Medical information processing apparatus, ultrasound diagnostic apparatus, and medical information processing method

Abstract

To improve image quality of an ultrasound image while avoiding generation of an artifact due to body movement.SOLUTION: A medical information processing apparatus of an embodiment comprises an acquisition unit, a detection unit, a selection unit, and a synthesis processing unit. The acquisition unit acquires a plurality of pieces of ultrasound data representing frames of a subject that are consecutive in a time direction. The detection unit detects movement of the subject. The selection unit selects, based on a detection result of the movement, a plurality of pieces of frame data representing synthesis targets from among the plurality of pieces of ultrasound data. The synthesis processing unit performs synthesis processing on the selected plurality of pieces of frame data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The embodiments disclosed in this specification relate to a medical information processing apparatus, an ultrasound diagnostic apparatus, and a medical information processing method. [Background technology]

[0002] Conventionally, techniques for improving the image quality of ultrasound images containing minute flow paths and fluids (e.g., blood flow, contrast agents, etc.) have been known. For example, Patent Document 3 discloses a method for generating an image showing the boundary of a flow path (e.g., the shape of a blood vessel) by removing clutter contained in successive ultrasound reception signals within a predetermined time period and averaging the fluid signals after clutter removal over the predetermined time period. Furthermore, Patent Document 2 discloses a method for improving the resolution (pixel density) of an ultrasound image, performing super-resolution processing to sharpen the peaks of speckle patterns in the ultrasound image, and superimposing multiple speckle patterns with sharpened peaks over a predetermined time period to display a fluid flow path with high image quality.

[0003] With the above-mentioned technology, the more frames used, the higher the quality of the image that can be generated. However, the more frames used, the longer the time it takes to collect the frame data, which increases the possibility of subject movement causing artifacts in the image. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-20881 [Patent Document 2] Special Publication No. 2019-526350 [Patent Document 3] Japanese Patent Application Laid-Open No. 2001-178720 Summary of the Invention [Problem to be solved by the invention]

[0005] One of the problems that the embodiments disclosed herein aim to solve is to improve the image quality of ultrasound images while avoiding the occurrence of artifacts due to body movement. However, the problems solved by the embodiments disclosed herein are not limited to the above problem. Problems corresponding to the effects of the configurations described in the embodiments below can also be considered as other problems solved by the embodiments disclosed herein. [Means for solving the problem]

[0006] A medical image processing apparatus according to an embodiment includes an acquisition unit, a detection unit, a selection unit, and a synthesis processing unit. The acquisition unit acquires a plurality of ultrasound data representing frames of a subject that are consecutive in the time direction. The detection unit detects motion of the subject. The selection unit selects a plurality of frame data representing a synthesis target from the plurality of ultrasound data based on a result of the motion detection. The synthesis processing unit performs synthesis processing on the selected plurality of frame data. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of an ultrasound diagnostic apparatus according to the first embodiment. [Figure 2] FIG. 2 is a flowchart showing the procedure of processing performed by the ultrasonic diagnostic apparatus according to the first embodiment. [Figure 3] FIG. 3 is a diagram for explaining motion detection according to the first embodiment. [Figure 4A] FIG. 4A is a diagram illustrating an example of the selection process according to the first embodiment. [Figure 4B] FIG. 4B is a diagram illustrating an example of the selection process according to the first embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of data processing according to the first embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of the extraction process according to the first embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of a processing result according to the comparative example. [Figure 8] FIG. 8 is a diagram illustrating an example of data processing according to the second embodiment. [Figure 9] FIG. 9 is a diagram for explaining an example of data processing according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of a medical information processing device, an ultrasound diagnostic device, and a medical information processing method according to the present application will be described in detail with reference to the accompanying drawings. Note that the medical information processing device, the ultrasound diagnostic device, and the medical information processing method according to the present application are not limited to the embodiments shown below.

