Information processing device, ultrasound diagnostic device, and program

Abstract

To support the acquisition of a moving image of a part suitable for stress electrocardiography.SOLUTION: According to an embodiment, an information processing device includes a first acquisition part, a division part, a calculation part, and a selection part. The first acquisition part acquires an ultrasonic moving image photographed while switching cross-sectional surfaces of the heart of a subject. The division part divides the ultrasonic moving image into a plurality of partial moving images on the basis of electrocardiographic waves of the subject. The calculation part calculates reliability showing the possibility that an observation object included in the heart of the subject is included in the partial moving images. The selection part selects the partial moving images on the basis of the reliability.SELECTED DRAWING: Figure 1

Description

Technical Field

[0006] , , , , , ,

[0001] The embodiments disclosed in this specification and the drawings relate to an information processing apparatus, an ultrasonic diagnostic apparatus, and a program.

[0002] Conventionally, an ultrasonic diagnostic apparatus is used for a stress echocardiogram examination that captures ultrasonic images of the heart in a state where a load is applied due to exercise or drugs. In the stress echocardiogram examination, medical staff perform a diagnosis by comparing the ultrasonic image before the load and the ultrasonic image after the load.

[0003] Here, when a predetermined period elapses from the state where a load is applied to the heart, the heart returns to its normal state. Therefore, medical staff such as doctors capture ultrasonic images within a specified time from the state where a load is applied to the heart in the stress echocardiogram examination. Then, the medical staff search for a moving image of a desired cross-section and a part suitable for the stress echocardiogram examination from the captured ultrasonic images.

[0004] However, the ultrasonic images include moving images of a plurality of cross-sections and further include moving images of a plurality of heartbeats. Therefore, medical staff have required a lot of time to search for a moving image of a part suitable for the stress echocardiogram examination.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] One of the problems that the embodiments disclosed herein and in the drawings aim to solve is to assist in acquiring video footage of the portion of the cardiac examination suitable for stress echocardiography. However, the problems that the embodiments disclosed herein and in the drawings aim to solve are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described later can also be positioned as other problems. [Means for solving the problem]

[0007] The information processing device according to the embodiment comprises a first acquisition unit, a division unit, a calculation unit, and a selection unit. The first acquisition unit acquires an ultrasound video taken while switching scanning cross-sections of the subject's heart. The division unit divides the ultrasound video into a plurality of partial videos based on the electrocardiogram waveform of the subject. The calculation unit, Based on the degree to which each of the multiple frame images included in the aforementioned partial video contains a cross-section of the object to be observed within the subject's heart, In the aforementioned video section The cross-section of the object being observed is The selection unit calculates the confidence level indicating the likelihood of inclusion. For each cross-section of the object to be observed Select the aforementioned video segment. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram showing an example of an ultrasound diagnostic apparatus according to the first embodiment. [Figure 2] Figure 2 shows an example of the data structure of the application settings information. [Figure 3] Figure 3 shows an example of the data structure of weight coefficient information. [Figure 4] Figure 4 shows an example of the training state of a pre-trained model for cross-section recognition. [Figure 5] Figure 5 shows an example of the operational state of a pre-trained model for cross-section recognition. [Figure 6] Figure 6 shows an example of the training state of a pre-trained model for region detection. [Figure 7] Figure 7 shows an example of the operational state of a pre-trained model for region detection. [Figure 8]Figure 8 shows an example of the training state of a pre-trained model for position detection. [Figure 9] Figure 9 shows an example of the operational state of a pre-trained model for position detection. [Figure 10] Figure 10 shows an example of an image from a stress echocardiogram. [Figure 11] Figure 11 shows an example of a Doppler ultrasound image. [Figure 12] Figure 12 is a flowchart showing an example of a selection process performed by an ultrasound diagnostic device according to the first embodiment. [Figure 13] Figure 13 is a flowchart showing an example of the cross-sectional recognition process performed by the ultrasound diagnostic device according to the first embodiment. [Figure 14] Figure 14 is a flowchart showing an example of a first B-mode image selection process performed by an ultrasound diagnostic device according to the first embodiment. [Figure 15] Figure 15 is a flowchart showing an example of a second B-mode image selection process performed by the ultrasound diagnostic apparatus according to the first embodiment. [Figure 16] Figure 16 is a flowchart showing an example of the Doppler image selection process performed by the ultrasound diagnostic device according to the first embodiment. [Figure 17] Figure 17 is a flowchart showing an example of a measurement process performed by the ultrasound diagnostic device according to the first embodiment. [Modes for carrying out the invention]

[0009] The information processing device, ultrasonic diagnostic device, and program of the embodiment will be described below with reference to the drawings. In the following embodiment, parts with the same reference numerals perform similar operations, and redundant explanations will be omitted as appropriate.

[0010] (First embodiment) FIG. 1 is a block diagram showing an example of an ultrasonic diagnostic apparatus 1 according to the first embodiment. As shown in FIG. 1, the ultrasonic diagnostic apparatus 1 includes an ultrasonic probe 101, an input device 102, a display 103, a device main body 104, and an electrocardiograph electrode 105.

[0011] The device main body 104 includes a transmission / reception circuit 110, a B-mode processing circuit 111, a Doppler processing circuit 112, an image generation circuit 120, an image memory 130, a storage circuit 140, a NW (network) interface 150, an electrocardiogram unit 160, and a processing circuit 170. The device main body 104 is an example of an information processing device.

[0012] The ultrasonic probe 101 has, for example, a plurality of elements such as piezoelectric vibrators. These plurality of elements generate ultrasonic waves based on a drive signal supplied from the transmission / reception circuit 110 of the device main body 104. The ultrasonic probe 101 also receives a reflected wave from the subject P and converts it into an electrical signal. The ultrasonic probe 101 also has, for example, a matching layer provided on the piezoelectric vibrator and a backing material that prevents the propagation of ultrasonic waves backward from the piezoelectric vibrator. The ultrasonic probe 101 is detachably connected to the device main body 104.

[0013] When ultrasonic waves are transmitted from the ultrasonic probe 101 to the subject P, the transmitted ultrasonic waves are successively reflected at the discontinuous surfaces of the acoustic impedance in the body tissues of the subject P, and are received by the plurality of elements of the ultrasonic probe 101 as reflected wave signals. The amplitude of the received reflected wave signal depends on the difference in acoustic impedance at the discontinuous surface where the ultrasonic waves are reflected. When the transmitted ultrasonic pulse is reflected at the surface of a moving blood flow or a heart wall, etc., the reflected wave signal undergoes a frequency shift depending on the velocity component of the moving object with respect to the ultrasonic transmission direction due to the Doppler effect. Then, the ultrasonic probe 101 outputs the reflected wave signal to the transmission / reception circuit 110 of the device main body 104.

