Ultrasonic diagnosis apparatus, information processing apparatus, ultrasonic image generation method, ultrasonic image learning method and program

Abstract

To obtain an ultrasonic image with good image quality without saturation even in a case of saturation due to a strong echo.SOLUTION: An ultrasonic diagnosis apparatus 100 comprises a control unit 18. The control unit 18 generates trained data of a trained model obtained by machine learning using saturated ultrasonic image data based on a received signal of an ultrasonic probe, and unsaturated ultrasonic image data. Then, the control unit 18 generates unsaturated ultrasonic image data from saturated ultrasonic image data by using the trained data.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an ultrasonic diagnostic apparatus, an information processing apparatus, an ultrasonic image generation method, an ultrasonic image learning method, and a program.

Background Art

[0002] Conventionally, there is known an ultrasonic diagnostic apparatus that irradiates ultrasonic waves into a subject using an ultrasonic probe, receives the reflected waves, and analyzes them to display an ultrasonic image of the inside of the subject. The subject is a living body of a patient or the like.

[0003] The receiving circuit of the ultrasonic diagnostic apparatus is required to have a wide dynamic range. The gain setting of the ultrasonic diagnostic apparatus is adjusted so that the receiving circuit does not saturate. Also, the gain setting of the ultrasonic diagnostic apparatus is adjusted so that the gain is not too low and the signal-to-noise ratio (S / N ratio) does not deteriorate. However, a received signal larger than expected may be input to the receiving circuit. In this case, the receiving circuit may saturate. In this case, for example, in a tomographic image, image quality deterioration such as the image flowing horizontally may occur.

[0004] Here, with reference to FIG. 7, the gain adjustment of a conventional received signal will be described. FIG. 7 is a block diagram showing a vibrator 211 and a receiving unit 33 of a conventional ultrasonic probe. The receiving unit 33 includes a preamplifier 331, a gain variable amplifier 332, an AD (Analog to Digital) converter 333, and a beamformer 334. The vibrator 211 includes vibrators 211a to 211h. The preamplifier 331 includes preamplifiers 331a to 331h. The gain variable amplifier 332 includes gain variable amplifiers 332a to 332h. The AD converter 333 includes AD converters 333a to 333h.

[0005] In FIG. 7, the transmitter that transmits the drive signal to the vibrators 211a to 211h of the ultrasonic probe and the configuration that generates image data from the beam data and displays it on the display unit are not shown. Further, the vibrators 211a to 211h and the corresponding parts represent the vibrators of the ultrasonic probe typically, and are not limited to the number thereof (8).

[0006] The ultrasonic waves emitted from the vibrators 211a to 211h of the ultrasonic probe are reflected by the subject and received by the vibrators 211a to 211h as reflected ultrasonic waves (echoes) and converted into received signals. The received signals are electrical signals

[0007] Each of the received signals output from the vibrators 211a to 211h is amplified by the preamplifiers 331a to 331h, and further amplified by the gain variable amplifiers 332a to 332h. Each of the amplified received signals is converted into a digital signal by the AD converters 333a to 333h, and beamformed by the beamformer 334 to generate beam data.

[0008] Here, the gain control will be described. The ultrasonic waves received by the vibrators 211a to 211h have a wide dynamic range. In contrast, the dynamic ranges of the preamplifiers 331a to 331h, the gain variable amplifiers 332a to 332h, the AD converters 333a to 333h, and the beamformer 334 after being converted into received signals are narrow. In particular, the dynamic range after the AD converters 333a to 333h is narrow. Therefore, the gain variable amplifiers 332a to 332h adjust the amplification factor so that the digitized signals after AD conversion have appropriate amplitudes.

[0009] Generally, since the propagation distance is shorter as it is closer to the body surface of the subject, the amplitude of the received signal is larger. Also, the amplitude of the received signal decreases as it is deeper from the body surface. However, the amplitude of the received signal varies depending on the shape of the tissue interface and the magnitude of the impedance difference. Therefore, if a received signal larger than expected enters from the vibrators 211a to 211h, saturation will occur in the subsequent circuit.

[0010] Saturation may occur when ultrasonic waves are input to all the oscillators 211 or when ultrasonic waves are input to some of the oscillators. However, it is difficult to perform optimal gain control depending on the subject, the site to which the ultrasonic waves are applied, or the way the ultrasonic probe is applied.

[0011] When saturation occurs, for example, an error may occur in the coherent addition in the beam former 134, and a correct beam may not be formed. Also, in the image mode for displaying blood flow signals, small-amplitude blood flow echoes are superimposed on large-amplitude tissue echoes. For this reason, when the tissue echoes are saturated, the blood flow echoes may disappear. Also, if the gain is set too low to avoid saturation, the signal will be buried in noise and cannot be extracted.

[0012] Generally, the adjustment of the amplifier in the receiving unit is not included in the adjustment items by the user. As gain adjustments for ultrasonic diagnostic devices, there are overall gain adjustments and TCG (Time Control Gain) adjustments. However, these are not adjustments of the amplifier in the receiving unit. For this reason, when an ultrasonic image is saturated, it is often impossible to make an adjustment to eliminate the saturation. Also, it is difficult for the user to make appropriate adjustments. Furthermore, in order to ensure a sufficient dynamic range, it is necessary to increase the power consumption. In particular, in recent portable ultrasonic diagnostic devices, saturation is likely to occur in order to suppress the power consumption. The degradation of the image quality of the ultrasonic image when saturated is quite noticeable. Measures are required so as not to cause saturation or so that the image does not degrade even when saturated from the viewpoint of quality.