[0009] (First embodiment) Fig. 1 is a block diagram showing an example of the configuration of an ultrasound diagnostic device 10 according to the first embodiment. The ultrasound diagnostic device 10 is a device that generates ultrasound data based on received signals (reflected wave signals) received from an ultrasound probe 5. The ultrasound diagnostic device 10 shown in Fig. 1 is a device that can generate two-dimensional ultrasound data based on received signals received from a one-dimensional ultrasound probe 5 (an ultrasound probe in which transducers are arranged in one dimension), and can generate three-dimensional ultrasound data based on received signals received from a two-dimensional ultrasound probe 5 (an ultrasound probe in which transducers are arranged in two dimensions).

[0010] The ultrasonic probe 5 is, for example, an electronic scanning probe, and has a plurality of transducers 101 arranged one-dimensionally or two-dimensionally at its tip. The transducers 101 are piezoelectric elements (electromechanical transducers) that convert between electrical signals (voltage pulse signals) and ultrasound waves (acoustic waves). The ultrasonic probe 5 transmits ultrasound waves from the plurality of transducers 101 to a subject and receives reflected ultrasound waves from the subject via the plurality of transducers 101. The reflected acoustic waves reflect differences in acoustic impedance within the subject. When a transmitted ultrasound pulse is reflected by the surface of a moving blood flow, heart wall, or the like, the reflected ultrasound undergoes a frequency shift due to the Doppler effect, depending on the velocity signal component of the moving object relative to the ultrasound transmission direction.

[0011] The probe connection unit 103 connects to the ultrasonic probe 5 and transmits and receives ultrasonic waves to and from the ultrasonic probe 5. The connection means of the ultrasonic probe 5 by the probe connection unit 103 may be either wired or wireless. In the wired case, the probe connection unit 103 has a connector unit (receptacle) for connecting the connector (plug) of the ultrasonic probe 5. In the wireless case, it has a communication unit for wireless communication with the ultrasonic probe 5.

[0012] The ultrasonic diagnostic device 10 includes a transmission circuit 9, a reception circuit 11, and a medical information processing device 100.

[0013] The transmission circuit 9 is a transmission unit that outputs pulse signals (drive signals) to the multiple transducers 101. By applying pulse signals to the multiple transducers 101 with a time difference, ultrasonic waves with different delay times are transmitted from the multiple transducers 101, thereby forming a transmitted ultrasonic beam. The direction and focus of the transmitted ultrasonic beam can be controlled by selectively changing the transducer 101 to which the pulse signal is applied (i.e., the transducer 101 to be driven) or by changing the delay time (application timing) of the pulse signal. By sequentially changing the direction and focus of this transmitted ultrasonic beam, an observation area inside the subject is scanned. Furthermore, by changing the delay time of the pulse signal, a transmitted ultrasonic beam that is a plane wave (focused at a distance) or a diverging wave (focus point is in the opposite direction of the ultrasonic transmission direction for the multiple transducers 101) may be formed. Alternatively, a transmitted ultrasonic beam may be formed using one transducer or some of the multiple transducers 101. The transmission circuit 9 transmits a pulse signal with a predetermined drive waveform to the transducer 101, causing the transducer 101 to generate a transmitted ultrasonic wave having a predetermined transmission waveform.

[0014] The receiving circuit 11 is a receiving unit that inputs, as a received signal, an electrical signal output from the transducer 101 that has received the reflected ultrasound. The received signal is input to the processing circuit 110. In this embodiment, the analog signal output from the transducer 101 and the digital data obtained by sampling (digital conversion) the analog signal are both referred to as the received signal without any particular distinction.

[0015] The medical information processing device 100 is connected to a transmission circuit 9 and a reception circuit 11, and processes signals received from the reception circuit 11 and controls the transmission circuit 9. The medical information processing device 100 includes a processing circuit 110, a memory 120, an input device 130, and a display 140.

[0016] The memory 120 is composed of a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, etc. The memory 120 is a memory for storing data such as image data for display generated by the processing circuit 110. The memory 120 can also store received signals (reflected wave signals) output by the receiving circuit 11. In addition, the memory 120 stores, as necessary, control programs for transmitting and receiving ultrasound, image processing, and display processing, as well as various data such as diagnostic information (e.g., patient ID, doctor's findings, etc.), diagnostic protocols, and various body marks.

