Information processing device, abnormality determination method, learning method, and program

Abstract

To easily determine an attachment abnormality of a component.SOLUTION: An ultrasonic diagnostic device 100 as an information processing device comprises an acquisition unit including a transmission unit 12 to a signal processing unit 15, and a control unit 18. The acquisition unit acquires ultrasonic image data. The control unit 18 uses learned estimation data for estimating noise in an ultrasonic image based on an attachment abnormality of a component of the ultrasonic diagnostic device 100 to estimate noise from the acquired ultrasonic image data. The control unit 18 determines the attachment abnormality of a component corresponding to the noise of the ultrasonic diagnostic device 100. The control unit 18 outputs determination result information.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an abnormality determination method, a learning method, and a program. [Background technology]

[0002] Conventionally, there has been known an ultrasound diagnostic apparatus that uses an ultrasound probe to irradiate ultrasound into the interior of a subject, receives and analyzes the reflected waves, and displays an ultrasound image of the interior of the subject. The subject is, for example, a living patient.

[0003] In ultrasound diagnostic equipment, noise can appear in images due to various factors such as: -Clutter noise caused by living organisms (not a malfunction). - Artifacts caused by reflections (not malfunctions). Noise can be introduced when a component fails, when the electrical grounding is weak due to poor assembly, or when an electrical device is placed close to the ultrasonic probe cable, depending on the usage environment. All of these are specific to ultrasound diagnostic devices.

[0004] Doctors and technicians routinely check ultrasound images for image noise. Image noise is determined visually, and can be easily overlooked if the user is unfamiliar with the features. Furthermore, image noise can be so severe that only an experienced technician can distinguish between normal and abnormal. Using an ultrasound diagnostic system while image noise is present poses a risk of misdiagnosis. It is essential to avoid using the system in this state. If this occurs, service and repair personnel must reproduce the problem, isolate the cause, and replace parts. This prevents the user from using the ultrasound diagnostic system for diagnosis, resulting in a missed diagnostic opportunity. Given the social issue of a shortage of service personnel (especially skilled service personnel), it is desirable for inspections to be easily performed by both users and service personnel.

[0005] Also known is an ultrasound imaging system that uses a neural network to detect a fault condition in a transducer or system (see Patent Document 1). When a fault condition is detected, the ultrasound imaging system adjusts parameters to compensate for the image data. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2022-500164 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the conventional ultrasound imaging systems focus on detecting a dropped transducer or physical damage to the interconnecting cable, and therefore can only detect fault conditions due to component failures in the wiring leading to the transducer.

[0008] For this reason, the ultrasonic imaging system cannot detect component installation errors, such as loosening of the component installation over time, and component installation errors can only be detected by visual inspection by an experienced person.

[0009] An object of the present invention is to easily determine whether a part is mounted abnormally. [Means for solving the problem]

[0010] In order to solve the above problem, the information processing device of the invention described in claim 1 comprises: an acquisition unit that acquires ultrasound image data; and a control unit that estimates noise from the acquired ultrasound image data using trained estimation data for estimating noise in an ultrasound image due to an installation abnormality of a part of the ultrasound diagnostic device, determines an installation abnormality of a part of the ultrasound diagnostic device corresponding to the noise, and outputs information on the determination result.

[0011] The invention described in claim 2 is the information processing device described in claim 1, The part mounting abnormality is an abnormality caused by an assembly error during the manufacture of the ultrasonic diagnostic apparatus, or an abnormality caused by a problem with the mounting of a part that occurs over time even when the ultrasonic diagnostic apparatus is used normally.

[0012] The invention described in claim 3 is the information processing device described in claim 1, the estimation data is data for estimating an abnormality location of an installation abnormality of a part corresponding to an estimated noise, The control unit uses the estimation data to estimate an abnormality location due to an installation error of a component of the ultrasound diagnostic device from the acquired ultrasound image data, and outputs result information including the abnormality location.

[0013] The invention described in claim 4 is the information processing device described in claim 3, The control unit outputs repair information relating to repair of the abnormality at the abnormal location.

[0014] The invention described in claim 5 is the information processing device described in claim 1, the estimation data is data for estimating noise corresponding to an installation abnormality of a part that is outside an allowable range, The control unit uses the estimation data to estimate noise based on an installation abnormality of a part that is outside a tolerance range of the ultrasonic diagnostic apparatus, from the acquired ultrasonic image data.

[0015] The invention described in claim 6 is the information processing device described in claim 1, the acquisition unit acquires ultrasound image data associated with at least one setting information of an image mode and a setting parameter; The estimation data is prepared for each piece of setting information, The control unit uses estimation data corresponding to setting information of the acquired ultrasound image data to determine, from the ultrasound image data, an installation abnormality of a part of the ultrasound diagnostic device corresponding to the noise.

[0016] The invention described in claim 7 is the information processing device described in claim 1, The component mounting abnormality is an abnormality other than that of an analog component.

[0017] The invention described in claim 8 is the information processing device according to any one of claims 1 to 7, the information processing device is the ultrasound diagnostic device, The acquisition unit generates and acquires the ultrasound image data.

[0018] The information processing device of the invention described in claim 9 comprises: The system includes a control unit that performs machine learning on ultrasound image data with or without noise in the ultrasound image due to an abnormality in the installation of the ultrasound diagnostic device's components, and generates trained estimation data for estimating noise in the ultrasound image due to an abnormality in the installation of the ultrasound diagnostic device's components.

[0019] The abnormality determination method of the invention described in claim 10 comprises: an acquisition step of acquiring ultrasound image data; and a control step of estimating noise from the acquired ultrasound image data using trained estimation data for estimating noise in an ultrasound image due to an installation abnormality of a part of the ultrasound diagnostic device, determining an installation abnormality of a part of the ultrasound diagnostic device corresponding to the noise, and outputting information on the determination result.

[0020] The learning method of the invention described in claim 11 comprises: The system includes a control unit that performs machine learning on ultrasound image data with or without noise in the ultrasound image due to an abnormality in the installation of the ultrasound diagnostic device's components, and generates trained estimation data for estimating noise in the ultrasound image due to an abnormality in the installation of the ultrasound diagnostic device's components.

