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 accurately identify the examination site regardless of the examiner's skill level. [Solution] The ultrasound diagnostic device includes an image acquisition unit that inputs an ultrasound image when an examiner performs an ultrasound examination on a subject, and a part discrimination unit that determines the examination area shown in the ultrasound image based on posture information of the examiner and subject obtained by analyzing the reflected signal when a detection signal is transmitted from a distance measuring device to the examiner and subject, and the ultrasound image. The part discrimination unit has multiple learning models that correspond to multiple predetermined divisions of the human body and have learned the relationship between the ultrasound image of the subject and the examination area. Based on the posture information, one of the multiple learning models is selected, and based on the ultrasound image, the examination area is determined in the division corresponding to one learning model using one learning model.

Description

Technical Field

[0001] The present invention relates to an ultrasonic diagnostic apparatus for specifying an examination position of a subject and a control method for the ultrasonic diagnostic apparatus.

Background Art

[0002] Conventionally, an ultrasonic image representing a tomographic image in a subject has been captured using a so-called ultrasonic diagnostic apparatus. Usually, an examiner often determines the examination site of the subject currently being imaged by checking the ultrasonic image. However, since the appearance of the ultrasonic image varies due to various factors such as differences in the shape of the site depending on the subject and differences in the tomographic plane being scanned, the examiner may misjudge the examination site by simply checking the ultrasonic image, especially when the level of proficiency is low.

[0003] In order to prevent such misjudgment of the site, for example, as disclosed in Patent Documents 1 and 2, techniques for automatically determining the examination site by analyzing the ultrasonic image have been developed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] According to the techniques disclosed in Patent Documents 1 and 2, the examination site is automatically determined regardless of the proficiency of the examiner. However, there are cases where the analysis of the ultrasonic image is not performed normally for some reason, and there is room for improvement in the accuracy of determining the examination site.

[0006] This invention was made to solve the problems of the conventional methods, and aims to provide an ultrasound diagnostic device and a control method for the ultrasound diagnostic device that can accurately identify the examination site regardless of the examiner's skill level. [Means for solving the problem]

[0007] The above objective can be achieved with the following configuration. [1] An image acquisition unit that inputs an ultrasound image of the subject when the examiner performs an ultrasound examination on the subject, A body part discrimination unit determines the area being examined based on the ultrasound image obtained by analyzing the reflected signals when a detection signal is transmitted from the distance measuring device to the examiner and the subject, along with the posture information of the examiner and the subject, and the ultrasound image obtained by the image acquisition unit. Equipped with, The part identification unit is, It has multiple learning models that correspond to several distinct and defined sections of the human body and that have learned the relationship between the ultrasound image of the subject and the examination site. Based on posture information, select one learning model from among several learning models. An ultrasound diagnostic device that uses ultrasound images to identify the examination site within a category corresponding to a single learning model. [2] The learning model is Having multiple candidate sites for the examination site, Based on posture information, select at least one candidate body part from multiple candidate body parts. The ultrasound diagnostic apparatus according to [1], which outputs one of at least one candidate sites as the examination site based on an ultrasound image. [3] The ultrasound diagnostic apparatus according to [1] or [2], which is equipped with an information memory that stores posture information and ultrasound images linked to each other. [4] The image acquisition unit is: Ultrasound probe and An image generation unit generates an ultrasound image of a subject by transmitting and receiving an ultrasound beam using an ultrasound probe. An ultrasound diagnostic device as described in any of [1] to [3], including the following: [5] Obtain an ultrasound image of the subject, Based on posture information of the examiner and subject obtained by analyzing the reflected signals when a detection signal is transmitted from the distance measuring device to the examiner and subject, one learning model is selected from among several learning models that correspond to multiple different parts of the human body and have learned the relationship between the subject's ultrasound image and the area being examined shown in the ultrasound image. Based on ultrasound images, a single learning model identifies the examination site within the corresponding section of the learning model. A method for controlling an ultrasound diagnostic device. [6] The learning model is Having a plurality of candidate sites for the aforementioned inspection site, Based on the posture information, select at least one candidate body part from the plurality of candidate body parts. A control method for an ultrasound diagnostic apparatus according to [5], which outputs one of the at least one candidate sites as the examination site based on the ultrasound image. [7] A control method for an ultrasound diagnostic apparatus according to [5] or [6], which stores the posture information and the ultrasound image linked to each other. [8] A control method for an ultrasound diagnostic apparatus according to any one of [5] to [7], which generates an ultrasound image of a subject by transmitting and receiving an ultrasound beam using an ultrasound probe. [Effects of the Invention]