[0017] The input device 130 accepts various instructions and information input from an operator. The input device 130 is composed of input interface devices such as a trackball, switch buttons, a mouse, a keyboard, a touchpad that performs input operations by touching the operation surface, a touch monitor that integrates a display screen and a touchpad, a non-contact input circuit using an optical sensor, and a voice input circuit. The input interface devices are connected to the processing circuit 110 (described later) and convert input operations received from the operator into electrical signals and output them to the processing circuit 110. In this specification, the input interface device is not limited to devices equipped with physical operating components such as a mouse and a keyboard. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the device and outputs the electrical signals to the processing circuit 110 is also included as an example of an input interface device.

[0018] The display 140 displays a GUI (Graphical User Interface) for receiving input of imaging conditions and various images under the control of the processing circuitry 110. The display 135 is configured by a display interface device such as a liquid crystal display, for example.

[0019] The processing circuitry 110 controls each unit of the ultrasound diagnostic apparatus 10, thereby controlling the entire ultrasound diagnostic apparatus 10. For example, the processing circuitry 110 executes a program stored in the memory 120, thereby functioning as a control function 111, a signal processing function 112, a selection function 113, a synthesis processing function 114, an extraction function 115, and an output function 116. The processing circuitry 110 is realized, for example, by a processor. Here, the control function 111 is an example of an execution unit and an acquisition unit. The signal processing function 112 is an example of a detection unit. The selection function 113 is an example of a selection unit. The synthesis processing function 114 is an example of a synthesis processing unit. The extraction function 115 is an example of an extraction unit. Note that, although the processing circuitry 110 is described in FIG. 1 as being realized by a single unit, it may also be realized by combining multiple independent processors. Furthermore, specific functions may be configured by dedicated independent circuits, such as an ASIC (Application Specific Integrated Circuit).

[0020] Furthermore, the term "processor" used in the above description refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphical Processing Unit), an Application Specific Integrated Circuit (ASIC), 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 realizes its functions by reading and executing programs stored in memory 132.

[0021] The ultrasound diagnostic device 10 configured as described above improves the image quality of ultrasound images while avoiding the occurrence of artifacts due to body movement. The operation of the ultrasound diagnostic device 10 will be described in detail below with reference to Fig. 2. Fig. 2 is a flowchart showing the processing procedure of the ultrasound diagnostic device 10 according to the first embodiment.

[0022] The control function 111 acquires a plurality of ultrasound data each representing a frame of the subject. Specifically, the control function 111 controls the transmission circuitry 9 and the reception circuitry 11 to cause the ultrasound probe 5 to execute an ultrasound scan and acquire a plurality of frames of received signals (e.g., CH data) (step S101). More specifically, the control function 111 collects a plurality of frame data (a plurality of frame data within a predetermined time period) that are continuous in the time direction and obtained by executing an ultrasound scan, at a predetermined frame rate. The frame data includes amplitude values ​​(signal values) of signals corresponding to each pixel that constitutes one image (frame).

[0023] The signal processing function 112 performs delay-and-sum and quadrature detection processing on the collected received signals. The delay-and-sum processing is a process of adding together received signals from multiple transducers 101 by changing the delay time and weight for each transducer 101, and is also called DAS (Delay and Sum) beamforming. The quadrature detection processing is a process of converting received signals into in-phase and quadrature signals in the baseband to obtain measurement data representing IQ data and the absolute values ​​of the IQ data. Note that other processes such as adaptive beamforming, model-based processing, and machine learning may also be performed on the received signals. The signal processing function 112 also performs envelope detection processing, logarithmic compression processing, and the like to generate B-mode data that represents the signal intensity at each point in the observation region as brightness.