[0013] Therefore, an ultrasonic diagnostic device is known that determines saturation according to the odd-order harmonic components extracted from the received signals (reflected wave signals) received by each element of the ultrasonic probe (see Patent Document 1). This ultrasonic diagnostic device multiplies a weight coefficient to the received signal determined to be saturated (reduces the gain), and generates beam data (reflected wave data) from the multiplied received signal.

[0014] In addition, an ultrasonic diagnostic apparatus having a saturation estimation function for estimating saturation of a reflected wave signal via an ultrasonic probe and a transmission / reception circuit is known (see Patent Document 2). When there is an amplitude exceeding a threshold value in even one of the reflected wave signals, the saturation estimation function sets all of the packet data to zero. This ultrasonic diagnostic apparatus interpolates (fills in the holes) the moving body information at the observation points where packet data is missing from the surrounding moving body information.

[0015] In addition, an ultrasonic diagnostic apparatus having a learned model that generates output data based on a high ultrasonic pressure signal using input data based on a low ultrasonic pressure signal of an ultrasonic wave is known (see Patent Document 3). The ultrasonic diagnostic apparatus generates output data based on a high ultrasonic pressure signal by inputting the input data based on the low ultrasonic pressure signal of the ultrasonic wave acquired by inspection using the learned model.

Prior Art Documents

Patent Documents

[0016]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0017] The ultrasonic diagnostic apparatus of Patent Document 1 reduces the influence of signal deterioration in saturation. However, on the other hand, the amplitude balance between elements (channels) is disrupted. For this reason, there is a possibility that there will be a problem with the beam shape. In addition, the ultrasonic diagnostic apparatus of Patent Document 2 interpolates the observation points with saturation using the surrounding moving body information. For this reason, there is a demand for better image quality.

[0018] In addition, the ultrasonic diagnostic apparatus of Patent Document 3 does not reduce the influence of signal deterioration in saturation.

[0019] An object of the present invention is to obtain an ultrasonic image with good image quality without saturation even in the case of saturation due to strong echoes.

Means for Solving the Problems

[0020] To solve the above problems, the ultrasonic diagnostic apparatus according to the invention described in claim 1 includes a control unit that uses learned data of a model learned using image data with saturation and image data without saturation based on reception signals of an ultrasonic probe, and generates image data without saturation from the image data with saturation.

[0021] The invention described in claim 2 is the ultrasonic diagnostic apparatus according to claim 1, wherein the control unit determines whether the image data based on the reception signal of the vibrator of the ultrasonic probe is saturated or not using the learned data, and when it is saturated, reduces the gain of a gain variable amplifier that arbitrarily amplifies the reception signal so as not to saturate, and generates image data.

[0022] The invention described in claim 3 is the ultrasonic diagnostic apparatus according to claim 2, wherein the control unit determines whether the image data based on the reception signal of the ultrasonic probe includes a saturated region or not using the learned data, and when there is a saturated region, reduces the gain of the gain variable amplifier for the region that arbitrarily amplifies the reception signal so as not to saturate, and generates image data.

[0023] The invention described in claim 4 is the ultrasonic diagnostic apparatus according to claim 1, wherein the control unit determines whether the image data based on the reception signal of the ultrasonic probe is saturated or not using the learned data, and when it is saturated, estimates and generates image data of a region without saturation from the image data using the learned data.

[0024] The invention described in claim 5 is the ultrasonic diagnostic apparatus according to claim 4, The control unit determines, based on the learned data, whether the image data based on the reception signal of the ultrasonic probe includes a saturated region. When there is a saturated region, the control unit estimates and generates, based on the learned data, image data of a non-saturated region from the image data of the region, and generates image data including the non-saturated region.

[0025] The information processing apparatus according to claim 6 includes a control unit that performs machine learning using saturated image data based on the reception signal of the ultrasonic probe and non-saturated image data, and generates learned data of the model of the machine learning.

[0026] The invention according to claim 7 is the information processing apparatus according to claim 6, wherein the learned data includes determination data for determining whether the image data has saturation.

[0027] The invention according to claim 8 is the information processing apparatus according to claim 7, wherein the determination data is data for determining whether the image data includes a saturated region.

[0028] The invention according to claim 9 is the information processing apparatus according to any one of claims 6 to 8, wherein the learned data includes estimation data for generating non-saturated image data from saturated image data.

[0029] The invention according to claim 10 is the information processing apparatus according to claim 9, wherein the estimation data is data for estimating and generating non-saturated region image data from saturated region image data and generating image data including the non-saturated region.

[0030] The ultrasonic image generation method according to claim 11 Using the learned data of a model learned using data for an image with saturation and data for an image without saturation based on the reception signal of an ultrasonic probe, a control step of generating data for an image without saturation from the data for an image with saturation is included.

[0031] The ultrasonic image learning method according to claim 12 is A control step of performing machine learning using data for an image with saturation and data for an image without saturation based on the reception signal of an ultrasonic probe, and generating learned data of the model of the machine learning is included.

[0032] The program according to claim 13 is causing a computer to function as a control unit that generates data for an image without saturation from the data for an image with saturation, using the learned data of a model learned using data for an image with saturation and data for an image without saturation based on the reception signal of an ultrasonic probe.

[0033] The program according to claim 14 is causing a computer to function as a control unit that performs machine learning using data for an image with saturation and data for an image without saturation based on the reception signal of an ultrasonic probe, and generates learned data of the model of the machine learning.

Advantages of the Invention

[0034] According to the present invention, even in the case of saturation due to a strong echo, an ultrasonic image with good image quality without saturation can be obtained.