[0008] According to the present invention, an ultrasonic diagnostic apparatus includes an image acquisition unit that inputs an ultrasonic image of a subject by an examiner performing an ultrasonic examination on the subject, posture information of the examiner and the subject obtained by analyzing a reflection signal when a detection signal is transmitted to the examiner and the subject from a distance measuring device, and a part discrimination unit that discriminates an examination part shown in the ultrasonic image based on the ultrasonic image obtained by the image acquisition unit. The part discrimination unit has a plurality of learning models corresponding to a plurality of defined different sections of the human body and learning the relationship between the ultrasonic image of the subject and the examination part. Based on the posture information, one learning model is selected from the plurality of learning models, and based on the ultrasonic image, the examination part is discriminated in the section corresponding to the one learning model by the one learning model, so that the examination part can be discriminated with high accuracy regardless of the proficiency of the examiner.

Brief Description of Drawings

[0009] [Figure 1] It is a block diagram showing the configuration of an ultrasonic diagnostic apparatus according to an embodiment of the present invention. [Figure 2] It is a block diagram showing the configuration of a transmission / reception circuit in an embodiment of the present invention. [Figure 3] It is a block diagram showing the configuration of an image generation unit in an embodiment of the present invention. [[ID=十六]] [Figure 4] [[ID=十七]]It is a diagram schematically showing an example of the positional relationship among a distance measuring sensor unit, a subject, and an examiner in an embodiment of the present invention. [Figure 5] 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" shall 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, a diagnostic apparatus 2 connected to the ultrasonic probe 1, and a distance measuring sensor unit 3 connected to the diagnostic apparatus 2. The ultrasonic probe 1 includes a transducer array 11 and a transmission / reception circuit 12 connected to the transducer array 11. The distance measuring sensor unit 3 includes a transmission unit 31 and a reception unit 32.

[0012] The diagnostic apparatus 2 is connected to the ultrasonic probe 1 and displays an ultrasonic image captured by the ultrasonic probe 1. The diagnostic apparatus 2 is used, for example, for an examiner to confirm an ultrasonic image captured in real time by the ultrasonic probe 1.

[0013] The diagnostic apparatus 2 includes an image generation unit 21 connected to the transmission / reception circuit 12 of the ultrasonic probe 1, and a display control unit 22 and a monitor 23 are sequentially connected to the image generation unit 21. The image acquisition unit 41 is constituted by the image generation unit 21 and the ultrasonic probe 1. The diagnostic apparatus 2 also includes a signal analysis unit 24 connected to the reception unit 32 of the distance measuring sensor unit 3. A part discrimination unit 25 is connected to the image generation unit 21 and the signal analysis unit 24. The part discrimination unit 25 is connected to the display control unit 22. An information memory 26 is connected to the image generation unit 21 and the signal analysis unit 24. The device control unit 27 is connected to the transmission / reception circuit 12, the display control unit 22, the signal analysis unit 24, the part discrimination unit 25, and the information memory 26. An input device 28 is connected to the device control unit 27.

[0014] Furthermore, the display control unit 22, signal analysis unit 24, part discrimination unit 25, and device control unit 27 constitute the processor 29 for the diagnostic device 2. In addition, the signal analysis unit 24 and distance measuring sensor unit 3 of the diagnostic device 2 constitute the distance measuring device 42.

[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 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 device control unit 27. As shown in Figure 2, the transmitting / receiving circuit 12 includes a pulser 51 connected to the transducer array 11, and an amplifier 52, an AD (Analog to Digital) converter 53, and a beamformer 64 connected sequentially in series from the transducer array 11.

[0017] The pulser 51 includes, for example, multiple pulse generators, and based on a transmission delay pattern selected in response to a control signal from the device control unit 27, 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.

[0018] 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 52.