[0024] Furthermore, the signal processing function 112 detects the movement of the subject based on the plurality of ultrasound data (step S102). Specifically, the signal processing function 112 detects frame data in which the position or orientation of the object depicted in the image has changed significantly from the plurality of frame data acquired by the control function 111. For example, as shown in FIG. 3, the signal processing function 112 performs detection processing on frames f1 to f8 in the frame data 20, and detects frame f6 as frame data in which the position or orientation of the object depicted in the image has changed significantly. That is, the signal processing function 112 detects that movement has occurred in the subject between frames f5 and f6. Note that FIG. 3 is a diagram for explaining movement detection according to the first embodiment.

[0025] Here, the signal processing function 112 can detect the subject's motion using various methods. For example, the signal processing function 112 calculates an image quality evaluation index for each frame of multiple ultrasound data, and detects the subject's motion based on the calculated image quality evaluation index. In this case, the signal processing function 112 calculates the image quality evaluation index for each frame using inter-frame difference, optical flow, template matching, etc., and detects the subject's motion based on the calculated image quality evaluation index.

[0026] For example, when using inter-frame differences, the signal processing function 112 calculates the inter-frame differences between temporally adjacent frames (or between a reference frame and a frame to be analyzed) to calculate the amount of movement in each image, and detects frames in which the amount of movement exceeds a threshold as frames in which movement has occurred in the subject.

[0027] Furthermore, for example, when optical flow is used, the signal processing function 112 calculates a displacement vector indicating the movement of an object (e.g., an element having the same color) between temporally adjacent frames, and identifies a transition of the amount of movement between the multiple frames based on the calculated displacement vector. Then, the image processing function 215 detects the movement of the subject by extracting the timing at which the amount of movement exceeds a threshold value in the transition of the amount of movement.

[0028] Furthermore, for example, when template matching is used, the signal processing function 112 acquires a template image (e.g., an image including a part of an object) from a reference frame in a plurality of frame data. Furthermore, the signal processing function 112 sets a search area (e.g., an area within a predetermined range from the area where the template image was acquired) in each frame to search for a portion similar to the template image. The signal processing function 112 then performs template matching using the template image for the search area, calculates the similarity between each position in the search area and the template image, and identifies the position with the highest similarity. The signal processing function 112 identifies the position with the highest similarity in each frame, and detects frames in which the similarity at the identified position is below a threshold as frames in which movement of the subject has occurred.

[0029] Furthermore, for example, the signal processing function 112 can perform a principal component analysis on each frame of a plurality of ultrasound data, and detect the movement of the subject based on the analysis results. In this case, the signal processing function 112 can perform a principal component analysis on each frame to detect the movement of each subject. The characteristics of each frame are analyzed, and the signal processing function 112 detects frames in which a significant change occurs in the analysis results as frames in which movement has occurred in the subject.

[0030] Furthermore, for example, the signal processing function 112 detects the movement of the subject by inputting each frame of the plurality of ultrasound data to a trained model trained with a data set including training ultrasound data and information regarding the presence or absence of movement of the subject in the training ultrasound data. In this case, first, a trained model is prepared that has been trained using training data including a plurality of ultrasound data including minute flow paths and fluids (e.g., blood flow, contrast agent, etc.) and the presence or absence of movement of the subject in the plurality of ultrasound data.

[0031] For example, a trained model is prepared that outputs whether or not motion has occurred in response to input of multiple frame data by training using multiple sets of frame data in which no motion has occurred between adjacent frame data and multiple sets of frame data in which motion has occurred between frame data. The signal processing function 112 inputs multiple frame data acquired by the control function 111 into the trained model to detect the motion of the subject.

[0032] As described above, the signal processing function 112 detects the subject's motion using various methods, but the method for detecting the subject's motion is not limited to the above methods. In other words, any method may be used as long as it can detect the subject's motion.

[0033] The selection function 113 selects frame data representing frames to be synthesized from the plurality of ultrasound data based on the detection result of the subject's movement (step S103). Specifically, the selection function 113 selects, from the plurality of ultrasound data, a plurality of frame data in which the subject's movement is relatively small, as frame data representing frames to be synthesized. For example, as shown in FIG. 4A, the selection function 113 selects frames f1 to f5 from frames f1 to f8 as frames to be synthesized. Note that FIG. 4A is a diagram showing an example of the selection process according to the first embodiment.