Brief Description of the Drawings

[0035]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0036] Hereinafter, with reference to the drawings, the first and second embodiments of the present invention will be described in detail in order. However, the scope of the invention is not limited to the illustrated examples.

[0037] (First Embodiment) With reference to FIGS. 1 to 5, the first embodiment of the present invention will be described. First, with reference to FIGS. 1 to 5, the apparatus configuration of this embodiment will be described. FIG. 1 is a schematic diagram of an ultrasonic diagnostic apparatus 100 of this embodiment. FIG. 2 is a block diagram showing the functional configuration of the ultrasonic diagnostic apparatus 100. FIG. 3 is a block diagram of the ultrasonic diagnostic apparatus 100 showing the internal configuration of the receiving unit 13.

[0038] As shown in FIG. 1, the ultrasonic diagnostic apparatus 100 is provided in a medical facility such as a hospital, emits ultrasonic waves to a subject such as a patient's living body, and generates ultrasonic image data. Further, the ultrasonic diagnostic apparatus 100 estimates ultrasonic image data using a learned model of machine learning. Further, the ultrasonic diagnostic apparatus 100 is configured to display a color Doppler image in color Doppler mode. The color Doppler image in color Doppler mode is a display image of ultrasonic image data obtained by superimposing color Doppler (color flow) image data and B (Brightness) image data. The color Doppler image data is ultrasonic image data of a tomographic image showing the blood flow state of the subject in color. The B image data is ultrasonic image data of a tomographic image showing the tissue of the subject and the like in luminance.

[0039] The ultrasonic diagnostic apparatus 100 includes an ultrasonic diagnostic apparatus main body 1 and an ultrasonic probe 2. The ultrasonic probe 2 is connected to the ultrasonic diagnostic apparatus main body 1. The ultrasonic probe 2 transmits ultrasonic waves (transmitted ultrasonic waves) into the subject and receives reflected waves (reflected ultrasonic waves: echoes) of the ultrasonic waves reflected in the subject. The ultrasonic probe 2 has an ultrasonic probe main body 21, a cable 22, and a connector 23. The ultrasonic probe main body 21 is the header part of the ultrasonic probe 2 and transmits and receives ultrasonic waves. The cable 22 is connected to the ultrasonic probe main body 21 and the connector 23. The cable 22 is a cable through which a drive signal for the ultrasonic probe main body 21 and a reception signal of ultrasonic waves flow. The connector 23 is a plug connector for connecting to a connector (not shown) of a receptacle of the ultrasonic diagnostic apparatus main body 1.

[0040] The ultrasonic diagnostic apparatus main body 1 is connected to the ultrasonic probe main body 21 via the connector 23 and the cable 22. The ultrasonic diagnostic apparatus main body 1 transmits a drive signal of an electrical signal to the ultrasonic probe main body 21 to cause the ultrasonic probe main body 21 to transmit transmitted ultrasonic waves to the subject. The ultrasonic probe 2 generates a reception signal, which is an electrical signal, in response to the reflected ultrasonic waves from the subject received by the ultrasonic probe main body 21. The ultrasonic diagnostic apparatus main body 1 images the internal state of the subject as ultrasonic image data based on the reception signal generated by the ultrasonic probe 2.

[0041] The ultrasonic probe main body 21 has a vibrator 211 (Fig. 2) on the tip side. The vibrator 211 has vibrators 211a to 211h (Fig. 3). However, the eight vibrators 211a to 211h are representative of each vibrator of the vibrator 211 for simplicity of explanation. The number of each vibrator of the vibrator 211 can be arbitrarily set and is, for example, 192 in actuality.

[0042] Each oscillator of the oscillator 211 is, for example, arranged in a one-dimensional array in the scanning direction (azimuth direction). Note that each oscillator of the oscillator 211 may be arranged in a two-dimensional array. In the present embodiment, a linear scanning type electronic scan probe is adopted as the ultrasonic probe 2. However, the ultrasonic probe 2 may be either an electronic scanning method or a mechanical scanning method. Further, the ultrasonic probe 2 may be any of a linear scanning method, a sector scanning method, or a convex scanning method. The ultrasonic diagnostic apparatus main body 1 and the ultrasonic probe 2 may be configured to perform wireless communication instead of wired communication via the cable 22. This wireless communication is, for example, UWB (Ultra Wide Band).

[0043] The operation input unit 11 is a control panel or the like that receives various operation inputs from users such as doctors and technicians. The operation input unit 11 has operation elements such as push buttons, encoders, lever switches, joysticks, trackballs, keyboards, touch pads, and multi-function switches.

[0044] The display unit 17 has a display panel such as an LCD (Liquid Crystal Display), an organic EL (Electro-Luminescence) display, or an inorganic EL display. The display unit 17 displays display information such as ultrasonic image data on the display panel.

[0045] As shown in FIG. 2, the ultrasonic diagnostic apparatus main body 1 includes an operation input unit 11, a transmission unit 12, a reception unit 13, a signal processing unit 14, an image processing unit 15, a display control unit 16, a display unit 17, a control unit 18, and a storage unit 19.

[0046] The operation input unit 11 receives various operation inputs from the user and outputs the operation signals to the control unit 18. The operation input unit 11 may be configured to include a touch panel that is integrally formed on the display screen of the display unit 17 and receives touch inputs from the user. Further, the operation input unit 11 receives operation inputs for the display mode of the blood flow components (blood flow velocity, power, dispersion) in the color Doppler mode. Further, the operation input unit 11 receives operation inputs for the ROI (Region of Interest) for the color Doppler image.