[0014] In this embodiment, the ultrasonic probe 101 is assumed to be a one-dimensional array probe in which a plurality of ultrasonic transducers are arranged along a predetermined direction. However, the ultrasonic probe 101 is not limited to this example, and may be a two-dimensional array probe (a probe in which a plurality of ultrasonic transducers are arranged in a two-dimensional matrix) or a mechanical 4D probe (a probe capable of performing ultrasonic scanning while mechanically shaking the row of ultrasonic transducers in a direction perpendicular to its arrangement direction), as long as it is capable of acquiring volume data.

[0015] The electrocardiograph electrode 105 is an electrode attached to the subject P in an electrocardiograph. The electrocardiograph electrode 105 is connected to the main unit 104 of the device via a cable. The electrocardiograph electrode 105 receives a weak electrical potential from the heart. The electrocardiograph electrode 105 then outputs the received electrical potential to the main unit 104 of the device.

[0016] The input device 102 is implemented by input means such as a mouse, keyboard, buttons, panel switches, touch command screen, foot switch, trackball, or joystick. The input device 102 receives various setting requests from the operator of the ultrasound diagnostic device 1 and transmits the received setting requests to the main unit 104 of the device.

[0017] The display 103 may, for example, display a GUI (Graphical User Interface) for the operator of the ultrasound diagnostic device 1 to input various setting requests using the input device 102, or display ultrasound images shown by ultrasound image data generated in the device body 104. The display 103 is implemented using an LCD monitor, a CRT (Cathode Ray Tube) monitor, or the like.

[0018] The transmitting / receiving circuit 110, under the control of the processing circuit 170, causes the ultrasonic probe 101 to transmit ultrasound and the ultrasonic probe 101 to receive ultrasound (reflected ultrasound waves). In other words, the transmitting / receiving circuit 110 performs ultrasonic scanning via the ultrasonic probe 101.

[0019] More specifically, the transmitting / receiving circuit 110 transmits ultrasound to the ultrasonic probe 101 under the control of the processing circuit 170. The transmitting / receiving circuit 110 includes, for example, a trigger generation circuit, a delay circuit, and a pulser circuit (not shown). The trigger generation circuit repeatedly generates trigger pulses to form the transmitted ultrasound at a predetermined rate frequency. The delay circuit provides each trigger pulse with a delay time necessary to focus the ultrasound into a beam shape for each channel and to determine the transmission directivity. The pulser circuit applies a drive pulse to the ultrasonic probe 101 at a timing based on this trigger pulse.

[0020] Furthermore, the transmitting and receiving circuit 110 generates reflected ultrasonic data, which is ultrasonic data based on the reflected wave signal received by the ultrasonic probe 101. The transmitting and receiving circuit 110 then stores the generated reflected ultrasonic data in a buffer memory.

[0021] More specifically, the reflected ultrasonic waves transmitted by the ultrasonic probe 101 reach a piezoelectric transducer inside the ultrasonic probe 101, where they are converted from mechanical vibrations into electrical signals (reflected wave signals) and input to the transmitting / receiving circuit 110. The transmitting / receiving circuit 110 includes, for example, a preamplifier, an A / D (Analog to Digital) converter, and a quadrature detection circuit, and performs various processing on the reflected wave signals received by the ultrasonic probe 101 to generate reflected ultrasonic data. In this embodiment, "acquiring ultrasonic data" includes obtaining ultrasonic data by transmitting and receiving ultrasonic waves.

[0022] Reflected ultrasonic data is two-dimensional data consisting of multiple data points from multiple sample points arranged along the depth direction on a scan line (hereinafter referred to as a raster), arranged along the raster direction, for a number of times equal to the number of rasters.

[0023] The preamplifier amplifies the reflected wave signal for each channel and performs gain adjustment (gain correction). The A / D converter converts the gain-corrected reflected wave signal into a digital signal by A / D conversion. The quadrature detection circuit converts the A / D-converted reflected wave signal into a baseband in-phase signal (I signal, I) and a quadrature-phase signal (Q signal, Q).

[0024] The quadrature detection circuit then stores the I and Q signals as reflected ultrasonic data in a buffer memory. Hereafter, the I and Q signals will be collectively referred to as the IQ signals. Also, since the IQ signals are A / D converted digital data, they are also called IQ data.

[0025] The B-mode processing circuit 111 performs processes such as logarithmic amplification, envelope detection, and logarithmic compression on the reflected ultrasonic wave data read from the buffer memory to generate data (B-mode data) in which the signal intensity is represented by brightness.

[0026] The Doppler processing circuit 112 performs frequency analysis on reflected ultrasonic wave data stored in the buffer memory to generate data (Doppler data) that extracts motion information based on the Doppler effect of moving objects within the ROI (Region of Interest) set in the scan area. A moving object is, for example, blood. For example, the Doppler processing circuit 112 can perform a color Doppler method, also known as color flow mapping (CFM).

[0027] The image generation circuit 120 generates ultrasound image data based on the data generated by the B-mode processing circuit 111 and the Doppler processing circuit 112. The image generation circuit 120 stores the generated ultrasound image data in the image memory 130.

[0028] More specifically, the image generation circuit 120 generates B-mode image data based on the B-mode data generated by the B-mode processing circuit 111.

[0029] Furthermore, the image generation circuit 120 generates Doppler image data based on the Doppler data generated by the Doppler processing circuit 112. The Doppler image data is an example of blood flow image data in this embodiment. The image generation circuit 120 generates Doppler image data based on the intensity information and phase change information included in the Doppler data generated by the Doppler processing circuit 112.

[0030] Doppler image data can be, for example, velocity image data, dispersion image data, power image data, or a combination thereof. For example, the image generation circuit 120 generates blood flow image data, in which blood flow information is displayed in color, from Doppler data as blood flow information. In this case, the image generation circuit 120 visualizes blood flow as Doppler image data by determining the display position based on the signal power of the blood flow and the display color based on the velocity and direction of the blood flow.

[0031] The image memory 130 stores various image data generated by the image generation circuit 120. For example, the image memory 130 can be implemented using semiconductor memory elements such as RAM or flash memory, a hard disk, or an optical disc.

[0032] The memory circuit 140 is implemented by, for example, a magnetic or optical storage medium, a semiconductor memory element such as flash memory, a hard disk, or a processor-readable storage medium such as an optical disc. The memory circuit 140 stores a program for realizing ultrasonic transmission and reception, various data, etc. The program and various data may be pre-stored in the memory circuit 140, for example. Alternatively, the program and various data may be stored and distributed on a non-transient storage medium, for example, and then read from the non-transient storage medium and installed in the memory circuit 140.