[0021] The program of the invention described in claim 12 is Computer, an acquisition unit that acquires ultrasound image data; a control unit that estimates noise from the acquired ultrasound image data using trained estimation data for estimating noise in the ultrasound image due to an installation abnormality of a part of the ultrasound diagnostic device, determines an installation abnormality of the part of the ultrasound diagnostic device corresponding to the noise, and outputs information on the determination result; Function as.

[0022] The invention described in claim 13 is a program comprising: Computer, a control unit that performs machine learning on ultrasound image data of whether or not noise in the ultrasound image is due to an abnormality in the installation of a component of the ultrasound diagnostic device, and generates trained estimation data for estimating noise in the ultrasound image due to an abnormality in the installation of the component of the ultrasound diagnostic device; Function as. [Effects of the Invention]

[0023] According to the present invention, an abnormality in the installation of a part can be easily detected. [Brief explanation of the drawings]

[0024] [Figure 1] 1 is a schematic diagram of an ultrasonic diagnostic apparatus according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram showing the functional configuration of the ultrasound diagnostic apparatus. [Figure 3] 1 is a schematic diagram showing the board configuration and noise paths of the ultrasonic diagnostic device main body. FIG. [Figure 4] 1 is a schematic diagram showing the board configuration and noise paths of the ultrasonic diagnostic device main body. FIG. [Figure 5] FIG. 2 is a schematic diagram showing the configuration of a board and a power supply of the ultrasound diagnostic device main body. [Figure 6] 10 is a flowchart showing a first learning process. [Figure 7] 10 is a flowchart showing a first abnormality determination process. [Figure 8] 10 is a flowchart showing a second learning process. [Figure 9] 10 is a flowchart showing a second abnormality determination process. [Figure 10] 10 is a flowchart showing a third learning process. [Figure 11] 10 is a flowchart showing a third abnormality determination process. DETAILED DESCRIPTION OF THE INVENTION

[0025] Hereinafter, first to fourth embodiments of the present invention will be described in detail in order with reference to the drawings, however, the scope of the invention is not limited to the illustrated examples.

[0026] (First embodiment) A first embodiment of the present invention will be described with reference to Figures 1 to 7. First, the device configuration of this embodiment will be described with reference to Figures 1 and 2. Figure 1 is a schematic diagram of an ultrasound diagnostic device 100 of this embodiment. Figure 2 is a block diagram showing the functional configuration of the ultrasound diagnostic device 100.

[0027] 1, an ultrasound diagnostic device 100 is installed in a medical facility such as a hospital, and emits ultrasound waves to a subject such as a living patient to generate ultrasound image data. The ultrasound diagnostic device 100 also uses a machine learning trained model to identify abnormalities in the device itself from the ultrasound image data.

[0028] The ultrasound diagnostic device 100 includes an ultrasound diagnostic device main body 1 and an ultrasound probe 2. The ultrasound probe 2 is connected to the ultrasound diagnostic device main body 1. The ultrasound probe 2 transmits ultrasound waves (transmitted ultrasound waves) into a subject and receives reflected waves of the ultrasound waves reflected within the subject (reflected ultrasound waves: echoes). The ultrasound probe 2 has an ultrasound probe main body 21, a cable 22, and a connector 23. The ultrasound probe main body 21 is the head of the ultrasound probe 2 and transmits and receives ultrasound waves. The cable 22 is connected to the ultrasound probe main body 21 and the connector 23. The cable 22 is a cable through which a drive signal for the ultrasound probe main body 21 and a received ultrasound signal flow. The connector 23 is a plug connector for connecting to a receptacle connector (not shown) of the ultrasound diagnostic device main body 1.

[0029] The ultrasound diagnostic device main body 1 is connected to the ultrasound probe main body 21 via a connector 23 and a cable 22. The ultrasound diagnostic device main body 1 transmits an electrical drive signal to the ultrasound probe main body 21, causing the ultrasound probe main body 21 to transmit ultrasound waves to the subject. The ultrasound probe 2 generates a reception signal, which is an electrical signal, in response to the ultrasound reflected from inside the subject and received by the ultrasound probe main body 21. The ultrasound diagnostic device main body 1 creates an image of the internal state of the subject as ultrasound image data based on the reception signal generated by the ultrasound probe 2.

[0030] The ultrasound probe main body 21 has transducers (not shown) at the tip side. The number of transducers can be set arbitrarily, and in practice, it is, for example, 192. The transducers are arranged, for example, in a one-dimensional array in the scanning direction (azimuth direction). The transducers may also be arranged in a two-dimensional array. In this embodiment, a linear scanning electronic scanning probe is adopted as the ultrasound probe 2. However, the ultrasound probe 2 may be either an electronic scanning type or a mechanical scanning type. The ultrasound probe 2 may also be any of a linear scanning type, a sector scanning type, or a convex scanning type. The ultrasound diagnostic device main body 1 and the ultrasound probe 2 may be configured to communicate wirelessly instead of by wire via a cable 22. This wireless communication may be UWB (Ultra Wide Band) or the like.

[0031] The operation input unit 11 is a control panel or the like that accepts various operation inputs from users such as doctors, technicians, etc. The operation input unit 11 has operation elements such as push buttons, encoders, lever switches, joysticks, trackballs, keyboards, touchpads, and multifunction switches.

[0032] The display unit 17 has a display panel such as an LCD (Liquid Crystal Display), an organic EL (Electro-Luminescence) display, an inorganic EL display, etc. The display unit 17 displays display information such as ultrasound image data on the display panel.

[0033] As shown in FIG. 2, the ultrasound diagnostic device main body 1 includes an operation input unit 11, a transmission unit 12, a receiving amplifier 131, an AD (Analog to Digital) converter 132, a beam former 14, a signal processing unit 15, a display unit 17, a control unit 18, a storage unit 19, and a communication unit 16. The receiving amplifier 131, the AD converter 132, and the beam former 14 function as a receiving unit. The analog block 30 is a block of circuit elements in the ultrasound diagnostic device 100 that transmits and receives ultrasound to and from a subject and performs processing using analog signals. The analog block 30 includes the transmission unit 12, the ultrasound probe 2, the receiving amplifier 131, and the AD converter 132. The digital block 40 is a block of circuit elements in the ultrasound diagnostic device 100 (ultrasound diagnostic device main body 1) that performs processing using digital signals. The digital block 40 includes the beam former 14, the signal processing unit 15, and the control unit 18.