[0047] The transmission unit 12 supplies a drive signal, which is an electrical signal, to the ultrasonic probe 2 in accordance with the control of the control unit 18 to generate transmission ultrasonic waves in the ultrasonic probe 2. The transmission unit 12 includes, for example, a clock generation circuit, a delay circuit, and a pulse generation circuit. The clock generation 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 vibrator 211 and delays the transmission of the drive signal by the set delay time. The delay circuit focuses the transmission beam constituted by the transmission ultrasonic waves due to the delay. The pulse generation circuit generates a pulse signal as a drive signal at a predetermined period. The transmission unit 12 drives a continuous part (for example, 64) of a plurality (for example, 192) of vibrators 211 arranged in the ultrasonic probe 2 to generate transmission ultrasonic waves. Then, the transmission unit 12 scans by shifting the vibrator 211 to be driven in the scanning direction every time transmission ultrasonic waves are generated.

[0048] Further, the transmission unit 12 generates a drive signal corresponding to the ROI input by the operation input unit 11 for the color Doppler image data in accordance with the control of the control unit 18. The drive signal corresponding to the ROI of the color Doppler image is a drive signal for transmitting a plurality of ultrasonic waves on the same scanning line. Further, the transmission unit 12 generates a drive signal corresponding to the entire region for the B (Brightness) mode image data in accordance with the control of the control unit 18.

[0049] The receiving unit 13 receives a received signal, which is an electrical signal, from the ultrasonic probe 2 in accordance with the control of the control unit 18. Here, with reference to FIG. 3, the internal configuration of the receiving unit 13 will be described. The receiving unit 13 includes a preamplifier 131, a gain variable amplifier 132, an AD converter 133, and a beamformer 134. The preamplifier 131 includes preamplifiers 131a to 131h. The gain variable amplifier 132 includes gain variable amplifiers 132a to 132h. The AD converter 133 includes AD converters 133a to 133h. The number of each of the preamplifier 131, the gain variable amplifier 132, and the AD converter 133 corresponds to the number (8) corresponding to the vibrators 211a to 211h, but is not limited thereto.

[0050] The preamplifiers 131a to 131h amplify the voltage of the received signal generated by the vibrators 211a to 211h with a predetermined gain value (amplification factor) set in advance. The gain variable amplifiers 132a to 132h amplify the voltage of the received signal amplified by the preamplifiers 131a to 131h with an arbitrary gain value (amplification factor) in accordance with the control of the control unit 18. The AD converters 133a to 133h convert the analog received signal amplified by the gain variable amplifiers 132a to 132h into a digital received signal.

[0051] The beamformer 134 gives a delay time to the received signals A / D-converted by the AD converters 133a to 133h for each individual path corresponding to each vibrator 211 to adjust the phase. The beamformer 134 adds (coherent addition) the received signals after these processes to generate beam data.

[0052] When generating B-mode image data, the signal processing unit 14 performs envelope detection processing, logarithmic compression, etc. on the beam data from the receiving unit 13 in accordance with the control of the control unit 18. The signal processing unit 14 further adjusts the dynamic range and gain of the beam data after these operations to perform luminance conversion. The signal processing unit 14 generates B-mode image data composed of pixels having luminance values as received energy by this luminance conversion. That is, the B-mode image data represents the strength of the received signal by luminance.

[0053] Also, the signal processing unit 14 generates color Doppler image data. In this case, the signal processing unit 14 generates color Doppler image data of the ROI according to the control of the control unit 18 and in response to the acoustic line data input from the receiving unit 13. This ROI is the ROI input via the operation input unit 11. The signal processing unit 14 includes an orthogonal demodulation circuit, a corner turn control unit, an MTI (Moving Target Indicator) filter, a correlation calculation unit, a data conversion unit, a noise removal spatial filter unit, an inter-frame filter, and a color Doppler image conversion unit for the color Doppler mode.

[0054] The orthogonal demodulation circuit orthogonally demodulates the received signal in the color Doppler mode input from the receiving unit 13 according to the control of the control unit 18. The orthogonal demodulation circuit calculates the phase difference between the acquired received signal in the color Doppler mode and the reference signal by orthogonal demodulation, and acquires the (complex) Doppler signals I, Q. The corner turn control unit arranges the Doppler signals I, Q input from the orthogonal demodulation circuit according to the control of the control unit 18. The arrangement is an arrangement in the depth direction from the ultrasonic probe to the subject and the ensemble direction of the number of repetitions n of ultrasonic transmission and reception for each same acoustic line (line). The corner turn control unit stores the arranged Doppler signals I, Q in a memory (not shown) and reads out the Doppler signals I, Q in the ensemble direction for each depth. The received signal (Doppler signals I, Q) contains not only the signal components of the blood flow necessary for color flow image generation but also unnecessary information such as blood vessel walls and tissues (clutter components). The MTI filter filters the Doppler signals I, Q input from the corner turn control unit to remove the clutter components according to the control of the control unit 18.

[0055] The correlation calculation unit calculates the real part D and the imaginary part N of the average value S of the autocorrelation operation of the Doppler signal from the Doppler signals I and Q from the MTI filter according to the control of the control unit 18. The Doppler signals I and Q are complex Doppler signals z. The average value S of the autocorrelation operation of the Doppler signal is the average value of the phase difference vectors. The data conversion unit calculates the blood flow components from the Doppler signals I and Q from the MTI filter, the real part D and the imaginary part N of the average value S of the autocorrelation operation according to the control of the control unit 18. The blood flow components are blood flow velocity, power, and variance.