[0033] For example, the memory circuit 140 stores application setting information 141 and weighting coefficient information 142. Figure 2 shows an example of the data structure of the application setting information 141. The application setting information 141 is a setting of whether or not to use evaluation items in the evaluation of each plane of interest. The plane of interest is the plane of observation of the subject for which medical professionals perform image diagnosis in a stress echocardiogram. For example, the first plane of interest is, for example, a four-chamber plane, the second plane of interest is, for example, a two-chamber plane, the third plane of interest is, for example, a parasternal approach left ventricle long-axis plane, and the fourth plane of interest is, for example, a parasternal approach left ventricle short-axis plane. The evaluation items are evaluation items used to evaluate whether the subject is suitable for image diagnosis by medical professionals. Examples of evaluation items include reliability, heart rate, and clarity of the observed subject. Reliability is the degree to which the plane of interest is suitable. Heart rate is the heart rate of subject P. In stress echocardiography, healthcare professionals increase the heart rate of the subject P by applying stress to their heart. They then perform imaging diagnostics on the heart at this elevated heart rate. Therefore, a low heart rate may not be suitable for imaging diagnostics. Thus, heart rate is included as an evaluation criterion. The ranking of the clarity of the observed objects is the order of clarity of the objects observed by healthcare professionals during imaging diagnostics. Examples of observed objects include the heart wall, tricuspid valve, mitral valve, and aortic valve.

[0034] The application setting information 141 is information in which weight coefficients are set for each evaluation item corresponding to each section of interest. Figure 3 shows an example of the data structure of the weight coefficient information 142. As shown in Figure 3, the application setting information 141 has weight coefficients set for each evaluation item of each section of interest.

[0035] The NW interface 150 controls communication between the main unit 104 and external devices. Specifically, the NW interface 150 receives various types of information from external devices and outputs the received information to the processing circuit 170. For example, the NW interface 150 can be implemented using a network card, network adapter, NIC (Network Interface Controller), etc.

[0036] The electrocardiogram unit 160 acquires the weak electrical potential of the subject P's heart from the electrocardiograph electrodes 105. Based on the electrical potential acquired from the electrocardiograph electrodes 105, the electrocardiogram unit 160 generates the subject P's electrocardiogram (ECG). The electrocardiogram unit 160 then stores the ECG waveform in a memory circuit 140 or the like. The electrocardiogram unit 160 may be located outside the main unit 104 of the device. In this case, the electrocardiogram unit 160 is connected to the main unit 104 of the device via a cable or the like.

[0037] The processing circuit 170 controls the entire processing of the ultrasound diagnostic device 1. Specifically, the processing circuit 170 controls the processing of the transmitting / receiving circuit 110, B-mode processing circuit 111, Doppler processing circuit 112, image generation circuit 120, and electrocardiogram unit 160, etc., based on various setting requests input from the operator via the input device 102 and various control programs and data read from the memory circuit 140. The processing circuit 170 also controls the display of ultrasound images.

[0038] Furthermore, the processing circuit 170 executes the image acquisition function 171, the segmentation function 172, the cross-sectional recognition function 173, the B-mode image selection function 174, the Doppler image selection function 175, the image measurement function 176, and the display control function 177. Here, for example, each of the processing functions that are components of the processing circuit 170, namely the image acquisition function 171, the segmentation function 172, the cross-sectional recognition function 173, the B-mode image selection function 174, the Doppler image selection function 175, the image measurement function 176, and the display control function 177, are stored in the memory circuit 140 in the form of programs that can be executed by a computer. The processing circuit 170 is a processor. For example, the processing circuit 170 reads the program from the memory circuit 140 and executes it to realize the function corresponding to each program. In other words, the processing circuit 170 in the state in which each program has been read has the functions shown in the processing circuit 170 of Figure 1. In Figure 1, the processing functions performed by the image acquisition function 171, segmentation function 172, cross-sectional recognition function 173, B-mode image selection function 174, Doppler image selection function 175, image measurement function 176, and display control function 177 are described as being realized by a single processor. However, it is also acceptable to configure the processing circuit 170 by combining multiple independent processors, with each processor executing a program to realize the functions. Furthermore, in Figure 1, the processing circuit 170 is described as storing programs corresponding to each processing function, but it is also acceptable to distribute multiple memory circuits and configure the processing circuit 170 to read the corresponding programs from individual memory circuits.

[0039] In the above explanation, the term "processor" refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphical Processing Unit), an Application Specific Integrated Circuit (ASIC), or a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor functions by reading and executing a program stored in the memory circuit 140. Alternatively, instead of storing the program in the memory circuit 140, the processor may be configured to directly incorporate the program into its circuitry. In this case, the processor functions by reading and executing the program incorporated into the circuitry.

[0040] The image acquisition function 171 acquires ultrasound video by switching between scanning cross-sections of the subject P's heart. The image acquisition function 171 is an example of the first acquisition unit. The image acquisition function 171 is a part of the first acquisition unit. For example, the ultrasound video consists of images taken in no particular order from a four-chamber section, a two-chamber section, a parasternal approach left ventricle long-axis section, and a parasternal approach left ventricle short-axis section within a specified period after the heart has been subjected to stress. The image acquisition function 171 also acquires electrocardiogram waveform information from the electrocardiogram unit 160, which shows the electrocardiogram waveform synchronized with the heart included in the ultrasound video. The image acquisition function 171 then associates the ultrasound video with the electrocardiogram waveform information.

[0041] Furthermore, when the image acquisition function 171 receives an operation to acquire an ultrasound video including Doppler data, it acquires an ultrasound video including Doppler data relating to the blood flow velocity of the subject P. The image acquisition function 171 is an example of a second acquisition unit. Here, the ultrasound diagnostic device 1 acquires B-mode ultrasound images at regular intervals in order to allow the operator to specify the position where Doppler data is being acquired. The ultrasound diagnostic device 1 then displays the acquired B-mode ultrasound images. The ultrasound video including Doppler data has one or more B-mode ultrasound images acquired at regular intervals.

[0042] The division function 172 divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of subject P indicated by the electrocardiogram waveform information associated with the ultrasound video. The division function 172 is an example of a division section. A partial video is a video of ultrasound images for each period during which a medical professional performs an image diagnosis. For example, a partial video may be a video of the duration of one heartbeat, or a video of the end-systolic portion. Furthermore, the period during which a medical professional performs an image diagnosis may be arbitrarily changed by operation. Here, medical professionals cannot make an appropriate diagnosis in the case of videos of arrhythmias. Therefore, the division function 172 excludes the divided videos of arrhythmias from evaluation.

[0043] The cross-sectional recognition function 173 recognizes which cross-section of the subject P is captured in the partial video. Here, the partial video has multiple frame images. The ultrasound diagnostic device 1 displays the video by changing the frame image for each frame rate. The cross-sectional recognition function 173 recognizes which cross-section of the observation target is captured in the video by performing image recognition processing on the frame images included in the partial video.