[0034] The operation input unit 11 accepts various operation inputs from the user and outputs the operation signals to the control unit 18. The operation input unit 11 may be formed integrally with the display screen of the display unit 17 and may include a touch panel that accepts touch inputs from the user.

[0035] The transmitter 12, under the control of the controller 18, supplies a drive signal, which is an electrical signal, to the ultrasonic probe 2, causing the ultrasonic probe 2 to generate a transmission ultrasonic wave. The transmitter 12 includes, for example, a clock generating circuit, a delay circuit, and a pulse generating circuit. The clock generating circuit generates a clock signal that determines the transmission timing and transmission frequency of the drive signal. The delay circuit sets a delay time for each individual path corresponding to each transducer, and delays the transmission of the drive signal by the set delay time. The delay circuit focuses a transmission beam formed by the transmission ultrasonic wave using the delay. The pulse generating circuit generates a pulse signal as a drive signal at a predetermined period. The transmitter 12 generates a transmission ultrasonic wave by, for example, driving a continuous portion (e.g., 64 transducers) of multiple transducers (e.g., 192 transducers) arranged in the ultrasonic probe 2. The transmitter 12 then performs scanning by shifting the driven transducers in the scanning direction each time a transmission ultrasonic wave is generated.

[0036] The receiving amplifier 131 receives a received signal, which is an analog electrical signal received from the ultrasonic probe 2. The receiving amplifier 131 amplifies the signal at an arbitrary gain value (amplification factor) under the control of the control unit 18. The AD converter 132 converts the analog received signal amplified by the receiving amplifier 131 into a digital received signal.

[0037] The beam former 14 adjusts the time phase by providing a delay time for each individual path corresponding to each transducer to the digital received signals that have been AD converted by the AD converter 132. The beam former 14 adds (phased addition) these processed received signals to generate sound ray data.

[0038] The signal processing unit 15 performs envelope detection processing, logarithmic compression, and the like on the sound ray data from the beam former 14 under the control of the control unit 18. The signal processing unit 15 further performs brightness conversion on the sound ray data after these processing by adjusting the dynamic range and gain. Through this brightness conversion, the signal processing unit 15 generates B (brightness) mode image data made up of pixels having brightness values ​​as received energy. In other words, B mode image data represents the strength of received signals by brightness. Furthermore, the signal processing unit 15 may be configured to generate image data for other image modes, such as color Doppler mode and pulse Doppler mode, in addition to B mode.

[0039] The signal processing unit 15 also has an image memory unit (not shown). The image memory unit is configured with, for example, a semiconductor memory such as a DRAM (Dynamic Random Access Memory). The signal processing unit 15 stores B-mode image data in the image memory unit on a frame-by-frame basis under the control of the control unit 18. The signal processing unit 15 reads out the ultrasound image data stored in the image memory unit one frame at a time at predetermined time intervals.

[0040] The signal processing unit 15 has, for example, a DSC (Digital Scan Converter). The signal processing unit 15 performs processing such as coordinate conversion on the ultrasound image data (B-mode image data) under the control of the control unit 18 to convert the data into an image signal for display. The signal processing unit 15 outputs the image signal to the display unit 17.

[0041] The display unit 17 displays an ultrasound image on a display panel in accordance with the image signal output from the signal processing unit 15 under the control of the control unit 18. The display unit 17 also displays various display information input from the control unit 18 on the display panel.

[0042] The control unit 18 includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The control unit 18 reads various processing programs stored in the ROM, loads them into the RAM, and controls each unit of the ultrasound diagnostic apparatus 100 in cooperation with the CPU. The ROM is composed of non-volatile memory such as a semiconductor. The ROM stores a system program corresponding to the ultrasound diagnostic apparatus 100, various processing programs executable on the system program, and various data such as a gamma table. In particular, the ROM stores a first learning program for executing a first learning process (described below) and a first anomaly discrimination program for executing a first anomaly discrimination process (described below). These programs are stored in the RAM in the form of computer-readable program codes. The CPU sequentially executes operations in accordance with the program codes in the RAM. The RAM forms a work area for temporarily storing various programs executed by the CPU and data related to these programs.

[0043] The storage unit 19 is a storage unit such as an HDD (Hard Disk Drive) or SSD (Solid State Drive) that stores information such as ultrasound image data in a writable and readable manner. In particular, the storage unit 19 stores estimation data as a trained model of machine learning. The estimation data is data for estimating image noise based on an installation abnormality of a component of the ultrasound diagnostic device 100 from the ultrasound image data.

[0044] The communication unit 16 is, for example, a network card connected to a communication network such as a LAN (Local Area Network). For example, the control unit 18 transmits and receives information to and from external devices on the communication network via the communication unit 16.

[0045] The analog block 30 uses a highly sensitive receiving amplifier 131 to handle weak signals from the subject's living body. On the other hand, the digital block 40 performs high-speed calculations using digital signals of several volts, which can become a source of (electrical) noise. In the digital block 40, noise generation is minimized as much as possible, and measures are usually taken to prevent generated noise from leaking into the analog block 30.

[0046] An example of noise generation in the ultrasonic diagnostic device 100 will be described with reference to Figures 3 to 5. Figure 3 is a schematic diagram showing the board configuration and noise paths of the ultrasonic diagnostic device main body 1. Figure 4 is a schematic diagram showing the board configuration and noise paths of the ultrasonic diagnostic device main body 1. Figure 5 is a schematic diagram showing the board and power supply configuration of the ultrasonic diagnostic device main body 1.

[0047] 3 and 4, the ultrasound diagnostic device main body 1 includes an analog board 31, a stack 32, a digital board 41, a stack 42, a frame 51, and a signal line 52. The analog board 31 is a board on which circuit elements of the analog block 30 are mounted. The digital board 41 is a board on which circuit elements of the digital block 40 are mounted. The stack 32 is a metal fixture for attaching the analog board 31 to the frame 51. The stack 42 is a metal fixture for attaching the digital board 41 to the frame 51. The frame 51 is a metal frame of the ultrasound diagnostic device main body 1. The signal line 52 is a signal line that electrically connects a terminal of the conductive pattern of the analog board 31 to a terminal of the conductive pattern of the digital board 41.