[0056] The noise removal spatial filter unit filters the power, the blood flow velocity, and the variance calculated by the data conversion unit according to the control of the control unit 18. The inter-frame filter performs inter-frame filtering of the blood flow components from the noise removal spatial filter unit according to the control of the control unit 18. The inter-frame filter selects the blood flow components that make up the color Doppler image from the noise removal spatial filter unit corresponding to the display mode of the color Doppler mode from the operation input unit 11. The inter-frame filter performs filtering so as to smooth the change between the frames of the selected blood flow components and leave an afterimage. The color Doppler image conversion unit generates the color flow image data of the ROI by color mapping the blood flow components from the inter-frame filter according to the control of the control unit 18. For example, in the color Doppler image data according to the blood flow velocity, the blood flow flowing in the direction toward the ultrasonic probe 2 is represented in red. In the color Doppler image data, the blood flow flowing in the receding direction is represented in blue.

[0057] The image processing unit 15 has an image memory unit 15a. The image memory unit 15a is composed of a semiconductor memory such as a DRAM (Dynamic Random Access Memory), for example. The image processing unit 15 stores the B-mode image data and the color Doppler image data transmitted from the signal processing unit 14 in the image memory unit 15a in units of frames according to the control of the control unit 18. The image processing unit 15 transmits the B-mode image data and the color Doppler image data stored in the image memory unit 15a to the display control unit 16 one frame at a time at predetermined intervals.

[0058] The display control unit 16 is, for example, a DSC (Digital Scan Converter). The display control unit 16 performs processes such as coordinate conversion on the ultrasonic image data input from the image processing unit 15 according to the control of the control unit 18, and converts it into an image signal for display. In particular, in the color Doppler mode, the display control unit 16 performs a process of superimposing color Doppler image data and B-mode image data. The display control unit 16 outputs the image signal to the display unit 17.

[0059] The display unit 17 displays an ultrasonic image on the display panel according to the image signal output from the display control unit 16 according to the control of the control unit 18. Further, the display unit 17 displays various display information input from the control unit 18 on the display panel.

[0060] 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 out various processing programs stored in the ROM and expands them in the RAM, and controls each part of the ultrasonic diagnostic apparatus 100 in cooperation with the expanded programs and the CPU. The ROM is composed of a non-volatile memory such as a semiconductor. The ROM stores a system program corresponding to the ultrasonic 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 learning program for executing the learning process described later and a first image display program for executing the first image display process described later. These programs are stored in the RAM in the form of computer-readable program codes. The CPU sequentially executes operations according to the program codes on the RAM. The RAM forms a work area for temporarily storing various programs executed by the CPU and data related to these programs.

[0061] The storage unit 19 is a storage unit such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) that can write and read information such as ultrasonic image data. In particular, the storage unit 19 stores learned data as a learned model of machine learning. The learned data includes determination data and estimation data. The determination data is data for determining the presence or absence of a saturated region at an arbitrary position in the ultrasonic image data. The estimation data is data for estimating and generating ultrasonic image data of a non-saturated region from ultrasonic image data of a saturated region.

[0062] Next, with reference to FIGS. 4 and 5, the operation of the ultrasonic diagnostic apparatus 100 of the present embodiment will be described. FIG. 4 is a flowchart showing a learning process. FIG. 5 is a flowchart showing a first image display process.

[0063] First, with reference to FIG. 4, the learning process executed by the ultrasonic diagnostic apparatus 100 will be described. The learning process is a process of acquiring ultrasonic image data with or without saturation as teacher data for a subject such as a patient serving as a learning sample and performing machine learning.

[0064] In the ultrasonic diagnostic apparatus 100, for example, an execution instruction for the learning process is input from a user via the operation input unit 11. The control unit 18 executes the learning process according to the learning program stored in the ROM in response to the execution instruction.

[0065] First, the control unit 18 acquires ultrasonic image data without saturation and stores it in the storage unit 19 under the control of the transmission unit 12 to the display control unit 16 (step S11). The ultrasonic image data without saturation in step S11 is color Doppler image data. The ultrasonic image data without saturation can be obtained, for example, by controlling the gain values of the gain variable amplifiers 132a to 132h to be low.

[0066] The control unit 18 acquires ultrasonic image data having saturation (having an area) under the control of the transmission unit 12 to the display control unit 16 and stores it in the storage unit 19 (step S12). The ultrasonic image data having saturation in step S12 is also color Doppler image data. The ultrasonic image data having saturation can be obtained, for example, by controlling the gain values of the gain variable amplifiers 132a to 132h to be high. However, a configuration may be adopted in which ultrasonic image data with or without saturation is generated in advance by the own device or another ultrasonic diagnostic device and stored in the storage unit 19. In this configuration, in steps S11 and S12, the control unit 18 reads and acquires ultrasonic image data with or without saturation from the storage unit 19.

[0067] Also, a configuration may be adopted in which ultrasonic image data with or without saturation, which has been photographed in advance by the own device or another ultrasonic diagnostic device, is stored in a device such as a server. In this configuration, in steps S11 and S12, for example, the control unit 18 receives and acquires ultrasonic image data with or without saturation from the device via a communication unit (not shown).

[0068] The control unit 18 determines whether the number of accumulated data of the ultrasonic image data with or without saturation 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 sufficient number of accumulated data for machine learning of the ultrasonic image data with or without saturation. Machine learning, for example, estimates the boundary of the feature amount using the ultrasonic image data with or without saturation accumulated in the storage unit 19 as teacher data. Also, machine learning generates determination data for ultrasonic image data with or without saturation as learned data using the boundary. Furthermore, machine learning is assumed to generate estimation data for estimating ultrasonic image data without saturation from ultrasonic image data with saturation as learned data using the boundary.