[0044] For example, the cross-section recognition function 173 recognizes which cross-section of the observed object is being photographed using a pre-trained model for cross-section recognition. The pre-trained model for cross-section recognition is generated, for example, by a deep learning CNN (Convolutional Neural Network). Note that the pre-trained model for cross-section recognition is not limited to a CNN and may be generated by other methods.

[0045] Figure 4 shows an example of the training state of a pre-trained model for cross-sectional recognition. Figure 5 shows an example of the operational state of a pre-trained model for cross-sectional recognition. During training, the neural network accepts a cross-sectional image, which is an ultrasound image in which any cross-section has been captured, as input on the input side, and information about the observed object indicating which object the cross-section contained in the cross-sectional image belongs to, as input on the output side, and also accepts input of parameters for identifying the cross-section.

[0046] During operation, the cross-section recognition function 173 inputs frame images contained in a partial video and parameters for identifying cross-sections into a pre-trained model for cross-section recognition. The pre-trained model for cross-section recognition outputs observation target information, which quantifies which observation target cross-section the cross-section contained in the frame image corresponds to.

[0047] Furthermore, the cross-sectional recognition function 173 adds up the numerical values ​​of the observation target information for each frame image included in the partial video. This allows the cross-sectional recognition function 173 to calculate the probability that the cross-sections included in the partial video correspond to the cross-sections of each observation target. In other words, the cross-sectional recognition function 173 calculates the confidence level indicating the probability that the observation targets included in the heart of subject P are included in the partial video. The cross-sectional recognition function 173 is an example of a calculation unit. Specifically, the cross-sectional recognition function 173 calculates numerical values ​​indicating the probability that the cross-sections included in the partial video correspond to the cross-section of the heart wall, the cross-section of the tricuspid valve, the cross-section of the mitral valve, and the cross-section of the aortic valve.

[0048] The B-mode image selection function 174 allows medical professionals to select a portion of a video for image diagnosis.

[0049] The B-mode image selection function 174 calculates the clarity of the observed object, which is one of the evaluation items for a partial video. Here, clarity is a value that indicates the degree to which the observed object is clearly displayed. In B-mode ultrasound images, the brightness value increases when the intensity of the reflected ultrasound waves transmitted by the ultrasound probe 101 is high. And in B-mode ultrasound images, the visibility of the observed object improves when the brightness value is high. In other words, when the brightness value is high, it can be said that the area of ​​the observed object is clearly displayed. Therefore, the B-mode image selection function 174 detects the area of ​​the observed object from the frame images of the partial video. The B-mode image selection function 174 also calculates the average brightness value of the detected area of ​​the observed object. Then, the B-mode image selection function 174 determines the clarity based on the average brightness value of the area of ​​the observed object. For example, the B-mode image selection function 174 may determine the clarity by the average brightness value, by the value obtained by substituting the average brightness value into a calculation formula, by the value corresponding to the average brightness value using a correspondence table, or by any other method. Furthermore, the B-mode image selection function 174 is not limited to the average value of the brightness, but may also be the median value of the brightness, the mode value, or any other value.

[0050] For example, the B-mode image selection function 174 detects the region to be observed using a pre-trained model for region detection. The pre-trained model for region detection is generated, for example, by a deep learning CNN. Note that the pre-trained model for region detection is not limited to a CNN and may be generated by other methods.

[0051] Figure 6 shows an example of the training state of a pre-trained model for region detection. Figure 7 shows an example of the operational state of a pre-trained model for region detection. During training, the neural network accepts a cross-sectional image, which is an ultrasound image containing the object being observed, as input on the input side, and segmentation information indicating the region of the object being observed contained in the cross-sectional image as input on the output side, and also accepts input of parameters for region detection.

[0052] During operation, the B-mode image selection function 174 inputs the frame images included in the partial video and parameters for detecting regions into a pre-trained model for region detection. The pre-trained model for region detection outputs segmentation information indicating the region to be observed within the frame image.

[0053] The B-mode image selection function 174 calculates the average value of the brightness within the observed area indicated by the segmentation information in the frame image. The B-mode image selection function 174 is performed on all frame images in the partial video. The B-mode image selection function 174 then ranks the images according to the average brightness value. In this way, the B-mode image selection function 174 obtains the ranking of the clarity of the observed object in the partial video. Note that the B-mode image selection function 174 may calculate the median, mode, or other values ​​in addition to the average brightness value.

[0054] The B-mode image selection function 174 selects a portion of the video for medical professionals to perform image diagnosis when calculating the clarity of the observed subject, which is one of the evaluation items. For example, the B-mode image selection function 174 selects a portion of the video based on the confidence level indicating the likelihood that the observed subject is included in that portion of the video. The B-mode image selection function 174 is an example of a selection unit.

[0055] Here, the evaluation criteria for the partial video are set in the application setting information 141. The B-mode image selection function 174 then selects the partial video based on the scores of each evaluation criterion set in the application setting information 141. In other words, the B-mode image selection function 174 selects the partial video based on the sum of the reliability score and the scores of the evaluation criteria for the partial video.

[0056] The evaluation criteria may include ranking the clarity of the observed subject in the partial video. The clarity of the observed subject is determined by its luminance value. In other words, the B-mode image selection function 174 may select a partial video based on its reliability and the luminance value of the observed subject P in the frame images of the partial video.

[0057] Furthermore, the B-mode image selection function 174 may select partial videos by taking into account weighting coefficients for evaluation items. That is, the B-mode image selection function 174 may select partial videos based on the sum of the confidence level and the scores of the evaluation items multiplied by the weighting coefficients. For example, the B-mode image selection function 174 multiplies the evaluation items indicated by the application setting information 141 by the weighting coefficients indicated by the weighting coefficient information 142. The B-mode image selection function 174 also adds up each of the evaluation items multiplied by the weighting coefficients. In this way, the B-mode image selection function 174 calculates the scores for partial videos. Then, the B-mode image selection function 174 selects partial videos for each cross-section of interest based on the scores calculated for each cross-section of interest. That is, the B-mode image selection function 174 selects the partial video with the highest score as the partial video for the cross-section of interest.

[0058] The Doppler image selection function 175 extracts Doppler diagnostic images, which are the parts of the ultrasound image that medical professionals use for diagnosis. More specifically, the Doppler image selection function 175 extracts Doppler diagnostic images from an ultrasound video when the ultrasound video contains Doppler data. For example, the Doppler image selection function 175 extracts Doppler diagnostic images from the ultrasound video, which are the parts of the ultrasound image that are displayed for diagnosis, based on the distance from the position of the Doppler cursor that measures the blood flow velocity of the subject P to the object of observation. The Doppler image selection function 175 is an example of an extraction unit.