[0019] The amplification unit 52 amplifies the signals input from each ultrasonic transducer constituting the transducer array 11 and transmits the amplified signals to the AD conversion unit 53. The AD conversion unit 53 converts the signals transmitted from the amplification unit 52 into digital received data. The beamformer 64 performs so-called receive focus processing by adding each received data received from the AD conversion unit 53 with a corresponding delay. Through this receive focus processing, each received data converted by the AD conversion unit 53 is phase-aligned and added together, and a sound ray signal with a focused ultrasonic echo is obtained.

[0020] As shown in Figure 6, the image generation unit 21 has a configuration in which a signal processing unit 55, a DSC (Digital Scan Converter) 56, and an image processing unit 57 are connected in series in sequence.

[0021] The signal processing unit 55 receives the sound line signal from the transmitting / receiving circuit 12, applies a sound velocity value set by the device control unit 27 to correct for attenuation due to distance according to the depth of the ultrasonic reflection position, and then performs envelope detection processing to generate a B-mode image signal, which is tomographic image information of the tissue within the subject.

[0022] The DSC56 converts the B-mode image signal generated by the signal processing unit 55 into an image signal that follows the scanning method of a normal television signal (raster conversion). The image processing unit 57 performs various necessary image processing, such as gradation processing, on the B-mode image signal input from the DSC 56, and then sends the B-mode image signal to the display control unit 22, the part discrimination unit 25, and the information memory 26. The B-mode image signal processed in this way by the image processing unit 57 is called an ultrasound image.

[0023] The display control unit 22, under the control of the device control unit 27, performs predetermined processing on the ultrasound image etc. generated by the image generation unit 21 and displays it on the monitor 23. The monitor 23 displays various information under the control of the device control unit 27. The monitor 23 may include, for example, a display device such as an LCD (Liquid Crystal Display) or an organic EL display (Organic Electroluminescence Display).

[0024] The distance measuring sensor unit 3 is positioned near the examiner J and the subject K, as shown in Figure 4, for example, when the examiner J is performing an examination on the subject K using the ultrasound probe 1. The sensor unit 3 transmits a detection signal to the examiner J and the subject K, and receives reflected signals from them. The example in Figure 4 depicts the subject K lying on the examination table T, with the examiner J examining the subject K's arm with the ultrasound probe 1.

[0025] The transmitting unit 31 of the distance measuring sensor unit 3 transmits a detection signal to the inspector J and the subject K. The transmitting unit 31 is a so-called electromagnetic wave wireless transmitter and includes, for example, an antenna for transmitting electromagnetic waves, a signal source such as an oscillator circuit, a modulation circuit for modulating the signal, and an amplifier for amplifying the signal. The receiving unit 32 includes an antenna for receiving electromagnetic waves and receives reflected signals from the examiner J and the subject K.

[0026] The distance measuring sensor unit 3 can be configured, for example, with a radar that transmits and receives detection signals in the so-called Wi-Fi® standard, consisting of electromagnetic waves with a center frequency of 2.4 GHz or 5 GHz, or with a radar that transmits and receives broadband detection signals with a center frequency of 1.78 GHz. Furthermore, the distance measuring sensor unit 3 can also be configured with a so-called LIDAR (Light Detection and Ranging, or Laser Imaging Detection and Ranging) sensor that transmits short-wavelength electromagnetic waves such as ultraviolet light, visible light, or infrared light as detection signals.

[0027] The signal analysis unit 24 of the diagnostic device 2 analyzes the reflected signal received by the distance measuring sensor unit 3 to acquire posture information of the examiner J and the subject K. The posture information of the examiner J and the subject K includes, for example, information on the position of each part of the examiner J and the subject K, such as the head, shoulders, arms, waist, and legs.

[0028] The signal analysis unit 24 can acquire posture information of the examiner J and the subject K using a machine learning model that has learned the reflected signals when a detection signal is transmitted to the human body by the distance measuring sensor unit 3. Specifically, the signal analysis unit 24 uses, for example, “ZHAO, Mingmin, et al. Through-wall human pose estimation using radio signals. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018. p. 7356-7365.”, “VASILEIADIS, Manolis; BOUGANIS, Christos-Savvas; estimation from 3D cloud data using 3D convolutional neural networks. Computer Vision and Image Understanding, 2019, 185: 12-23.", "JIANG, Wenjun, et al. Towards 3D human pose construction using WiFi. In: Proceedings of the 26th Annual International Conference on Mobile Computing and Networking. 2020. p. 1-14.", or "WANG, Fei, et al. Person-in-WiFi: Posture information can be acquired using the method described in "Fine-grained person perception using WiFi. In: Proceedings of the IEEE / CVF International Conference on Computer Vision. 2019. pp. 5452-5461."