[0069] If it is less than the predetermined number (step S13; NO), the process proceeds to step S11. If it is equal to or greater than the predetermined number (step S13; YES), the control unit 18 performs machine learning using the ultrasonic image data with or without saturation in the storage unit 19 (step S14). The control unit 18 extracts the learned data from the learning result of the machine learning in step S14 and stores it in the storage unit 19 (step S15). The learning process ends. The learned data becomes the determination data and the estimation data for the color Doppler image data.

[0070] Next, with reference to FIG. 5, the first image display process executed by the ultrasonic diagnostic apparatus 100 will be described. The first image display process is a process of acquiring ultrasonic image data of a subject such as a patient to be diagnosed, and estimating and generating an estimated image data without saturation and displaying it when there is a saturated area.

[0071] In the ultrasonic diagnostic apparatus 100, after the learning process, for example, an execution instruction for the first image display process is input from the user via the operation input unit 11. The control unit 18 executes the first image display process according to the first image display program stored in the ROM in response to the execution instruction.

[0072] First, the control unit 18 acquires ultrasonic image data of the subject to be diagnosed under the control of the transmission unit 12 to the display control unit 16 (step S21). The ultrasonic image data in step S21 includes color Doppler image data and B-mode image data superimposed thereon. The control unit 18 reads the learned data from the storage unit 19 (step S22). The control unit 18 determines whether there is a saturated area in the ultrasonic image data in step S21 using the determination data of the learned data in step S22 (step S23). In step S23, it is assumed that the color Doppler image data in step S21 is determined. The control unit 18 discriminates whether there is a saturated area from the determination result in step S23 (step S24).

[0073] When there is a saturated region (step S24; YES), the process proceeds to step S25. The control unit 18 estimates and generates ultrasonic image data of the same-position region without saturation from the ultrasonic image data of the saturated region determined in step S24 (step S25). The learned data for determination in step S22 is used for this estimation and generation. In step S25, the control unit 18 generates ultrasonic image data with the saturated region replaced by a region without saturation. The ultrasonic image data including the estimated region without saturation is used as estimated image data. In step S25, the estimated image data is color Doppler image data without saturation.

[0074] After the execution of step S25, or when there is no saturation (step S24; NO), the process proceeds to step S26. The control unit 18 displays the ultrasonic image data in step S21 or the estimated image data in step S25 on the display unit 17 (step S26). When step S25 is not executed, the control unit 18 superimposes and displays the color Doppler image data and B-mode image data in step S21. When step S25 is executed, the control unit 18 superimposes and displays the color Doppler image data, which is the estimated image data in step S25, and the B-mode image data in step S21. The first image display process ends.

[0075] The image generation conditions for the ultrasonic image data with / without saturation in the learning process are preferably made to correspond to the image generation conditions of the ultrasonic image data in step S21 of the first image display process. The image generation conditions are various processing conditions related to the generation of ultrasonic image data, such as transmission conditions, reception conditions, image mode, and image processing. These image generation conditions may include the type and part of the subject.

[0076] In the present embodiment, saturation is not observed in individual oscillators 211a to 211h (channels), but saturation is determined (detected) from the image data after coherent addition. Further, in the present embodiment, machine learning is performed based on ultrasonic image data with saturation (having a region) and ultrasonic image data without saturation. Using the data for estimating the learned data, ultrasonic image data without saturation is estimated and generated from the ultrasonic image data including the region with saturation. Thereby, the image quality of the ultrasonic image data (color Doppler image data) is improved. Note that it may be configured such that machine learning is performed based on the ultrasonic image data with saturation in all regions and the ultrasonic image data without saturation. In this configuration, using the data for estimating the learned data, ultrasonic image data without saturation is estimated and generated from the ultrasonic image data with saturation in all regions.

[0077] According to the present embodiment, the ultrasonic diagnostic apparatus 100 includes a control unit 18. The control unit 18 generates learned data of a learned model that has been machine-learned using the ultrasonic image data with saturation and the ultrasonic image data without saturation based on the reception signal of the ultrasonic probe 2. The control unit 18 uses the learned data to generate ultrasonic image data without saturation from the ultrasonic image data with saturation. Further, the control unit 18 performs machine learning using the ultrasonic image data with saturation and the ultrasonic image data without saturation based on the reception signal of the ultrasonic probe 2. The control unit 18 generates learned data of the learned model of the machine learning.

[0078] Therefore, even in the case of saturation due to a strong echo, an ultrasonic image (color Doppler image) with good image quality without saturation can be obtained. In particular, in the color Doppler image data, disappearance of the blood flow echo can be prevented.

[0079] The control unit 18 determines whether the ultrasonic image data based on the reception signal of the ultrasonic probe 2 is saturated or not using the determination data for the learned data. The learned data includes the determination data for determining whether the ultrasonic image data is saturated or not. When it is determined that the data is saturated, the control unit 18 estimates and generates ultrasonic image data without saturation from the ultrasonic image data using the estimation data for the learned data. The learned data includes the estimation data for generating ultrasonic image data without saturation from the ultrasonic image data with saturation. Therefore, it is possible to more accurately determine whether the ultrasonic image data is saturated or not, and even in the case of saturation due to a strong echo, it is possible to obtain an ultrasonic image with better image quality without saturation.