[0059] For example, the Doppler image selection function 175 extracts the ultrasound video with the shortest distance from the Doppler cursor to the observation target from among ultrasound videos where the distance from the Doppler cursor to the observation target is less than a threshold, as the Doppler diagnostic partial image. The Doppler cursor is a cursor that indicates the position for measuring blood flow velocity.

[0060] The Doppler image selection function 175 detects the Doppler cursor and the object being observed from the ultrasound video. For example, the Doppler image selection function 175 detects the Doppler cursor and the object being observed using a pre-trained model for position detection. The pre-trained model for position detection is generated, for example, by deep learning YOLO (You Only Look Once). Note that the pre-trained model for position detection is not limited to YOLO and may be generated by other methods.

[0061] Figure 8 shows an example of the training state of a pre-trained model for position detection. Figure 9 shows an example of the operational state of a pre-trained model for position detection. During training, the neural network accepts a cross-sectional image, which is an ultrasound image containing the Doppler cursor and the object being observed, as input to the input side, and accepts Doppler cursor position information indicating the position of the Doppler cursor included in the cross-sectional image and object position information indicating the position of the object being observed included in the cross-sectional image as input to the output side, and accepts input of parameters for identifying the position of the object being detected.

[0062] During operation, the Doppler image selection function 175 inputs the ultrasound images included in the ultrasound video and parameters for identifying the location of the object to be detected into a pre-trained model for position detection. The pre-trained model for position detection outputs Doppler cursor position information, which indicates the position of the Doppler cursor included in the ultrasound image, and observation target position information, which indicates the position of the object being observed. The Doppler image selection function 175 may also detect the position of the Doppler cursor and the position of the object being observed using different pre-trained models.

[0063] The image measurement function 176 measures the object of observation of subject P. The image measurement function 176 is an example of a measurement unit. When an ultrasound video is acquired, the image measurement function 176 measures the blood flow of the object of observation based on the Doppler data. For example, as a first measurement process, the image measurement function 176 measures the tricuspid valve passage velocity, mitral valve passage velocity, aortic valve passage velocity, left ventricular outflow tract obstruction, regurgitation, etc.

[0064] For example, if the partial video includes a parasternal approach left ventricular long-axis section, the image measurement function 176 performs a second measurement process, measuring Global Logitudinal Strain, Transverse Strain, left ventricular ejection fraction, TAPSE, MAPSE, and dyssynchrony assessment.

[0065] For example, if the partial video includes a parasternal approach left ventricle short-axis section, the image measurement function 176 performs a third measurement process, measuring circumferential strains, radial strains, etc.

[0066] The display control function 177 displays various screens. For example, when the image acquisition function 171 acquires an ultrasound video, the display control function 177 displays the stress echocardiogram image G1, which is an arrangement of the partial videos selected by the B-mode image selection function 174.

[0067] Figure 10 shows an example of a stress echocardiogram image G1. The display control function 177 displays a stress echocardiogram image G1 which is created by dividing a partial video taken before the subject P was subjected to stress and an ultrasound video taken after the subject P was subjected to stress, and then arranging the partial video selected by the B-mode image selection function 174. The display control function 177 is an example of a display control unit. More specifically, the stress echocardiogram image G1 comprises a first target display area G111, a first partial video display area G112, a first electrocardiogram waveform display area G113, a second target display area G121, a second partial video display area G122, a second electrocardiogram waveform display area G123, a third target display area G131, a third partial video display area G132, a third electrocardiogram waveform display area G133, a fourth target display area G141, a fourth partial video display area G142, and a fourth electrocardiogram waveform display area G143.

[0068] The first target display area G111 is the area that displays the plane of interest corresponding to the partial video in the first partial video display area G112. The first partial video display area G112 is the area that displays the partial video. The first electrocardiogram waveform display area G113 is the area that displays the electrocardiogram waveform corresponding to the partial video. In the stress echocardiogram image G1 shown in Figure 10, the display control function 177 displays the first plane of interest before the stress in the first target display area G111, the first partial video display area G112, and the first electrocardiogram waveform display area G113.

[0069] Furthermore, the display control function 177 displays the first cross-section of interest after load in the second target display area G121, the second partial video display area G122, and the second electrocardiogram waveform display area G123. In other words, the display control function 177 displays the partial video selected by the B-mode image selection function 174 and the electrocardiogram waveform corresponding to the partial video.

[0070] The display control function 177 displays the second plane of interest before the load in the third target display area G131, the third partial video display area G132, and the third electrocardiogram waveform display area G133. The display control function 177 also displays the second plane of interest after the load in the fourth target display area G141, the fourth partial video display area G142, and the fourth electrocardiogram waveform display area G143. In other words, the display control function 177 displays the partial video selected by the B-mode image selection function 174 and the electrocardiogram waveform corresponding to the partial video. Thus, the display control function 177 displays the partial video of the first plane of interest after the load and the partial video of the second plane of interest even if a medical professional has not entered an operation to select a partial video.

[0071] Note that the stress echocardiogram image G1 shown in Figure 10 displays two planes of interest: the first plane of interest and the second plane of interest. However, the display control function 177 may display a stress echocardiogram image G1 with one plane of interest, or a stress echocardiogram image G1 with three planes of interest, or a stress echocardiogram image G1 with four or more planes of interest.

[0072] For example, when the image acquisition function 171 acquires Doppler video data, the display control function 177 displays the Doppler diagnostic partial image selected by the Doppler image selection function 175 and the Doppler examination image G2 which shows the Doppler data corresponding to the Doppler diagnostic partial image.

[0073] Figure 11 shows an example of a Doppler examination image G2. The Doppler examination image G2 has a Doppler diagnostic partial image display area G21 and a Doppler data display area G22. The display control function 177 displays the Doppler diagnostic partial image selected by the Doppler image selection function 175 in the Doppler diagnostic partial image display area G21. The Doppler diagnostic partial image includes a Doppler cursor image G211 showing the Doppler cursor and an observation target image G212 showing the observation target. In the Doppler diagnostic partial image selected by the Doppler image selection function 175, the display control function 177 displays a flow velocity waveform, which is a waveform representation of the blood flow velocity at the position of the Doppler cursor, in the Doppler data display area G22.

[0074] Next, we will explain the various processes performed by the ultrasound diagnostic device 1.

[0075] Figure 12 is a flowchart showing an example of a selection process performed by the ultrasound diagnostic device 1 according to the first embodiment.

[0076] The splitting function 172 performs image splitting processing to divide the ultrasound video into one or more partial videos (step S1).

[0077] The cross-section recognition function 173 performs the cross-section recognition process shown in Figure 13 (step S2).