[0034] Here, the selection function 113 can perform various selection processes depending on the situation in which a high-quality image is to be generated. Specifically, the selection function 113 performs different selection processes depending on whether a high-quality image is to be generated after an ultrasound scan is completed, or after a set number of frames of ultrasound data are acquired during an ultrasound scan, or whether a high-quality image is to be generated while ultrasound data is being acquired by an ultrasound scan.

[0035] When generating a high-quality image by extracting and synthesizing signals from multiple frame data, the more frames used, the higher the quality of the generated image. Therefore, when generating a high-quality image after an ultrasound scan is completed, or when generating a high-quality image after acquiring a predetermined number of frames of ultrasound data during an ultrasound scan, the selection function 113 selects, from the set number of frames of ultrasound data, the frame data with the larger number of frames, with the boundary at the time point when the subject's movement is detected by the signal processing function 112, as the frame data representing the frame to be synthesized.

[0036] For example, as shown in FIG. 4A, if one set for generating a high-quality image is set to eight frames, and the signal processing function 112 detects the occurrence of motion between frames f5 and f6, the selection function 113 selects frames f1 to f5, which have the largest number of frames with the point at which the motion occurred as the boundary, as frames to be synthesized. If, for example, motion is detected between frames f2 and f3 in FIG. 4A, the selection function 113 selects frames f3 to f8 as frames to be synthesized. If no subject motion is detected in frames f1 to f8, the selection function 113 selects frames f1 to f8 as frames to be synthesized. For convenience of explanation, FIG. 4A shows a case where eight frames are set as the number of frames for generating a high-quality image, but in reality, the number of frames (number of packets) is set to, for example, several tens to several hundred.

[0037] On the other hand, when generating a high-quality image while acquiring ultrasound data through an ultrasound scan, the selection function 113 selects, from the multiple ultrasound data acquired over time, frame data acquired before the point at which the subject's movement was detected by the signal processing function 112, as frame data representing the frame to be synthesized.

[0038] For example, when the frame data 20 shown in Fig. 4A is acquired in order starting from frame f1, the signal processing function 112 determines whether or not there is movement of the subject each time frame data is acquired. That is, after acquiring frame f1, the signal processing function 112 determines whether or not there is movement of the subject each time frame data from frame f2 onwards is acquired. Here, in Fig. 4A, since the occurrence of movement is detected in frame f6, the selection function 113 selects frames f1 to f5 up until the occurrence of movement is detected as frames to be synthesized.

[0039] In addition, if no movement of the subject is detected from frame f1 to frame f8 in Figure 4A, the selection function 113 selects frames f1 to f8 as frames to be synthesized when the number of acquired frames reaches the set number of frames, which is 8 frames (i.e., when frame f8 is acquired).

[0040] 4A, when motion is detected between frames f5 and f6, the signal processing function 112 uses frame f6 as a reference and detects whether or not there is motion in subsequent frames (frames f7, f8, and frames f9 and subsequent frames not shown). The selection function 113 selects frames to be synthesized by performing the above-mentioned selection process on frames f6 and subsequent frames.

[0041] As described above, the greater the number of frames to be synthesized, the higher the image quality that can be generated. Therefore, in order to select as many frames as possible as frame data to be synthesized, the selection function 113 can select, as frame data representing frames to be synthesized, multiple frame data that have a similarity equal to or greater than a reference level between frames represented by multiple ultrasound data.

[0042] For example, as shown in FIG. 4B , after a movement occurs between frames f5 and f6, another movement may occur between frames f7 and f8, causing the rendering state of the object in frame f8 to return to a state similar to that of the object in frame f5. In such a case, in order to select as many frames as possible as the frame data to be synthesized, the selection function 113 calculates the similarity between the frame data and performs a selection process to include frame data whose calculated similarity is equal to or greater than a threshold value in the frame data to be synthesized. Here, the similarity between the frame data is calculated using a known similarity calculation method as appropriate. Note that FIG. 4B is a diagram illustrating an example of the selection process according to the first embodiment.