[0048] As shown in Fig. 3, analog board 31 is attached to frame 51 via stack 32. Digital board 41 is attached to frame 51 via stack 42. Furthermore, as shown by the noise path indicated by the arrow in Fig. 3, electrical noise generated on digital board 41 is prevented from reaching analog board 31.

[0049] However, consider a structure in which the digital board 41 is attached to the frame 51 with screws using the stack 42, as shown in Figure 4. In this structure, the screws may become loose due to vibrations during transportation of the ultrasound diagnostic device 100, causing the device to float. In this case, the path for electrical noise that was conducted from the digital board 41 to the frame 51 disappears. Therefore, as shown by the noise path indicated by the arrow in Figure 4, electrical noise travels along the signal line 52 and finds its way into the analog board 31, causing image noise in the ultrasound image on the display screen.

[0050] As shown in FIG. 5 , in terms of the power supply configuration, the ultrasound diagnostic apparatus main body 1 further includes a main power supply 61 and power supply units 62 and 63. The main power supply 61 is a main power supply mounted on the frame 51. The power supply unit 62 is a power supply mounted on the analog board 31. The power supply unit 62 converts the voltage of the power supply supplied from the main power supply 61 and supplies it to the circuit elements of the analog board 31. The power supply unit 63 is a power supply mounted on the digital board 41. The power supply unit 63 converts the voltage of the power supply supplied from the main power supply 61 and supplies it to the circuit elements of the digital board 41.

[0051] As described above, the analog board 31 and the digital board 41 each have their own power supply units 62, 63 to supply stable voltage to the circuit elements. The power supply units 62, 63 are each supplied with power from the main power supply 61. The analog board 31 requires circuit elements with low noise. For this reason, a series power supply or the like is used for the power supply unit 62. The digital board 41 is supplied with a large current. For this reason, a DC (Direct Current)-DC converter such as a switching regulator is often used for the power supply unit 63.

[0052] The DC-DC converter is equipped with an oscillator, which generates switching noise due to the principle of on-off switching of current. The analog board 31 is designed to prevent switching noise from leaking in. However, changes / deterioration over time in the power supply unit 63 can cause abnormalities, such as a change in the switching frequency, and the switching noise can be transmitted to the analog board 31. In this case, the switching noise can affect the generation of image noise in the ultrasound image.

[0053] Components that correspond to component installation abnormalities in the ultrasonic diagnostic device 100 are not limited to the stack 42 and the power supply unit 62. Component installation abnormalities include abnormalities that occur due to changes over time / deterioration over time even during normal use as described above, as well as abnormalities other than component failure, such as improper assembly of the ultrasonic diagnostic device during manufacturing. As shown in Figures 3 to 5, component installation abnormalities are assumed to be installation abnormalities in components other than the components (analog components) on the analog board 31.

[0054] Next, the operation of the ultrasound diagnostic apparatus 100 of this embodiment will be described with reference to Figures 6 and 7. Figure 6 is a flowchart showing the first learning process. Figure 7 is a flowchart showing the first abnormality discrimination process.

[0055] The first learning process executed by the ultrasound diagnostic device 100 will be described with reference to Figure 6. The first learning process is a machine learning process that acquires ultrasound image data with and without image noise due to component installation abnormalities as training data. Here, we will describe a case where noise occurs in the development process or manufacturing process, or where noise is generated by intentionally creating an installation abnormality.

[0056] Assume beforehand that the ultrasound diagnostic device 100 is in a state where there are no abnormalities in the installation of parts. In the ultrasound diagnostic device 100, for example, an instruction to execute a first learning process is input from an operator via the operation input unit 11. In response to the execution instruction, the control unit 18 executes the first learning process in accordance with the first learning program stored in the ROM. The operator is the developer or manufacturer of the ultrasound diagnostic device 100.

[0057] First, the control unit 18 acquires ultrasound image data without image noise due to an installation error of a part (hereinafter simply referred to as "image noise") and stores it in the storage unit 19 (step S11). In step S11, the control unit 18 controls the transmission unit 12 to the signal processing unit 15 to scan the subject and generate and acquire ultrasound image data without image noise.

[0058] The control unit 18 acquires ultrasound image data with image noise and stores it in the storage unit 19 (step S12). Here, the presence of image noise can occur when an operator generates image noise due to a defect in the assembly of the ultrasound diagnostic device or intentionally causes a defect in the installation of a component. The above is a learning method for determining whether image noise is present or absent during the development and manufacturing processes. However, it is also possible to adapt to cases where image noise occurs in the market, for example. In this case, the ultrasound diagnostic device 100 is communicatively connected to an image server via a communication network. The image server receives and stores ultrasound image data with image noise generated based on a component installation abnormality from an ultrasound diagnostic device in which a component installation abnormality has occurred. This ultrasound diagnostic device is preferably the same model as the ultrasound diagnostic device 100. In step S12, the control unit 18 requests and receives ultrasound image data with image noise from the image server via the communication unit 16. The image server may also be configured to store ultrasound image data with and without image noise. In this configuration, in step S11, the control unit 18 requests the image server via the communication unit 16 to receive and acquire ultrasound image data without image noise.

[0059] The control unit 18 determines whether the number of accumulated data of ultrasound image data with / without image noise stored in the storage unit 19 is equal to or greater than a predetermined number (step S13). The predetermined number in step S13 is a number of accumulated data that is sufficient for machine learning of ultrasound image data with / without image noise. The machine learning, for example, estimates the boundary of the feature amount using the ultrasound image data with / without image noise accumulated in the storage unit 19 as training data. Furthermore, the machine learning generates, as estimation data, a trained model for estimating the presence / absence of image noise based on an installation abnormality of a part in the ultrasound image data using the boundary.

[0060] If the number is less than the predetermined number (step S13; NO), the process proceeds to step S11. If the number is equal to or greater than the predetermined number (step S13; YES), the control unit 18 performs machine learning using ultrasound image data with / without image noise in the storage unit 19 (step S14). The control unit 18 extracts estimation data from the learning results of the machine learning in step S14 (step S15). The estimation data in step S15 is a trained model that estimates the presence or absence of image noise based on component installation abnormalities from the input ultrasound image data. The control unit 18 stores the estimation data extracted in step S15 in the storage unit 19 (step S16). The first learning process ends.