[0080] The control unit 18 determines whether the ultrasonic data based on the reception signal of the ultrasonic probe 2 includes a saturated region using the determination data. The determination data is data for determining whether the ultrasonic image data includes a saturated region. When there is a saturated region, the control unit 18 estimates and generates ultrasonic image data of a region without saturation from the ultrasonic image data of the region using the estimation data. The control unit 18 generates ultrasonic image data including the region without saturation. The estimation data is data for estimating and generating ultrasonic image data of a region without saturation from the ultrasonic image data of a saturated region and generating ultrasonic image data including the region without saturation. Therefore, it is possible to more accurately determine whether the ultrasonic image data includes a saturated region. Also, even in the case of saturation due to a strong echo, it is possible to obtain an ultrasonic image with better image quality without saturation and reduce the image processing load.

[0081] (Second Embodiment) Referring to FIG. 6, a second embodiment of the present invention will be described. FIG. 6 is a flowchart showing the second image display process.

[0082] The first embodiment was configured to generate ultrasonic image data without saturation from ultrasonic image data with saturation (color Doppler image data) using estimation data. In this embodiment, when ultrasonic image data with saturation (color Doppler image data) is acquired, the gain of the receiving unit 13 is decreased to generate ultrasonic image data without saturation.

[0083] In this embodiment, as the device configuration, an ultrasonic diagnostic apparatus 100 is used. However, it is assumed that a learning program and a second image display program for executing the second image display process described later are stored in the ROM of the control unit 18.

[0084] Next, with reference to FIG. 6, the operation of the ultrasonic diagnostic apparatus 100 of this embodiment will be described. With reference to FIG. 6, the second image display process executed by the ultrasonic diagnostic apparatus 100 will be described. The second image display process is a process of acquiring ultrasonic image data of a subject such as a patient for diagnosis, and when there is saturation, reducing the gain to generate and display ultrasonic image data without saturation. Also, the description of the learning process is omitted because it is the same as that of the first embodiment.

[0085] It is assumed that a gain initial value, which is the initial value of each gain value for color Doppler image data of the gain variable amplifier 132, is stored in the storage unit 19 in advance.

[0086] In the ultrasonic diagnostic apparatus 100, after the learning process, for example, an execution instruction for the second image display process is input from the user via the operation input unit 11. In response to the execution instruction, the control unit 18 executes the second image display process according to the second image display program stored in the ROM.

[0087] First, the control unit 18 reads and obtains the initial gain value from the storage unit 19 (step S31). The control unit 18 sets each gain value of the gain variable amplifier 132 according to the initial gain value in step S31 or the decreased gain value in step S36 (step S32). That is, at the first execution of step S32, the initial gain value is set. At the execution of step S32 after that, the decreased gain value in step S36 is set.

[0088] The control unit 18 obtains ultrasonic image data of the subject to be diagnosed by controlling the transmission unit 12 to the display control unit 16 including the control of the gain value set in step S32 (step S33). In step S33, the gain variable amplifiers 132a to 132h are controlled by the gain value set in step S32. Color Doppler image data and B-mode image data corresponding to the set gain value are generated. Steps S34, S35, and S36 are the same as steps S22, S23, and S24 of the first image display process in FIG. 5, respectively.

[0089] When there is saturation (step S36; YES), the control unit 18 decreases the currently set gain value corresponding to the region determined to have saturation by a predetermined amount (step S37). The process proceeds to step S32. When there is no saturation (step S36; NO), the control unit 18 displays the ultrasonic image data acquired in the immediately preceding step S33 on the display unit 17 (step S38). In step S38, the color Doppler image data and B-mode image data acquired in the immediately preceding step S33 are displayed in an overlapping manner. The second image display process ends.

[0090] As described above, according to the present embodiment, the control unit 18 determines whether the ultrasonic image data based on the reception signal of the ultrasonic probe 2 is saturated or not by using the determination data of the learned data. When there is saturation, the control unit 18 decreases the gain of the gain variable amplifier 132 that arbitrarily amplifies the reception signal so as not to saturate, and generates ultrasonic image data. Therefore, even in the case of saturation due to a strong echo, it is possible to obtain an ultrasonic image (color Doppler image) with good image quality without saturation. In addition, since ultrasonic image data without saturation is not generated by image processing, the image processing burden can be reduced.

[0091] The control unit 18 determines whether the ultrasonic image data based on the reception signal of the ultrasonic probe 2 includes a saturated region by using the determination data. When there is a saturated region, the control unit 18 decreases the gain of the gain variable amplifier 132 for the region that arbitrarily amplifies the reception signal so as not to saturate, and generates ultrasonic image data. Therefore, even in the case of saturation due to a strong echo, it is possible to obtain an ultrasonic image with good image quality without saturation, and the processing burden of gain setting can be reduced.

[0092] In the above description, an example in which the ROM of the control unit 18 is used as a computer-readable medium of the program according to the present invention has been disclosed, but the present invention is not limited to this example. As other computer-readable media, non-volatile memories such as flash memories and portable recording media such as CD-ROMs can be applied. In addition, a carrier wave is also applied to the present invention as a medium for providing the data of the program according to the present invention via a communication line.

[0093] Note that the description in the above embodiment is an example of an ultrasonic diagnostic apparatus, an information processing apparatus, an ultrasonic image generation method, an ultrasonic image learning method, and a program according to the present invention, and the present invention is not limited thereto.

[0094] Each of the above embodiments performs machine learning on ultrasonic image data as image data to generate learned data. Further, the first embodiment has a configuration for generating estimated image data without saturation from ultrasonic image data with saturation as image data. However, the present invention is not limited to this configuration. As the image data, it may be configured to use sound ray data or intermediate data generated between sound ray data generation and image data generation.