[0029] Furthermore, the signal analysis unit 24 can set a coordinate system with the position of one part of the subject K as the origin, and acquire the 3D coordinates of the positions of each part of the subject K as posture information of the subject K. For example, the signal analysis unit 24 can set a 3D coordinate system by setting the position of the subject K's neck as the origin of the coordinate system, the axis along the straight line passing through the subject's left and right shoulders as the first axis, the axis perpendicular to the first axis and along the straight line passing through the subject K's head and torso as the second axis, and the axis perpendicular to both the first and second axes as the third axis. As a result, even if the positional relationship between the distance measuring sensor unit 3 and the subject K differs from one examination to the next, the positions of each part of the subject K can be represented in the same 3D coordinate system.

[0030] Here, for example, the position on subject K where the tip of examiner J's arm is located can be identified as the position on subject K to which examiner J is making contact with the ultrasound probe 1. Therefore, when the human body is divided into multiple sections, the section on subject K currently being examined can be identified based on the posture information of subject K and examiner J.

[0031] Furthermore, it is possible to identify the location of the subject K as seen in an ultrasound image by using, for example, machine learning or deep learning that has learned the relationship between the characteristics of tissue structures visible in ultrasound images and the names of the corresponding parts of the subject K, or by using image analysis such as template matching. However, due to differences in the shape and size of anatomical structures among subjects, or because clear ultrasound images are not obtained due to the examiner J's low skill level, attempting to identify the examination site based solely on ultrasound images sometimes resulted in incorrect identification of the examination site.

[0032] Therefore, in order to improve the accuracy of identifying the examination area, the area discrimination unit 25 identifies the examination area shown in the ultrasound image based on both the posture information of the subject K and examiner J acquired by the signal analysis unit 24 and the ultrasound image of the subject K generated by the image generation unit 21. As a result, when identifying the examination area based on the ultrasound image, the classification of the subject K's body, which is identified from the posture information of the subject K and examiner J, is taken into consideration, enabling high-precision identification of the examination area.

[0033] In this process, the body part discrimination unit 25 can determine the current body part of the subject based on the posture information of the subject K and examiner J acquired by the signal analysis unit 24 and the ultrasound image generated by the image generation unit 21. This is done using a learning model that has learned the relationship between the posture of the subject K and examiner J, the ultrasound image taken in that posture of the subject K and examiner J, and the body part of the subject being examined.

[0034] More specifically, the learning model of the body part discrimination unit 25 has a predetermined set of candidate body parts to be examined, and can calculate the probability that the body part visible in the ultrasound image is one of these candidate body parts. In this process, the learning model can, for example, weight the probability of at least one candidate body part corresponding to a body part of the subject identified from the posture information of the subject K and the examiner J. The learning model can then identify the candidate body part with the highest probability among the multiple candidate body parts calculated in this way as the body part to be examined.

[0035] The region discrimination unit 25 can use models that follow algorithms such as ResNet (Residual Neural Network), DenseNet (Dense Convolutional Network), AlexNet, Baseline, Batch Normalization, Dropout Regularization, NetWidth Search, or NetDepth Search as learning models. The region discrimination unit 25 can also use models that follow these algorithms in appropriate combinations.

[0036] In this way, the body part discrimination unit 25 uses not only the ultrasound image of the subject K but also the posture information of the subject K and the examiner J as auxiliary information to determine the current examination area. Therefore, the examination area can be determined with high accuracy regardless of the examiner J's skill level. Furthermore, it can determine the examination area more accurately than, for example, determining the examination area based solely on the ultrasound image.