[0061] 7, the first abnormality discrimination process executed by the ultrasound diagnostic apparatus 100 will be described. The first abnormality discrimination process is a process of scanning a subject to acquire ultrasound image data, discriminating component installation abnormalities from the ultrasound image data, and displaying the result information.

[0062] After the first learning process, the ultrasound diagnostic device 100 is delivered to a medical facility and becomes available for use by users such as doctors and technicians. After the first learning process, an instruction to execute a first abnormality discrimination process is input to the ultrasound diagnostic device 100 from the user via, for example, the operation input unit 11. In response to the execution instruction, the control unit 18 executes the first abnormality discrimination process in accordance with the first abnormality discrimination program stored in the ROM.

[0063] First, the control unit 18 generates and acquires ultrasound image data of the subject under the control of the transmission unit 12 to the signal processing unit 15 in response to setting information input to the operation input unit 11 (step S21). The setting information is setting conditions for generating ultrasound image data, such as an image mode and setting parameters. The control unit 18 reads out the learned estimation data from the storage unit 19 (step S22).

[0064] Using the estimation data from step S22, control unit 18 estimates the presence or absence of image noise due to component installation abnormality in the ultrasound image data from step S21 (step S23). In step S23, control unit 18 determines the presence or absence of component installation abnormality based on the presence or absence of the image noise. Control unit 18 generates result information indicating whether or not there is a component installation abnormality from the determination result of step S23, and displays the result information on display unit 17 (step S24). The first abnormality determination process then ends.

[0065] As described above, according to this embodiment, the ultrasound diagnostic device 100 as an information processing device includes the transmitter 12 to the signal processor 15 (and the ultrasound probe 2) as an acquisition unit, and the controller 18. The transmitter 12 to the signal processor 15 generate and acquire ultrasound image data. The controller 18 estimates image noise from the acquired ultrasound image data using estimation data for estimating noise in the ultrasound image based on an installation abnormality of a component of the device itself. The controller 18 determines whether there is an installation abnormality of the component corresponding to the noise of the ultrasound diagnostic device 100. The controller 18 displays information on the determination result on the display 17.

[0066] The control unit 18 performs machine learning of ultrasound image data on whether or not noise is present in the ultrasound image based on an installation abnormality of the parts of the ultrasound diagnostic device 100. The control unit 18 generates trained estimation data for estimating image noise in the ultrasound image based on an installation abnormality of the parts of the ultrasound diagnostic device 100.

[0067] Therefore, it is possible to easily detect any installation abnormalities in the parts of the ultrasonic diagnostic device 100, and the resulting information can be presented to users, service personnel, etc. Furthermore, the ultrasonic diagnostic device 100 itself can perform abnormality detection that can only be determined visually by an experienced person, thereby reducing the number of personnel required.

[0068] The component installation abnormality is an abnormality caused by an assembly error during the manufacture of the ultrasound diagnostic device 100, or an abnormality caused by a problem with the component installation that occurs over time even when the device is used normally. Therefore, it is possible to determine component installation abnormalities that are not appropriate for ultrasound image diagnosis, such as a screw being left untightened or a cable being in contact with an unintended location, even if the component is not faulty.

[0069] An abnormality in the installation of a component is an abnormality other than that of an analog component. Here, an analog component refers to a component that constitutes a block (so-called analog front end) that receives signals from the transducer of the ultrasonic probe 2. Therefore, an abnormality in the installation of a component other than an analog component can be detected. For example, an abnormality in the operating state of a power supply component can be detected.

[0070] (Second embodiment) A second embodiment of the present invention will be described with reference to Figures 8 and 9. Figure 8 is a flowchart showing a second learning process. Figure 9 is a flowchart showing a second abnormality determination process.

[0071] The first embodiment described above is configured to determine whether or not there is an installation abnormality in the components of the ultrasound diagnostic device 100. The present embodiment is configured to determine whether or not there is an installation abnormality in the components, and also to estimate the location of the abnormality if there is an abnormality.

[0072] In this embodiment, the apparatus configuration is assumed to be an ultrasound diagnostic apparatus 100. However, it is assumed that the ROM of the control unit 18 stores a second learning program for executing a second learning process (to be described later) and a second abnormality discrimination program for executing a second abnormality discrimination process (to be described later).

[0073] Next, the operation of the ultrasound diagnostic device 100 of this embodiment will be described with reference to Figures 8 and 9. The second learning process executed by the ultrasound diagnostic device 100 will be described with reference to Figure 8.

[0074] In this embodiment, ultrasound image data with image noise based on abnormal component installation for learning is stored in the image server in association with setting information and abnormal locations. Setting information refers to the image mode and setting parameters used when generating the ultrasound image data. Image modes include B-mode, color Doppler mode, pulse Doppler mode, etc. Setting parameters include the frequency used for transmission and reception, gain settings, and the scale of the image to be displayed. The abnormal locations are information about the locations where abnormal component installation occurred in the ultrasound diagnostic device 100.

[0075] In the ultrasound diagnostic device 100, for example, an instruction to execute the second learning process is input by the operator via the operation input unit 11. In response to the execution instruction, the control unit 18 executes the second learning process in accordance with the second learning program stored in the ROM.

[0076] First, the control unit 18 receives, from the operator via the operation input unit 11, input of setting information corresponding to the estimation data to be generated and a designation of an abnormality location (step S31). The control unit 18 acquires ultrasound image data without image noise based on an abnormality in the component installation, corresponding to the setting information input in step S11, and stores the acquired data in the storage unit 19 (step S32). In step S32, the control unit 18 scans the subject according to the setting information under the control of the transmission unit 12 to the signal processing unit 15. Through the scan, the control unit 18 generates and acquires ultrasound image data without image noise, corresponding to the setting information.

[0077] The control unit 18 acquires ultrasound image data with image noise corresponding to the setting information and abnormality location input in step S31 and stores the acquired data in the storage unit 19 (step S33). Regarding the state with image noise due to an abnormality in an abnormality location, the operator may generate image noise due to a defect in the abnormality location during assembly of the ultrasound diagnostic device, or may intentionally generate image noise by causing a defect in the installation of a component at the abnormality location. The image server receives and stores ultrasound image data with image noise based on the abnormality in the abnormality location generated in this state from the ultrasound diagnostic device 100 in which an abnormality in the component installation has occurred. In this case, in step S33, the control unit 18 requests, receives, and acquires ultrasound image data with image noise corresponding to the input setting information and abnormality location from the image server via the communication unit 16. The image server may also be configured to store ultrasound image data with / without image noise corresponding to the setting information and abnormality location. In this configuration, in step S32, the control unit 18 requests, via the communication unit 16, the image server to receive and acquire ultrasound image data without image noise that corresponds to the input setting information.