[0095] Also, in each of the above embodiments, the ultrasonic diagnostic apparatus 100 as an information processing apparatus is configured to perform machine learning using ultrasonic image data with or without saturation. However, the present invention is not limited to this configuration. For example, a server as an information processing apparatus may be provided on a communication network connected to the ultrasonic diagnostic apparatus 100. The server acquires ultrasonic image data with or without saturation generated by the ultrasonic diagnostic apparatus 100, performs machine learning, and generates estimation data. The server transmits the estimation data to the ultrasonic diagnostic apparatus 100 for storage.

[0096] Further, the first embodiment has a configuration for performing machine learning on color Doppler image data with or without saturation in the color Doppler mode. The acquired color Doppler image data is determined as to whether there is saturation. When there is saturation, color Doppler image data without saturation is estimated and generated. The estimated color Doppler image data without saturation is displayed with the B-mode image data superimposed thereon. However, the present invention is not limited to this configuration. For example, it may be configured to perform machine learning on color Doppler image data with or without saturation and B-mode image data. The acquired color Doppler image data is determined as to whether there is saturation (in the region). When there is saturation, color Doppler image data without saturation is estimated and generated. The acquired B-mode image data is determined as to whether there is saturation (in the region). When there is saturation, B-mode image data without saturation is estimated and generated. The estimated color Doppler image data without saturation and the B-mode image data without saturation are superimposed and displayed.

[0097] Further, for example, it may be configured to machine-learn the superimposed image data of the color Doppler image data with / without saturation and the B-mode image data. The acquired superimposed image data is determined as to whether or not there is saturation (in the region). In the case of saturation, the superimposed image data without saturation is estimated and generated for display. Further, it may be configured to machine-learn the ultrasonic image data with / without saturation in another image mode such as B-mode other than the color Doppler mode. The acquired ultrasonic image data is determined as to whether or not there is saturation (in the region). In the case of saturation, the ultrasonic image data without saturation is estimated and generated for display. The above determination can also be applied to the second embodiment.

[0098] In addition, regarding the detailed configuration and detailed operation of the ultrasonic diagnostic apparatus 100 in the above embodiment, it can be appropriately changed within a range not departing from the spirit of the present invention.

Explanation of Signs

[0099] 100 Ultrasonic diagnostic apparatus 1 Ultrasonic diagnostic apparatus main body 11 Operation input unit 12 Transmission unit 13, 33 Reception unit 131, 131a~131h, 331, 331a~331h Preamplifier 132, 132a~132h, 332, 332a~332h Gain variable amplifier 133, 133a~133h, 333, 333a~333h AD converter 134, 334 Beamformer 14 Signal processing unit 15 Image processing unit 151 Image memory unit 16 Display control unit 17 Display unit 18 Control unit 19 Storage unit 2 Ultrasonic probe 21 Ultrasonic probe main body 211, 211a~211l Vibrator 22 Cable 23 Connector

Claims

1. An ultrasonic diagnostic apparatus comprising a control unit that uses learned data of a model learned using image data with saturation and image data without saturation based on a reception signal of an ultrasonic probe to generate image data without saturation from the image data with saturation.

2. The control unit determines, based on the learned data, whether image data based on a reception signal of a vibrator of an ultrasonic probe has saturation, and when there is saturation, decreases the gain of a gain variable amplifier that arbitrarily amplifies the reception signal so as not to saturate, and generates image data. The ultrasonic diagnostic apparatus according to claim 1.

3. The control unit determines, based on the learned data, whether image data based on the reception signal of the ultrasonic probe includes a saturated region, and when there is a saturated region, decreases the gain of the gain variable amplifier that arbitrarily amplifies the reception signal in the region so as not to saturate, and generates image data. The ultrasonic diagnostic apparatus according to claim 2.

4. The control unit determines, based on the learned data, whether image data based on a reception signal of an ultrasonic probe has saturation, and when there is saturation, estimates and generates image data of a region without saturation from the image data based on the learned data. The ultrasonic diagnostic apparatus according to claim 1.

5. The control unit determines, based on the learned data, whether image data based on the reception signal of the ultrasonic probe includes a saturated region, and when there is a saturated region, estimates and generates image data of a region without saturation from the image data of the region based on the learned data, and generates image data including the region without saturation. The ultrasonic diagnostic apparatus according to claim 4.

6. An information processing apparatus comprising a control unit that performs machine learning using image data with saturation and image data without saturation based on a reception signal of an ultrasonic probe, and generates learned data of a model of the machine learning.

7. The learned data includes determination data for determining whether image data has saturation. The information processing apparatus according to claim 6.

8. The determination data is data for determining whether image data includes a saturated region. The information processing apparatus according to claim 7.

9. The learned data includes estimation data for generating image data without saturation from image data with saturation. The information processing apparatus according to any one of claims 6 to 8.

10. The data for estimation is data for generating image data for a region without saturation by estimating and generating the image data for the region without saturation from the image data for the region with saturation, and is the data for generating the image data including the region without saturation, according to the information processing apparatus of Claim 9.

11. An ultrasonic image generation method including a control step of generating image data without saturation from image data with saturation using learned data of a model learned using image data with saturation based on a reception signal of an ultrasonic probe and image data without saturation.

12. An ultrasonic image learning method including a control step of performing machine learning using image data with saturation based on a reception signal of an ultrasonic probe and image data without saturation, and generating learned data of the model of the machine learning.

13. A computer, a control unit that generates image data without saturation from image data with saturation using learned data of a model learned using image data with saturation based on a reception signal of an ultrasonic probe and image data without saturation, a program for causing the computer to function as the control unit.

14. A computer, a control unit that performs machine learning using image data with saturation based on a reception signal of an ultrasonic probe and image data without saturation, and generates learned data of the model of the machine learning, a program for causing the computer to function as the control unit.

Images