[0078] The Doppler image selection function 175 determines whether or not the ultrasound video contains Doppler data (step S3).

[0079] If Doppler data is not available (Step S3; No), the B-mode image selection function 174 executes the B-mode image selection process described later (Step S4). The B-mode image selection process includes the first B-mode image selection process shown in Figure 14 and the second B-mode image selection process shown in Figure 15.

[0080] On the other hand, if Doppler data is available (Step S3; Yes), the Doppler image selection function 175 executes the Doppler image selection process shown in Figure 16 (Step S5).

[0081] Based on the above, the ultrasound diagnostic device 1 terminates the selection process.

[0082] Figure 13 is a flowchart showing an example of the cross-sectional recognition process performed by the ultrasound diagnostic device 1 according to the first embodiment.

[0083] The B-mode image selection function 174 determines whether or not the evaluation of all the partial videos to be evaluated has been completed (step S11).

[0084] If there are any parts of the video to be evaluated that have not yet been evaluated (Step S11; No), the B-mode image selection function 174 determines whether or not all frame images included in the parts of the video that have not yet been evaluated have been evaluated (Step S12).

[0085] If the evaluation of all frame images is complete (step S12; Yes), the B-mode image selection function 174 proceeds to step S11. This causes the B-mode image selection function 174 to evaluate any remaining video segments that have not yet been evaluated.

[0086] If the evaluation of all frame images is not complete (step S12; No), the B-mode image selection function 174 performs cross-sectional analysis on the frame images of the partial video that have not yet been evaluated (step S13). In other words, the B-mode image selection function 174 inputs the frame images to the trained model for cross-sectional recognition. As a result, the trained model for cross-sectional recognition outputs confidence information that quantifies the confidence level of which cross-section of the observed object the frame image corresponds to.

[0087] The B-mode image selection function 174 adds up the confidence values ​​of each of the multiple frame images that the partial video has (step S14). As a result, the B-mode image selection function 174 calculates the total confidence value of the multiple frame images that the partial video has. Then, the B-mode image selection function 174 proceeds to step S12.

[0088] In step S11, if there are no un-evaluated video segments among all the video segments to be evaluated, that is, if the evaluation of all video segments to be evaluated is complete (step S11; Yes), the B-mode image selection function 174 ranks the video segments to be evaluated in order according to their reliability (step S15).

[0089] As a result, the ultrasound diagnostic device 1 terminates the cross-sectional recognition process.

[0090] Figure 14 is a flowchart showing an example of a first B-mode image selection process performed by the ultrasound diagnostic device 1 according to the first embodiment.

[0091] The B-mode image selection function 174 determines whether or not the evaluation of all the partial videos to be evaluated has been completed (step S21).

[0092] If the evaluation of all partial videos is not complete (step S21; No), the B-mode image selection function 174 performs segmentation processing on each of the frame images in the partial video (step S22). That is, the B-mode image selection function 174 inputs the frame images to the trained model for region detection. As a result, the trained model for cross-sectional recognition detects each region of the object to be observed in the frame image.

[0093] The B-mode image selection function 174 calculates the average brightness value of the first observation target region in the frame image (step S23). The B-mode image selection function 174 also determines the sharpness based on the average brightness value of the first observation target region.

[0094] The B-mode image selection function 174 calculates the average brightness value of the second observation target region in the frame image (step S24). The B-mode image selection function 174 also determines the sharpness based on the average brightness value of the second observation target region.

[0095] The B-mode image selection function 174 calculates the average brightness value of the third observation target region in the frame image (step S25). The B-mode image selection function 174 also determines the sharpness based on the average brightness value of the third observation target region.

[0096] The B-mode image selection function 174 calculates the average brightness value of the fourth observation target region in the frame image (step S26). The B-mode image selection function 174 also determines the sharpness based on the average brightness value of the fourth observation target region. Then, the B-mode image selection function 174 proceeds to step S21. In the flowchart shown in Figure 14, the processes in steps S23 to S26 are executed in the order of the first observation target, second observation target, third observation target, and fourth observation target, but this order is just an example and can be changed arbitrarily. Also, the processes in each step may be executed simultaneously by parallel processing or the like.

[0097] In step S21, if the evaluation of all partial videos is completed (step S21; Yes), the B-mode image selection function 174 ranks the partial videos to be evaluated in order according to the clarity of the first observation target area (step S27).

[0098] The B-mode image selection function 174 ranks the partial video to be evaluated in order according to the clarity of the second observation target area (step S28).

[0099] The B-mode image selection function 174 ranks the partial video to be evaluated in order according to the clarity of the third observation target area (step S29).

[0100] The B-mode image selection function 174 ranks the partial video to be evaluated in order of clarity according to the region of the fourth observation target (step S30). In the flowchart shown in Figure 14, the processes in steps S27 to S30 are performed in the order of the first observation target, second observation target, third observation target, and fourth observation target, but this order is just an example and can be changed arbitrarily. Also, the processes in each step may be performed simultaneously by parallel processing or the like.

[0101] As a result, the ultrasound diagnostic device 1 terminates the first B-mode image selection process.

[0102] Figure 15 is a flowchart showing an example of a second B-mode image selection process performed by the ultrasound diagnostic apparatus 1 according to the first embodiment. The second B-mode image selection process selects partial videos based on the processing results of the first B-mode image selection process shown in Figure 14. For example, the second B-mode image selection process selects partial videos based on the clarity ranking determined by the first B-mode image selection process.

[0103] The B-mode image selection function 174 reads the settings indicated by the applied setting information 141 and the weight coefficient information 142 (step S41).

[0104] The B-mode image selection function 174 determines whether or not partial videos have been selected for all sections of interest (step S42). If no partial videos have been selected (step S42; No), the B-mode image selection function 174 determines whether or not the scores for all partial videos have been calculated (step S43).

[0105] If the calculation of scores for all partial videos has not been completed (Step S43; No), the B-mode image selection function 174 calculates the score for each partial video using the calculation formula for the first section of interest (Step S44). That is, the B-mode image selection function 174 calculates the score using a calculation formula based on the application setting information 141 and the weighting coefficient information 142.

[0106] More specifically, the B-mode image selection function 174 multiplies the evaluation items indicated by the application setting information 141 by the weight coefficients indicated by the weight coefficient information 142. The B-mode image selection function 174 then calculates a score by adding up the multiplied values. For example, the B-mode image selection function 174 calculates the score using the formula: Score = Confidence (Rank) × Weight Coefficient + Heart Rate (Rank) × Weight Coefficient + Rank of the clarity of the first observed object × Weight Coefficient + Rank of the clarity of the second observed object × Weight Coefficient + Rank of the clarity of the third observed object × Weight Coefficient + Rank of the clarity of the fourth observed object × Weight Coefficient. Note that for parameters related to rank, the score increases as the rank increases. Also, the rank of clarity may be a rank corresponding to the average brightness value.