[0078] In the market, image noise may occur due to an abnormality other than those assumed above. In this case, the image noise is recorded, and ultrasound image data with the image noise is acquired by an image server via a communication network, or acquired during repair of the ultrasound diagnostic device 100. At the same time, the abnormality is identified by a service person, and the data is used as ultrasound image data specifying the abnormality.

[0079] Step S34 is the same as step S13 in the first learning process of Fig. 6. If the number is equal to or greater than the predetermined number (step S34; YES), the control unit 18 receives input from the operator via the operation input unit 11 as to whether or not all of the setting information and abnormal locations have been designated. In response to the input, the control unit 18 determines whether or not all of the setting information and abnormal locations have been designated (step S35). If they have not been designated (step S34; NO), the process proceeds to step S31.

[0080] If it has been specified (step S34; YES), the process proceeds to step S36. The control unit 18 performs machine learning using the setting information and ultrasound image data with / without image noise for each abnormality location in the storage unit 19 (step S36). The control unit 18 extracts estimation data for each setting information and abnormality location from the learning result of the machine learning in step S36 (step S37). The estimation data in step S37 is a learned model that estimates the presence or absence of image noise based on each setting information and an installation abnormality of a component at each abnormality location from the input ultrasound image data. Furthermore, the estimation data in step S37 is a learned model that estimates an abnormality location corresponding to an installation abnormality of a component with image noise when it is estimated that image noise is present. The control unit 18 stores the setting information and estimation data for each abnormality location extracted in step S37 in the storage unit 19 (step S38). The second learning process ends.

[0081] 9, the second abnormality discrimination process executed by the ultrasound diagnostic apparatus 100 will be described. The second abnormality discrimination process is a process of scanning the subject to acquire ultrasound image data, discriminating component installation abnormalities from the ultrasound image data, estimating the location of the abnormality, and displaying the resulting information.

[0082] After the second learning process, the ultrasound diagnostic device 100 is delivered to a medical facility and becomes available for use by the user. In the ultrasound diagnostic device 100, for example, a command to execute the second abnormality discrimination process is input from the user via the operation input unit 11. In response to the command, the control unit 18 executes the second abnormality discrimination process in accordance with the second abnormality discrimination program stored in the ROM.

[0083] First, the control unit 18 receives input of setting information from the user via the operation input unit 11 (step S41). In step S41, the control unit 18 generates and acquires ultrasound image data of the subject corresponding to the input setting information under the control of the transmission unit 12 to the signal processing unit 15. The control unit 18 reads out, from the storage unit 19, the learned estimation data of each abnormality location corresponding to the setting information of step S41 (step S42).

[0084] The control unit 18 uses the estimation data of step S42 to estimate the presence or absence of image noise in the ultrasound image data of step S41, based on a component installation abnormality (step S43). In step S43, the control unit 18 determines the presence or absence of a component installation abnormality based on the presence or absence of the image noise. In step S43, the control unit 18 uses the estimation data of step S42 to estimate an abnormality location corresponding to the abnormality, if any, of the component installation abnormality. The control unit 18 generates result information on the presence or absence of a component installation abnormality and the abnormal location from the determination and estimation results of step S43, and displays the result information on the display unit 17 (step S44). The second abnormality determination process then ends.

[0085] As described above, according to this embodiment, the estimation data is data for estimating the location of the component installation abnormality corresponding to the estimated image noise. Using the estimation data, the control unit 18 estimates the location of the component installation abnormality in the ultrasound diagnostic device 100 from the acquired ultrasound image data, and displays the result information including the location of the abnormality on the display unit 17. This allows the user, service personnel, etc. to easily recognize the location of the component installation abnormality, making it easy to deal with the abnormality.

[0086] The transmitter 12 to the signal processor 15 (and the ultrasound probe 2) generate and acquire ultrasound image data associated with the image mode and the setting information of the setting parameters. Estimation data is prepared for each setting information. The controller 18 uses the estimation data corresponding to the setting information of the acquired ultrasound image data to determine, from the ultrasound image data, a component installation abnormality corresponding to the image noise of the ultrasound diagnostic device 100. This allows the use of estimation data appropriately prepared for each setting information, allowing the image noise corresponding to the setting information of the ultrasound image data to be accurately estimated. This improves the accuracy of determining image noise due to component installation abnormalities. Furthermore, by specifying a specific image mode (B mode, color Doppler mode, pulse Doppler mode, etc.) that makes it easy to estimate image noise, the accuracy of determining component installation abnormalities can be further improved.

[0087] In this embodiment, a trained model for estimating setting information of input ultrasound image data may be further used. In step S42, estimation data corresponding to the setting information estimated using the trained model for setting information in this embodiment is read from the storage unit 19.

[0088] (Third embodiment) A third embodiment of the present invention will be described with reference to Fig. 10. Fig. 10 is a flowchart showing a third learning process.

[0089] The first embodiment described above is configured to determine whether or not there is a component installation abnormality based on image noise in the ultrasound image data of the ultrasound diagnostic device 100. However, even when there is image noise, there may be cases where a component installation abnormality occurs within the tolerance range for diagnosis. This embodiment is configured to determine whether or not there is a component installation abnormality when there is image noise in the ultrasound image data that causes a component installation abnormality outside the tolerance range.

[0090] In this embodiment, the apparatus configuration is assumed to be an ultrasound diagnostic apparatus 100. However, it is assumed that the ROM of the control unit 18 stores a third learning program for executing a third learning process, which will be described later, and a first abnormality discrimination program.

[0091] Next, the operation of the ultrasound diagnostic device 100 of this embodiment will be described with reference to Fig. 10. The third learning process executed by the ultrasound diagnostic device 100 will be described with reference to Fig. 10.

[0092] In this embodiment, it is assumed that ultrasound image data with image noise for learning is assigned an abnormality presence / absence label and stored in the image server. The abnormality presence / absence label is a label indicating whether or not there is an installation abnormality of a part outside the allowable range of image noise in the ultrasound image data to which the abnormality presence / absence label is assigned. For this reason, even in the case of ultrasound image data with image noise, an abnormality presence / absence label indicating that there is no installation abnormality of a part outside the allowable range may be assigned.