[0043] The signal processing function 112 generates blood flow data by extracting information derived from blood flow in the measurement data. For example, the signal processing function 112 applies an MTI (Moving Target Indicator) filter to the selected plurality of frame data to be synthesized (step S104). This reduces information derived from tissue that is stationary between frames or tissue with little movement (tissue signal components (clutter)), and extracts information derived from blood flow (blood flow signal components). The MTI filter may be a filter with fixed filter information, such as a Butterworth-type IIR (Infinite Impulse Response) filter or a polynomial regression filter. The MTI filter may be an adaptive filter that changes coefficients according to the input signal using eigenvalue decomposition or singular value decomposition.

[0044] The signal processing function 112 can also decompose the frame data into multiple bases using eigenvalue decomposition or singular value decomposition, and extract a specific base to remove tissue-derived information and extract blood flow-derived information. The signal processing function 112 can also use the Doppler processing function 110i to calculate a velocity vector for each coordinate in the received signal data and obtain a blood flow vector that represents the magnitude and direction of blood flow by using techniques such as vector Doppler, speckle tracking, and vector flow mapping. In addition to the methods exemplified here, other methods can be used to extract blood flow-derived information or remove tissue-derived information contained in the frame data (measurement data).

[0045] The signal processing function 112 can also estimate the amount of displacement of the object between multiple frame data to be combined, and correct each frame data to be combined based on the estimated result. Specifically, the signal processing function 112 calculates the amount of displacement of the object between frames from the multiple frame data to be combined, and corrects the frame data based on the calculated amount of displacement.

[0046] The synthesis processing function 114 performs synthesis processing on the selected frame data. Specifically, the synthesis processing function 114 generates added data by adding together multiple frames represented by the frame data (step S105). For example, as shown in FIG. 5, the synthesis processing function 114 generates added data 21 by adding together multiple frame data 20 (frames f1 to f5) to be synthesized, which were selected by the processing in step S103 and from which clutter components have been removed by the processing in step S104. Then, the synthesis processing function 114 sequentially generates added data from the frame data to be synthesized that are sequentially selected. Note that FIG. 5 is a diagram for explaining an example of data processing according to the first embodiment.

[0047] The added data 21 has a higher S / N ratio (Signal-to-Noise Ratio) than each frame data 20 before addition. The added data 21 is, for example, blood flow data (power Doppler data) in which information derived from blood flow is emphasized. The added data 21 is also obtained by averaging fluid signals after clutter removal over a predetermined time period, and represents the boundary of a flow path (for example, the shape of a blood vessel). A frame represented by the added data 21 includes signal values ​​(amplitude values) representing speckle patterns caused by constructive and destructive interference between ultrasonic waves reflected from the fluid, and signal values ​​representing other noises (artifacts, etc.) that occur instantaneously or intermittently on the time axis.

[0048] The extraction function 115 extracts signal components representing the object from the added data (step S106). Specifically, the extraction function 115 continuously extracts signal components representing the object for each of the sequentially generated added data. Here, the object is something that is to be highly resolved, and includes, for example, at least one of blood, body tissue, and contrast agent. Furthermore, continuously extracting signal components representing the object means extracting the positions of signals (pixels) representing the object without leaving a predetermined interval in signal space or image space. For example, if the object is blood, the extraction function 115 extracts the positions of signals (pixels) representing blood without leaving a predetermined interval in signal space or image space.

[0049] For example, as shown in FIG. 5, the extraction function 115 performs an extraction process to continuously extract signal components representing the object from the added data 21 generated by the synthesis processing function 114, thereby obtaining signal components 22, which are extracted data representing the object.

[0050] 6 is a diagram illustrating an example of extraction processing according to the first embodiment. As shown in FIG. 6, the extraction function 115 extracts the position of a signal (pixel) representing an object (for example, blood) using the kernel 30. Specifically, the extraction function 115 places the kernel 30 at a position of interest in a frame (image) represented by the added data 21, and compares a pixel value (luminance value) corresponding to the position of interest with pixel values ​​(luminance values) corresponding to the positions of pixels in the range where the kernel 30 is placed (around the position of interest), thereby extracting the position of the signal representing the position of the object.