[0107] The B-mode image selection function 174 calculates the score for each partial video using a calculation formula for the second section of interest (step S45). That is, the B-mode image selection function 174 calculates the score using a calculation formula based on the applied setting information 141 and the weighting coefficient information 142. For example, the B-mode image selection function 174 calculates the score using the formula: Score = Confidence (rank) × Weighting coefficient + Heart rate (rank) × Weighting coefficient + Rank of clarity of the first observed object × Weighting coefficient + Rank of clarity of the third observed object × Weighting coefficient + Rank of clarity of the fourth observed object × Weighting coefficient.

[0108] The B-mode image selection function 174 calculates the score for each partial video using a calculation formula for the third plane of interest (step S46). That is, the B-mode image selection function 174 calculates the score using a calculation formula based on the applied setting information 141 and the weighting coefficient information 142. For example, the B-mode image selection function 174 calculates the score using the formula: Score = Confidence (rank) × Weighting coefficient + Heart rate (rank) × Weighting coefficient + Rank of clarity of the first observed object × Weighting coefficient + Rank of clarity of the third observed object × Weighting coefficient + Rank of clarity of the fourth observed object × Weighting coefficient.

[0109] The B-mode image selection function 174 calculates the score for each partial video using a calculation formula for the fourth plane of interest (step S47). That is, the B-mode image selection function 174 calculates the score using a calculation formula based on the application setting information 141 and the weight coefficient information 142. For example, the B-mode image selection function 174 calculates the score using the formula: Score = Confidence (rank) × Weight coefficient + Heart rate (rank) × Weight coefficient + Rank of clarity of the first observed object × Weight coefficient. In the flowchart shown in Figure 15, the processes are executed in the order of the first plane of interest, second plane of interest, third plane of interest, and fourth plane of interest from step S44 to step S47, but this order is just an example and can be changed arbitrarily. Also, the processes in each step may be executed simultaneously by parallel processing or the like.

[0110] If the calculation of scores for all partial videos is complete (step S43; Yes), the B-mode image selection function 174 selects a partial video for the first section of interest based on the scores (step S48). That is, the B-mode image selection function 174 selects the partial video with the highest score among the scores calculated for the first section of interest.

[0111] The B-mode image selection function 174 selects a partial video for the second section of interest based on the score (step S49). That is, the B-mode image selection function 174 selects the partial video with the highest score among the scores calculated for the second section of interest.

[0112] The B-mode image selection function 174 selects a partial video for the third section of interest based on the score (step S50). That is, the B-mode image selection function 174 selects the partial video with the highest score among the scores calculated for the third section of interest.

[0113] The B-mode image selection function 174 selects a partial video for the fourth section of interest based on the score (step S51). That is, the B-mode image selection function 174 selects the partial video with the highest score among the scores calculated for the fourth section of interest. In the flowchart shown in Figure 15, the processes are executed in the order of the first section of interest, second section of interest, third section of interest, and fourth section of interest from step S48 to step S51, but this order is just an example and can be changed arbitrarily. Also, the processes in each step may be executed simultaneously by parallel processing or the like.

[0114] If the selection of a partial video is complete (step S42; Yes), the B-mode image selection function 174 terminates the second B-mode image selection process.

[0115] Figure 16 is a flowchart showing an example of the Doppler image selection process performed by the ultrasound diagnostic device 1 according to the first embodiment.

[0116] The Doppler image selection function 175 determines whether or not the judgment has been completed for each ultrasound image included in the ultrasound video containing Doppler data (step S61).

[0117] If the determination is not complete (step S61; No), the Doppler image selection function 175 obtains the position of the Doppler cursor and the position of each observation target included in the ultrasound image (step S62). For example, the Doppler image selection function 175 inputs the ultrasound image to a trained model for position detection. As a result, the trained model for position detection outputs Doppler cursor position information indicating the position of the Doppler cursor and observation target position information indicating the position of the observation target included in the cross-sectional image.

[0118] The Doppler image selection function 175 determines, based on the position of the Doppler cursor included in the ultrasound image and the position of the first observation target, whether the distance from the first observation target to the Doppler cursor is less than a threshold and whether the distance from the first observation target to the Doppler cursor is the shortest possible distance (step S63).

[0119] If the distance from the first observation target to the Doppler cursor is less than the threshold, and the distance from the first observation target to the Doppler cursor is the shortest possible distance (Step S63; Yes), the Doppler image selection function 175 selects the ultrasound image to be judged as the Doppler diagnostic partial image of the first observation target (Step S64).

[0120] If the distance from the first observation target to the Doppler cursor is less than a threshold and the distance from the first observation target to the Doppler cursor is not the shortest distance (Step S63; No), the Doppler image selection function 175 determines, based on the position of the Doppler cursor included in the ultrasound image and the position of the second observation target, whether the distance from the second observation target to the Doppler cursor is less than a threshold and whether the distance from the second observation target to the Doppler cursor is the shortest distance (Step S65).

[0121] If the distance from the second observation target to the Doppler cursor is less than the threshold, and the distance from the second observation target to the Doppler cursor is the shortest possible distance (Step S65; Yes), the Doppler image selection function 175 selects the ultrasound image to be judged as the Doppler diagnostic partial image of the second observation target (Step S66).

[0122] If the distance from the second observation target to the Doppler cursor is less than the threshold and the distance from the second observation target to the Doppler cursor is not the shortest distance (Step S65; No), the Doppler image selection function 175 determines, based on the position of the Doppler cursor included in the ultrasound image and the position of the third observation target, whether the distance from the third observation target to the Doppler cursor is less than the threshold and whether the distance from the third observation target to the Doppler cursor is the shortest distance (Step S67).

[0123] If the distance from the third observation target to the Doppler cursor is less than the threshold, and the distance from the third observation target to the Doppler cursor is the shortest possible distance (Step S67; Yes), the Doppler image selection function 175 selects the ultrasound image to be judged as the Doppler diagnostic partial image of the third observation target (Step S68).

[0124] If the distance from the third observation target to the Doppler cursor is less than the threshold and the distance from the third observation target to the Doppler cursor is not the shortest distance (Step S67; No), the Doppler image selection function 175 determines, based on the position of the Doppler cursor included in the ultrasound image and the position of the fourth observation target, whether the distance from the fourth observation target to the Doppler cursor is less than the threshold and whether the distance from the fourth observation target to the Doppler cursor is the shortest distance (Step S69).

