional area of ​​the lumen A of the heart H reaches a maximum or minimum value in the cross-sectional area change curve C1 or C2. Therefore, the error detection unit 28 can detect an error only if, for example, the portions of the cross-sectional area change prediction curve D1 corresponding to the maximum and minimum values ​​do not overlap with the cross-sectional area change curve C2, or it can notify the user of the error only in this case. This allows the user to easily understand that a segmentation error has been detected in the ultrasound image U of the frame used to calculate the ejection fraction of the lumen A of the heart H.

[0073] Furthermore, while it is explained that the error detection unit 28 notifies of segmentation errors by displaying a message on the monitor 23, the notification method is not limited to displaying a message. For example, if the ultrasound diagnostic device is equipped with a speaker (not shown), the error detection unit 28 can also notify the user by sound emitted from the speaker. Also, if the ultrasound diagnostic device is equipped with a light-emitting unit (not shown), the error detection unit 28 can also notify the user by light emitted from the light-emitting unit. [Explanation of Symbols]

[0074] 1 Ultrasound probe, 2 Main unit, 11 Transducer array, 12 Transmit / receive circuit, 21 Image generation unit, 22 Display control unit, 23 Monitor, 24 Contour extraction unit, 25 Cross-sectional area calculation unit, 26 Cross-sectional area change curve generation unit, 27 Cross-sectional area change prediction curve generation unit, 28 Error detection unit, 29 Memory, 30 Error correction unit, 31 Main unit control unit, 32 Input device, 33 Image acquisition unit, 34 Processor, 41 Pulsar, 42 Amplifier unit, 43 AD conversion unit, 44 Beamformer, 45 Signal processing unit, 46 DSC, 47 Image processing unit, 4C Apical four-chamber section, A Lumen, B Contour, C1, C2 Cross-sectional area change curve, D1 Cross-sectional area change prediction curve, F: Edge, H: Heart, P: Control point, R: Image region, U: Ultrasound image.

Claims

1. An image acquisition unit that acquires multiple frames of ultrasound images of the subject's heart in a time-series sequence, A contour extraction unit performs segmentation on the multiple frames of ultrasound images and extracts the contour of the lumen of the heart for each frame. A cross-sectional area calculation unit calculates the cross-sectional area of ​​the lumen of the heart for each frame based on the contour extracted by the contour extraction unit, A cross-sectional area change curve generation unit generates a cross-sectional area change curve that represents the actual temporal change of the cross-sectional area based on the cross-sectional area calculated by the cross-sectional area calculation unit, An error detection unit detects the segmentation error performed in the contour extraction unit by comparing the cross-sectional area change curve generated by the cross-sectional area change curve generation unit with a predicted cross-sectional area change prediction curve. An ultrasound diagnostic device equipped with the following features.

2. The ultrasonic diagnostic apparatus according to claim 1, wherein the error detection unit notifies the user of the error detection result.

3. The ultrasonic diagnostic apparatus according to claim 1, wherein the error detection unit detects that the cross-sectional area change curve and the cross-sectional area change prediction curve are at least partially separated from each other as an error.

4. The ultrasonic diagnostic apparatus according to claim 1, further comprising a prediction curve generation unit that generates a prediction curve of cross-sectional change based on the cross-sectional change curve generated by the cross-sectional change curve generation unit.

5. The ultrasound diagnostic apparatus according to claim 1, further comprising an error correction unit for correcting the error in the segmentation performed in the contour extraction unit.

6. The error correction unit is, Multiple control points that can be operated by the user are set for the contour extracted by the contour extraction unit. The ultrasonic diagnostic apparatus according to claim 5, which corrects the error in the segmentation by deforming the contour based on the movement of at least one of the plurality of control points by the user.

7. The error correction unit is, In the image region specified by the user, the number of inflection points in the contour extracted by the contour extraction unit is calculated. In the aforementioned image region, the edge of the lumen of the heart is detected and the number of inflection points at the edge is calculated. The ultrasonic diagnostic apparatus according to claim 6, wherein the number of control points is set by comparing the number of inflection points in the contour with the number of inflection points in the edge.

8. By acquiring multiple ultrasound images of the subject's heart in a time-series sequence, Segmentation is performed on the ultrasound images of the multiple frames to extract the outline of the lumen of the heart for each frame. Based on the extracted contour, the cross-sectional area of ​​the lumen of the heart is calculated for each frame. Based on the calculated cross-sectional area, a cross-sectional area change curve is generated that represents the actual change in the cross-sectional area over time. The segmentation error is detected by comparing the generated cross-sectional area change curve with the predicted cross-sectional area change curve. A method for controlling an ultrasound diagnostic device.

Images