[0093] In the ultrasound diagnostic apparatus 100, for example, an instruction to execute the third learning process is input by the operator via the operation input unit 11. In response to the execution instruction, the control unit 18 executes the third learning process in accordance with the third learning program stored in the ROM.

[0094] First, the control unit 18 acquires ultrasound image data without image noise based on an installation abnormality of a part and stores it in the storage unit 19 (step S51). In step S51, the control unit 18 scans the subject according to the setting information under the control of the transmission unit 12 to the signal processing unit 15. Through the scan, the control unit 18 generates and acquires ultrasound image data without image noise that corresponds to the setting information.

[0095] The control unit 18 acquires ultrasound image data with image noise and stores it in the storage unit 19 (step S52). Regarding the state where image noise is within / outside the tolerance for diagnosis, the operator may generate image noise due to a defect in the assembly of the ultrasound diagnostic device or may intentionally generate image noise by causing a defect in the installation of a component. The operator determines whether the image noise is within or outside the tolerance and assigns an abnormality presence / absence flag representing the determination result to the ultrasound image data containing image noise. The image server receives and stores ultrasound image data with image noise and an abnormality presence / absence flag based on the abnormality generated in this state from the ultrasound diagnostic device in which the component installation abnormality occurred. In step S52, the control unit 18 requests, receives, and acquires ultrasound image data with image noise and an abnormality presence / absence flag from the image server via the communication unit 16. The image server may also be configured to store ultrasound image data with / without image noise. In this configuration, in step S52, the control unit 18 requests, receives, and acquires ultrasound image data without image noise from the image server via the communication unit 16.

[0096] In the field, image noise may occur due to abnormalities outside / within the tolerance range other than those assumed above, and a response may be required. In this case, the image noise is recorded, and ultrasound image data with the image noise is acquired by an image server via a communication network, or acquired when repairing the ultrasound diagnostic device 100. At the same time, a service technician assigns an abnormality presence / absence flag to the ultrasound image data indicating whether the part installation abnormality is within / outside the tolerance range. Furthermore, the abnormality may be identified and used as ultrasound image data specifying the abnormality.

[0097] Step S53 is the same as step S13 in the first learning process of Fig. 6. If the number is equal to or greater than the predetermined number (step S53; YES), the control unit 18 performs machine learning using ultrasound image data with / without image noise and labeled with anomaly presence / absence in the storage unit 19 (step S54). The control unit 18 extracts estimation data from the learning results of the machine learning in step S54 (step S55). The estimation data in step S55 is a trained model that estimates the presence or absence of image noise based on an attachment abnormality of a component outside the tolerance range from the input ultrasound image data. The control unit 18 stores the estimation data extracted in step S55 in the storage unit 19 (step S56). The third learning process ends.

[0098] After the third learning process, the ultrasound diagnostic device 100 is delivered to a medical facility and becomes available for use by a user. In the ultrasound diagnostic device 100, the first anomaly discrimination process is executed, as in the first embodiment. In step S23, image noise due to component installation abnormalities within the allowable range is estimated to be no image noise.

[0099] As described above, according to this embodiment, the estimation data is data for estimating image noise corresponding to an abnormality in a component installation outside the tolerance range, which corresponds to an abnormality presence / absence label indicating the presence of an abnormality. The control unit 18 uses the estimation data to estimate image noise based on an abnormality in a component installation outside the tolerance range of the ultrasound diagnostic device 100 from the acquired ultrasound image data. This makes it possible to prevent a component installation abnormality that is within the tolerance range for diagnosis despite the presence of image noise from being determined as an abnormality. Note that the person setting the ultrasound image data for learning can arbitrarily set the tolerance range using the abnormality presence / absence label.

[0100] (Fourth embodiment) A fourth embodiment of the present invention will be described with reference to Fig. 11. Fig. 11 is a flowchart showing a third abnormality determination process.

[0101] The second embodiment described above is configured to determine whether or not there is an installation abnormality in the parts of the ultrasound diagnostic device 100 and to display the result information on the abnormal part. The present embodiment is configured to display, if there is an installation abnormality in the parts of the ultrasound diagnostic device 100, how to repair the abnormal part.

[0102] In this embodiment, the apparatus configuration is assumed to be an ultrasound diagnostic apparatus 100. However, the ROM of the control unit 18 is assumed to store a second learning program and a fourth abnormality discrimination program for executing a fourth abnormality discrimination process described later.

[0103] The storage unit 19 also stores repair information associated with abnormal locations due to installation errors of components of the ultrasound diagnostic device 100. The repair information is information related to repairing each abnormal location, and includes repair methods, information on tools required for repair, and information identifying the abnormality for contacting a support center.

[0104] Next, the operation of the ultrasound diagnostic device 100 of this embodiment will be described with reference to Fig. 11. In the ultrasound diagnostic device 100, the second learning process is executed in the same manner as in the second embodiment.

[0105] 11, the third abnormality discrimination process executed by the ultrasound diagnostic apparatus 100 will be described. The third abnormality discrimination process is a process of scanning a subject to acquire ultrasound image data, discriminating component installation abnormalities from the ultrasound image data, and displaying repair information and result information.

[0106] After the second learning process, the ultrasound diagnostic device 100 is delivered to a medical facility and becomes available for use by the user. In the ultrasound diagnostic device 100, for example, a command to execute a third abnormality discrimination process is input from the user via the operation input unit 11. In response to the command, the control unit 18 executes the third abnormality discrimination process in accordance with the third abnormality discrimination program stored in the ROM.

[0107] Steps S71 to S73 are the same as steps S41 to S43 of the second abnormality determination process in Fig. 9. The control unit 18 reads out repair information corresponding to the abnormality location estimated in step S73 from the storage unit 19 (step S74). The control unit 18 executes step S75. In step S75, result information on the presence or absence of a component installation abnormality and the abnormal location is generated from the determination and estimation results of step S73. In step S75, the control unit 18 displays the generated result information and the repair information read out in step S74 on the display unit 17. The third abnormality determination process then ends.