[0037] Furthermore, the learning model of the body part discrimination unit 25 can be composed of multiple sub-learning models that correspond to a set of predetermined divisions of the human body and have learned the relationship between the ultrasound image of the subject K and the examination site. In this case, the learning model selects one sub-learning model from among the multiple sub-learning models that corresponds to the division of the subject K's body identified from the posture information, and by inputting the ultrasound image into the selected sub-learning model, it can discriminate the examination site in the corresponding division. More specifically, the learning model can be composed of sub-learning models for, for example, the abdomen, chest, and upper limbs, and when the division of the body identified from the posture information of the subject K and examiner J is the abdomen, it will discriminate one of the candidate sites located in the abdomen, such as the liver and kidneys, as the examination site.

[0038] Furthermore, the learning model can consist of a single learning model that selects at least one candidate body part from among multiple candidate body parts, which is included in the division of the human body identified based on the posture information of the subject K and examiner J, and outputs one of the selected candidate body parts as the examination body part based on the ultrasound image. For example, the learning model has the liver, kidneys, heart, lungs, and diaphragm as multiple defined candidate body parts, and when the abdomen of subject K is identified from the posture information of subject K and examiner J, the learning model can select the liver and kidneys, which correspond to the abdomen, as candidate body parts from among the multiple candidate body parts. In this case, the learning model calculates, for example, the probability that the liver is the examination body part and the probability that the kidney is the examination body part, and identifies the body part with the highest probability as the examination body part. In this case, for example, the process of calculating the probability is performed for only at least one of the multiple candidate body parts, so the computational load on the body part discrimination unit 25 can be reduced.

[0039] Furthermore, the learning model can, for example, calculate probabilities for multiple candidate sites based on ultrasound images, and then select at least one candidate site from those multiple sites based on the posture information of the subject K and the examiner J. For example, if the learning model has a set of candidate sites including the liver, kidneys, heart, lungs, and diaphragm, and the abdomen of subject K is identified from the posture information of subject K and examiner J, then the liver and kidneys can be selected as candidate sites from the liver, kidneys, heart, lungs, and diaphragm for which probabilities have been calculated. In this case, the learning model will, for example, determine the site with the highest probability among the probabilities of the liver being the examination site and the probability of the kidney being the examination site. For example, even if, for some reason, the probability of the heart, lungs, or diaphragm being the examination site becomes higher than the probability of the liver and kidney being the examination site, the liver and kidneys will still be selected as candidate sites based on the posture information, making it possible to accurately determine the examination site.

[0040] Furthermore, the site discrimination unit 25 can also discriminate the examination site without using a learning model. For example, the site discrimination unit 25 has template data representing the typical shape, etc., for each of several candidate sites, and can calculate the probability that the examination site is one of several candidate sites by comparing the anatomical structure shown in the ultrasound image with the multiple template data, using a method known as template matching, and can discriminate the examination site based on the calculated probability.

[0041] In this case, the body part discrimination unit 25 can assign weights to the probabilities of at least one body part corresponding to a body category identified from the posture information of the subject K and examiner J. Alternatively, the body part discrimination unit 25 can select at least one body part from a predetermined group of candidate body parts that corresponds to a body category identified from the posture information of the subject K and examiner J, and then determine the examination body part.

[0042] Furthermore, the body part discrimination unit 25 can display the name of the examined body part of the identified subject on the monitor 23.

[0043] The information memory 26 stores both the posture information of the subject K and examiner J acquired by the signal analysis unit 24 and the ultrasound image generated by the image generation unit 21 corresponding to that posture information, linked together under the control of the device control unit 27. The information memory 26 can link posture information and ultrasound images together, for example, by recording the examination position in the so-called header information of the ultrasound image under the control of the device control unit 27. The information memory 26 can also link posture information and ultrasound images together, for example, by using a so-called timestamp or so-called DICOM (Digital Imaging and Communications in Medicine), under the control of the device control unit 27. The posture information of the subject K and examiner J and the ultrasound image stored in the information memory 26 can also be read out by an input operation by the examiner J, for example, via the input device 28, and sent to the body part discrimination unit 25. This allows the body part discrimination unit 25 to accurately determine the examination area shown in the ultrasound image, even after the examination is completed, regardless of the examiner J's skill level.

[0044] For example, the information memory 26 can be a recording medium 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).