[0125] If the distance from the fourth observation target to the Doppler cursor meets the condition (step S69; Yes), the Doppler image selection function 175 selects the ultrasound image to be judged as the Doppler diagnostic partial image of the fourth observation target (step S70). If the distance from the fourth observation target to the Doppler cursor does not meet the condition (step S69; No), the Doppler image selection function 175 proceeds to step S61.

[0126] If the determination is complete (step S61; Yes), the Doppler image selection function 175 terminates the Doppler image selection process.

[0127] Figure 17 is a flowchart showing an example of a measurement process performed by the ultrasound diagnostic device 1 according to the first embodiment.

[0128] The image measurement function 176 determines whether or not the ultrasound video contains Doppler data (step S81).

[0129] If Doppler data is available (step S81; Yes), the image measurement function 176 performs the first measurement process (step S82).

[0130] If Doppler data is not available (Step S81; No), the image measurement function 176 determines whether the partial video includes a parasternal approach left ventricular long-axis view (Step S83).

[0131] If a parasternal approach left ventricular long-axis view is included (step S83; Yes), the image measurement function 176 performs a second measurement process (step S84).

[0132] If the parasternal approach left ventricle long-axis view is not included (step S83; No), the image measurement function 176 determines whether or not the partial video includes a parasternal approach left ventricle short-axis view (step S85).

[0133] If the parasternal approach left ventricle short-axis section is not included (step S85; No), the image measurement function 176 terminates the measurement process.

[0134] If a parasternal approach left ventricle short-axis section is included (step S85; Yes), the image measurement function 176 performs a third measurement process (step S86). Then, the image measurement function 176 terminates the measurement process.

[0135] As described above, the ultrasound diagnostic device 1 according to this embodiment acquires ultrasound videos by switching the scanning cross-section of the subject P's heart. The ultrasound diagnostic device 1 also divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform indicated by the electrocardiogram waveform information associated with the ultrasound video. The ultrasound diagnostic device 1 also calculates a confidence level indicating the likelihood that the object of observation included in the subject P's heart is included in the partial video. Then, the ultrasound diagnostic device 1 selects a partial video based on the confidence level. In this way, the ultrasound diagnostic device 1 selects a partial video that is the target of the image diagnosis performed by the medical professional, even if the medical professional has not selected a partial video. Therefore, the ultrasound diagnostic device 1 can assist in acquiring a video of the portion suitable for stress echocardiography.

[0136] In the above embodiment, the ranking of the clarity of the observed objects in the partial video was described as an example of an evaluation item for the partial video. However, the evaluation item for the partial video is not limited to the ranking of the clarity of the observed objects in the partial video; it may also be a value indicating clarity, or a value derived based on clarity.

[0137] According to at least one embodiment described above, it is possible to support the acquisition of video footage of the portion suitable for stress echocardiography.

[0138] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be implemented in a variety of other forms, and various omissions, substitutions, modifications, and combinations of embodiments are possible without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]

[0139] 1. Ultrasound diagnostic equipment 101 Ultrasound probe 104 Main unit of the device 105 Electrode for electrocardiograph 140 Memory circuit 141 Application Settings Information 142 Weighting coefficient information 150 NW (Network) Interfaces 160 ECG Unit 170 Processing Circuits 171 Image acquisition function 172-way split function 173 Section recognition function 174 B-mode image selection function 175 Doppler Image Selection Function 176 Image Measurement Function 177 Display Control Function G1 Exercise Echocardiography Images G2 Doppler scan image P Subject

Claims

1. A first acquisition unit that acquires an ultrasound video taken while switching scanning cross-sections of the heart of a subject, A division unit that divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of the subject, A calculation unit calculates a confidence level indicating the likelihood that the partial video contains a cross-section of the object to be observed, based on the degree to which each of the multiple frame images included in the partial video contains a cross-section of the object to be observed, which is included in the heart of the subject. A selection unit that selects the partial video for each cross-section of the object to be observed based on the reliability, A display control unit that displays side by side a partial video taken before the subject was subjected to a load and the partial video taken after the subject was subjected to a load, which is divided from the ultrasound video and selected in the selection unit. An information processing device equipped with the following features.

2. A first acquisition unit acquires ultrasound video while switching between scanning cross-sections of the subject's heart, A division unit that divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of the subject, A calculation unit that calculates the confidence level indicating the possibility that the object to be observed, which is included in the heart of the subject, is included in the partial video, A selection unit that selects the partial video based on the reliability and the scores of the evaluation items for the partial video, An information processing device equipped with the following features.

3. The selection unit selects the partial video based on the sum of the reliability score and the score of the evaluation item for the partial video. The information processing apparatus according to claim 2.

4. The selection unit selects the partial video based on the sum of the confidence level and the score of the evaluation item multiplied by the weight coefficient. The information processing apparatus according to claim 3.

5. A first acquisition unit acquires ultrasound video while switching between scanning cross-sections of the subject's heart, A division unit that divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of the subject, A calculation unit that calculates the confidence level indicating the possibility that the object to be observed, which is included in the heart of the subject, is included in the partial video, A selection unit that selects the partial video based on the aforementioned reliability and the brightness value of the subject being observed in the frame image of the partial video, An information processing device equipped with the following features.

6. A first acquisition unit acquires ultrasound video while switching between scanning cross-sections of the subject's heart, A second acquisition unit that acquires the ultrasound video including Doppler data relating to the blood flow velocity of the subject, A division unit that divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of the subject, A calculation unit that calculates the confidence level indicating the possibility that the object to be observed, which is included in the heart of the subject, is included in the partial video, A selection unit that selects the partial video based on the aforementioned reliability, An extraction unit extracts an image to be displayed from the ultrasonic video based on the distance from the position where the flow velocity is measured to the object to be observed, which is included in the ultrasonic image of the ultrasonic video. An information processing device equipped with the following features.

7. The system further includes a measuring unit for measuring the object of observation of the subject. An information processing apparatus according to any one of claims 1 to 6.

8. An ultrasound probe that receives reflected waves from the subject, A first acquisition unit acquires ultrasound video while switching between scanning cross-sections of the subject's heart, A division unit that divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of the subject, A calculation unit that calculates the confidence level indicating the possibility that the object to be observed, which is included in the heart of the subject, is included in the partial video, A selection unit that selects the partial video based on the reliability and the scores of the evaluation items for the partial video, An ultrasound diagnostic device equipped with the following features.

9. Computers, A first acquisition unit acquires ultrasound video while switching between scanning cross-sections of the subject's heart, A division unit that divides the ultrasound video into multiple partial videos based on the electrocardiogram waveform of the subject, A calculation unit that calculates the confidence level indicating the possibility that the object to be observed, which is included in the heart of the subject, is included in the partial video, A selection unit that selects the partial video based on the reliability and the scores of the evaluation items for the partial video, A program to make it work.

Images