[0108] As described above, according to this embodiment, control unit 18 displays repair information related to repairing an abnormality in an abnormal part of a part installation on display unit 17. This allows a user or a service technician to easily check the repair information for the abnormal part, and facilitates repairing the abnormal part.

[0109] The description of the above embodiments is merely an example of the ultrasound diagnostic device, information processing device, ultrasound image generating method, ultrasound image learning method, and program according to the present invention, and is not limited to this. For example, at least two of the above embodiments may be appropriately combined.

[0110] In each of the above embodiments, trained estimation data is generated by performing machine learning on ultrasound image data as image data. However, the present invention is not limited to this configuration. Sound ray data or intermediate data generated between sound ray data generation and image data generation may be used as image data.

[0111] In addition, in each of the above embodiments, the ultrasound diagnostic device 100 as an information processing device is configured to perform machine learning using ultrasound image data generated by the ultrasound diagnostic device 100 itself or received from a server. However, this configuration is not limited to this. For example, a configuration may be adopted in which a PC (Personal Computer) as an information processing device is provided on a communication network connected to the ultrasound diagnostic device 100 or the server. The control unit of the PC receives and acquires ultrasound image data from the ultrasound diagnostic device 100 or the server, performs machine learning, and generates estimation data. The control unit of the PC transmits the estimation data to the ultrasound diagnostic device 100 and stores it. Alternatively, the control unit of the PC receives and acquires ultrasound image data that is the target of anomaly detection from the ultrasound diagnostic device 100. The control unit of the PC uses the estimation data to determine whether there is an installation abnormality in a component of the ultrasound diagnostic device 100 from the received ultrasound image data and generate result information. The control unit of the PC displays the result information on the display unit of the device. Alternatively, the control unit of the PC transmits the result information to the ultrasound diagnostic device 100 and displays it on the display unit 17. According to this configuration, abnormal installation of parts of the ultrasound diagnostic device 100 can be detected by remote processing using a PC.

[0112] Furthermore, the detailed configuration and detailed operation of the ultrasound diagnostic device 100 in the above-described embodiment can also be modified as appropriate without departing from the spirit of the present invention. [Explanation of symbols]

[0113] 100 Ultrasound diagnostic equipment 1. Ultrasound diagnostic device 11 Operation input section 12 Transmitter 131 Receiving amplifier 132 AD converter 14 Beamformer 15 Signal processing section 16 Communications Department 17 Display section 18 Control Unit 19 Memory section 2 Ultrasonic probe 21 Ultrasonic probe body 22 Cable 23 Connector 30 Analog Blocks 31 Analog board 32 stacks 40 Digital Blocks 41 Digital Board 42 stacks 51 frames 52 Signal line 61 Main power supply 62,63 Power supply section

Claims

1. an acquisition unit that acquires ultrasound image data; a control unit that estimates noise from the acquired ultrasound image data using trained estimation data for estimating noise in an ultrasound image due to an installation abnormality of a part of the ultrasound diagnostic device, determines an installation abnormality of a part of the ultrasound diagnostic device corresponding to the noise, and outputs information on the determination result.

2. The information processing apparatus according to claim 1 , wherein the part installation abnormality is an abnormality caused by an assembly error during the manufacture of the ultrasonic diagnostic apparatus, or an abnormality caused by a defect in part installation that occurs over time even when the ultrasonic diagnostic apparatus is used normally.

3. the estimation data is data for estimating an abnormality location of an installation abnormality of a part corresponding to an estimated noise, The information processing device according to claim 1 , wherein the control unit uses the estimation data to estimate an abnormality location due to an installation error of a component of the ultrasound diagnostic device from the acquired ultrasound image data, and outputs result information including the abnormality location.

4. The information processing apparatus according to claim 3 , wherein the control unit outputs repair information relating to repair of the abnormality at the abnormal location.

5. the estimation data is data for estimating noise corresponding to an installation abnormality of a part that is outside an allowable range, The information processing apparatus according to claim 1 , wherein the control unit uses the estimation data to estimate noise based on an installation error of a part that is outside an allowable range for the ultrasonic diagnostic apparatus from the acquired ultrasonic image data.

6. the acquisition unit acquires ultrasound image data associated with at least one setting information of an image mode and a setting parameter; The estimation data is prepared for each piece of setting information, The information processing device according to claim 1 , wherein the control unit uses estimation data corresponding to setting information of the acquired ultrasound image data to determine, from the ultrasound image data, an installation abnormality of a part of the ultrasound diagnostic device corresponding to the noise.

7. The information processing apparatus according to claim 1 , wherein the component mounting abnormality is an abnormality in a component other than an analog component.

8. the information processing device is the ultrasound diagnostic device, The information processing apparatus according to claim 1 , wherein the acquisition unit generates and acquires the ultrasound image data.

9. An information processing device including a control unit that performs machine learning on ultrasound image data of whether or not there is noise in the ultrasound image due to an abnormality in the installation of a component of the ultrasound diagnostic device, and generates trained estimation data for estimating noise in the ultrasound image due to an abnormality in the installation of the component of the ultrasound diagnostic device.

10. an acquisition step of acquiring ultrasound image data; a control step of estimating noise from the acquired ultrasound image data using trained estimation data for estimating noise in ultrasound images due to abnormal installation of components of the ultrasound diagnostic device, determining whether there is an abnormal installation of the component corresponding to the noise, and outputting information on the determination result.

11. A learning method that includes a control unit that performs machine learning on ultrasound image data with or without noise in ultrasound images due to installation abnormalities of components in an ultrasound diagnostic device, and generates trained estimation data for estimating noise in ultrasound images due to installation abnormalities of components in an ultrasound diagnostic device.

12. Computer, an acquisition unit that acquires ultrasound image data; a control unit that estimates noise from the acquired ultrasound image data using trained estimation data for estimating noise in the ultrasound image due to an installation abnormality of a part of the ultrasound diagnostic device, determines an installation abnormality of the part of the ultrasound diagnostic device corresponding to the noise, and outputs information on the determination result; A program to function as a

13. Computer, a control unit that performs machine learning on ultrasound image data of whether or not there is noise in the ultrasound image due to an abnormality in the installation of a component of the ultrasound diagnostic device, and generates trained estimation data for estimating noise in the ultrasound image due to an abnormality in the installation of the component of the ultrasound diagnostic device; A program to function as a

Images