[0045] The device control unit 27 controls each part of the diagnostic device 2 according to a pre-recorded program or the like. The input device 28 receives input operations from the inspector J, etc., and sends the input information to the device control unit 27. The input device 28 is composed of, for example, a keyboard, mouse, trackball, touchpad, and touch panel, or other devices for the inspector to perform input operations.

[0046] The processor 29 of the diagnostic device 2, which includes a display control unit 22, a signal analysis unit 24, a part discrimination unit 25, and a device control unit 27, is composed of a CPU (Central Processing Unit) and a control program for causing the CPU to perform various processes. However, it may also be composed of an FPGA (Field Programmable Gate Array), a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), a GPU (Graphics Processing Unit), or other ICs (Integrated Circuits), or a combination thereof.

[0047] Furthermore, the display control unit 22, signal analysis unit 24, part discrimination unit 25, and device control unit 27 of the processor 29 can be partially or entirely integrated into a single CPU or the like.

[0048] Next, an example of the operation of the ultrasound diagnostic device according to the embodiment will be explained using the flowchart in Figure 5. First, in step S1, the distance measuring sensor unit 3 begins to continuously transmit detection signals to the subject K and the examiner J, and to continuously receive reflected signals from the subject K and the examiner J. At this time, the examiner J brings the ultrasound probe 1 into contact with the surface of the subject K's body.

[0049] Next, in step S2, the signal analysis unit 24 detects the subject K and the examiner J by analyzing the reflected signal received by the distance measuring sensor unit 3 in step S1.

[0050] In the subsequent step S3, the signal analysis unit 24 analyzes the reflected signal received by the distance measuring sensor unit 3 in step S1 to acquire posture information of the subject K and examiner J detected in step S2. At this time, the signal analysis unit 24 sets a coordinate system with one part of the subject K as the origin and can acquire the 3D coordinates of the positions of each part of the subject K as posture information of the subject K. For example, the signal analysis unit 24 can set a 3D coordinate system by setting the position of the subject K's neck as the origin of the coordinate system, the axis along the line passing through the subject's left and right shoulders as the first axis, the axis perpendicular to the first axis and along the line passing through the subject K's head and torso as the second axis, and the axis perpendicular to both the first and second axes as the third axis. As a result, even if the positional relationship between the distance measuring sensor unit 3 and the subject K differs from one examination to the next, the positions of each part of the subject K can be represented in the same 3D coordinate system.

[0051] The posture information of the subject K and examiner J obtained in step S3 is sent to the body part discrimination unit 25 and the information memory 26.

[0052] In step S4, with the ultrasound probe 1 in contact with the body surface of the subject K, the ultrasound probe 1 scans the inside of the subject K, and an ultrasound image representing a tomographic image of the inside of the subject K is acquired. At this time, the transmitting / receiving circuit 12 generates an acoustic ray signal by performing so-called reception focus processing under the control of the device control unit 27. The acoustic ray signal generated by the transmitting / receiving circuit 12 is sent to the image generation unit 21. The image generation unit 21 generates an ultrasound image using the acoustic ray signal sent from the transmitting / receiving circuit 12. The ultrasound image acquired in this way is sent to the site discrimination unit 25 and the information memory 26.

[0053] Here, the posture information of the subject K and examiner J sent to the information memory 26 in step S3, and the ultrasound image sent to the information memory 26 in step S4, can be linked together and stored in the information memory 26 under the control of the device control unit 27.

[0054] In step S5, the body part discrimination unit 25 determines the area to be examined in the ultrasound image based on both the posture information of the subject K and examiner J acquired in step S3, and the ultrasound image generated by the image generation unit 21 in step S4. The body part discrimination unit 25 has, for example, a learning model that has been pre-learned the relationship between the posture of the subject K and examiner, the ultrasound image acquired in accordance with that posture, and the area to be examined shown in that ultrasound image. By inputting the posture information of the subject K and examiner J and the ultrasound image into the learning model, the body part to be examined can be determined.

[0055] Here, the body segment of subject K from which the ultrasound image is being taken can be identified based on the posture information of subject K and examiner J. Therefore, when determining the examination site based on the ultrasound image, for example, the accuracy of determining the examination site can be improved by taking into account the identified body segment of subject K.

[0056] The site identification unit 25 displays the information of the identified examination site on the monitor 23. By checking the information of the examination site displayed on the monitor 23, the examiner J can easily understand the current examination site and proceed with the examination of the subject.

[0057] In step S6, the device control unit 27 determines whether or not to terminate the inspection. For example, if the inspector J inputs an instruction to terminate the inspection via the input device 28, the device control unit 27 determines to terminate the current inspection. Alternatively, if the inspector J does not input an instruction to terminate the inspection via the input device 28, the device control unit 27 determines to continue the current inspection.

[0058] If it is determined in step S6 to continue the test, the process returns to step S3. In this way, as long as it is determined in step S6 to continue the test, the process from step S3 to step S6 is repeated.

[0059] Furthermore, if it is determined in step S6 that the examination should be terminated, the device control unit 27 controls each part of the ultrasound diagnostic device to terminate the examination, and the operation of the ultrasound diagnostic device according to the flowchart in Figure 5 ends.

[0060] Here, after the examination is completed, for example, based on input operations by the examiner J via the input device 28, the posture information and ultrasound images stored in association with each other can be read from the information memory 26. In this case, the body part discrimination unit 25 can determine the examination area shown in the read ultrasound image based on the posture information and ultrasound images of the subject K and examiner J read from the information memory 26. As a result, for example, when a doctor or other professional checks the ultrasound image and makes a diagnosis of the subject, they can accurately grasp the examination area shown in the ultrasound image by confirming the name of the examination area determined by the body part discrimination unit 25, thereby improving the accuracy of the diagnosis.

[0061] As described above, according to the ultrasound diagnostic apparatus according to the embodiment of the present invention, the site discrimination unit 25 discriminates the examination site of the subject based on both the posture information of the subject K and examiner J acquired by the signal analysis unit 24 and the ultrasound image generated by the image generation unit 21. Therefore, the examination site can be determined with high accuracy regardless of the skill level of the examiner J. Furthermore, the ultrasound diagnostic apparatus according to the embodiment of the present invention can improve the accuracy of examination site determination compared to cases where the examination site is determined based solely on the ultrasound image.

[0062] Although the signal analysis unit 24 is described as being provided in the diagnostic device 2, for example, a distance measuring device 42 independent of the diagnostic device 2 can be configured using the distance measuring sensor unit 3 and the signal analysis unit 24. In this case, the signal analysis unit 24 of the distance measuring device 42 acquires posture information of the subject K and the examiner J, and the acquired posture information is sent to the body part discrimination unit 25 of the diagnostic device 2. Therefore, even in this case, similar to when the diagnostic device 2 is equipped with the signal analysis unit 24, the body part discrimination unit 25 determines the examination area based on both the posture information of the subject K and the examiner J acquired by the signal analysis unit 24 and the ultrasound image generated by the image generation unit 21.

[0063] Furthermore, although the distance measuring sensor unit 3 is shown to be installed near the examiner J and the subject K, as shown in Figure 2, for example, the installation location of the distance measuring sensor unit 3 is not particularly limited as long as the detection signal transmitted from the distance measuring sensor unit 3 reaches the examiner J and the subject K. The distance measuring sensor unit 3 can also be installed, for example, on the ceiling of the room in which the examiner J is conducting the examination of the subject K.

[0064] Furthermore, although it is explained that the signal analysis unit 24 sets a coordinate system with the position of one part of the subject K as the origin, the part discrimination unit 25 can also set a coordinate system with the position of one part of the subject K as the origin instead of the signal analysis unit 24. In this case, the signal analysis unit 24 can, for example, acquire the 3D coordinates of each part of the subject K with an arbitrary position, such as within the room where the distance measuring sensor unit 3 is installed, as the origin, based on the reflected signal transmitted from the receiving unit 32 of the distance measuring sensor unit 3. The part discrimination unit 25 converts the 3D coordinates of each part of the subject K acquired by the signal analysis unit 24 into a representation in a 3D coordinate system with the position of one part of the subject K as the origin. Even in such a case, for example, even if the positional relationship between the distance measuring sensor unit 3 and the subject K differs from one examination to the next, the positions of each part of the subject K can be represented in the same 3D coordinate system.

[0065] Furthermore, while the flowchart in Figure 5 shows the process proceeding in the order of step S3 and then step S4, steps S3 and S4 can also be processed in parallel.

[0066] Furthermore, while the flowchart in Figure 5 explains that an ultrasound image is generated in step S4 each time posture information is acquired in step S3, it is also possible that posture information is acquired once in step S3 each time a certain number of ultrasound images are generated in step S4. Alternatively, it is also possible that a single ultrasound image is generated in step S4 each time posture information is acquired multiple times in step S3.

[0067] Furthermore, although the image generation unit 21 is described as being provided in the diagnostic device 2, it can also be provided in the ultrasound probe 1 instead.

[0068] Furthermore, the diagnostic device 2 may be a stationary type, a portable type, or a handheld type consisting of a smartphone or tablet computer, etc. Thus, the type of equipment constituting the diagnostic device 2 is not particularly limited. Furthermore, the ultrasound probe 1 and the diagnostic device 2 can be connected to each other by wire or wirelessly. [Explanation of Symbols]

[0069] 1 Ultrasound probe, 2 Diagnostic device, 3 Distance sensor unit, 11 Transducer array, 12 Transmit / receive circuit, 21 Image generation unit, 22 Display control unit, 23 Monitor, 24 Signal analysis unit, 25 Site discrimination unit, 26 Information memory, 27 Device control unit, 28 Input device, 29 Processor, 31 Transmitter, 32 Receiver, 41 Image acquisition unit, 42 Distance measuring device, 51 Pulsar, 52 Amplifier, 53 AD conversion unit, 54 Beamformer, 55 Signal processing unit, 56 DSC, 57 Image processing unit, J Examiner, K Subject, T Examination table.

Claims

1. An image acquisition unit that inputs an ultrasound image of a subject when an examiner performs an ultrasound examination on the subject, A part determination unit determines the inspection area shown in the ultrasound image based on the posture information of the inspector and the subject obtained by analyzing the reflected signal when a detection signal is transmitted from the distance measuring device to the inspector and the subject, and the ultrasound image obtained by the image acquisition unit. Equipped with, The aforementioned part discrimination unit is It has multiple learning models that correspond to multiple distinct and defined sections of the human body and that have learned the relationship between the ultrasound image of the subject and the examination site. Based on the posture information, one learning model is selected from the plurality of learning models. An ultrasound diagnostic device that, based on the ultrasound image, identifies the examination site in a section corresponding to the learning model using the learning model.

2. The aforementioned learning model, Having a plurality of candidate sites for the aforementioned inspection site, Based on the posture information, select at least one candidate body part from the plurality of candidate body parts. The ultrasound diagnostic apparatus according to claim 1, which outputs one of the at least one candidate sites as the site to be examined based on the ultrasound image.

3. The ultrasound diagnostic apparatus according to claim 1, further comprising an information memory for storing the posture information and the ultrasound images in a linked manner.

4. The image acquisition unit, Ultrasound probe and An image generation unit generates an ultrasonic image of the subject by transmitting and receiving an ultrasonic beam using the ultrasonic probe. An ultrasound diagnostic apparatus according to any one of claims 1 to 3, including the following:

5. We acquire ultrasound images of the subject, Based on the posture information of the examiner and the subject obtained by analyzing the reflected signals when a detection signal is transmitted from the distance measuring device to the examiner and the subject, one learning model is selected from among multiple learning models that correspond to multiple different parts of the human body and have learned the relationship between the ultrasound image of the subject and the examination area shown in the ultrasound image. Based on the ultrasound image, the one learning model determines the examination site in the section corresponding to the one learning model. A method for controlling an ultrasound diagnostic device.

6. The aforementioned learning model, Having a plurality of candidate sites for the aforementioned inspection site, Based on the posture information, select at least one candidate body part from the plurality of candidate body parts. A control method for an ultrasound diagnostic apparatus according to claim 5, wherein one of the at least one candidate sites is output as the examination site based on the ultrasound image.

7. A control method for an ultrasound diagnostic apparatus according to claim 5, wherein the posture information and the ultrasound image are linked and stored together.

8. A control method for an ultrasound diagnostic apparatus according to any one of claims 5 to 7, which generates an ultrasound image of a subject by transmitting and receiving an ultrasound beam using an ultrasound probe.

Images