[0051] For example, as shown in FIG. 6, when a kernel 30 based on a ratio of 3 vertical x 1 horizontal is used, the extraction function 115 positions the center (second square) of the kernel 30 as a position of interest, and if the pixel value corresponding to this position is greater than or equal to the pixel values ​​corresponding to the positions at both ends of the kernel 30 (first square, third square), extracts the signal component at the position of interest as a signal component representing the object. The extraction function 115 sets each pixel of the added data 21 as a position of interest and sequentially executes the extraction process described above. That is, the extraction function 115 determines whether or not a signal component is included for each pixel included in the added data 21. In this way, the extraction function 115 continuously extracts signal components representing the object.

[0052] 6 shows an example in which a signal component representing an object is extracted by arranging kernels 30 in four directions (vertical, horizontal, diagonally upper right, and diagonally upper left) for each position in a frame represented by added data, but the shape and size of the kernel are not limited to this. Also, while the example in FIG. 6 shows an example in which an object is extracted pixel by pixel, a signal component may be extracted for each position in a frame in groups each having multiple pixels. When extracting a signal component in groups each having multiple pixels, the average or integrated value of the signal values ​​at each position constituting the group may be used as the pixel value (luminance value) corresponding to each position in the frame.

[0053] The output function 116 outputs composite data based on the signal components extracted by the extraction function 115 and the added data (S107). For example, as shown in FIG. 5, the output function 116 outputs composite data 23 based on the added data 21 and the signal component 22. Here, the output function 116 generates the composite data 23 by performing a synthesis process or a correction process on the signal component 22 and the added data 21. The synthesis process is a process of synthesizing (adding) the signal component 22 (extracted data) and the added data 21 (power signal data). The correction process is a process of correcting the added data 21 (power signal data) based on the signal component 22 (extracted data).

[0054] As described above, the ultrasound diagnostic device 10 adjusts the number of frames used for analysis (the number of frames to be synthesized) according to the presence or absence of movement in each frame, thereby making it possible to improve the image quality of ultrasound images while avoiding the occurrence of artifacts due to body movement even when movement occurs in the subject.

[0055] Fig. 7 is a diagram showing an example of a processing result according to a comparative example. Fig. 7 shows a case where the number of frames used for analysis is not adjusted according to the presence or absence of motion in each frame. As shown in Fig. 7, when a synthesis process is performed using eight frames (frames f1 to f8) in which subject movement occurs between frames f5 and f6 as frame data to be synthesized, the generated added data 41 will depict the subject movement as an artifact. Furthermore, when signal components are extracted from such added data 41, signal components 42 containing artifacts will be extracted, as shown in Fig. 7.

[0056] As described above, according to the first embodiment, the control function 111 acquires a plurality of ultrasound data sets, each representing a frame of the subject. The signal processing function 112 detects the motion of the subject based on the plurality of ultrasound data sets. The selection function 113 selects frame data representing frames to be synthesized from the plurality of ultrasound data sets based on the detection result of the subject's motion. The synthesis processing function 114 performs synthesis processing on the selected frame data sets. Therefore, the ultrasound diagnostic apparatus 10 according to the first embodiment can selectively change the frame data sets to be synthesized in accordance with the motion of the subject, thereby enabling the image quality of ultrasound images to be improved while avoiding the occurrence of artifacts due to body motion.

[0057] Furthermore, according to the first embodiment, the synthesis processing function 114 performs synthesis processing to generate sum data by adding together multiple frames represented by frame data. The extraction function 115 extracts signal components representing the target object from the sum data. Therefore, the ultrasound diagnostic apparatus 10 according to the first embodiment can generate high-quality images that are free of artifacts.

[0058] Furthermore, according to the first embodiment, the signal processing function 112 calculates an image quality evaluation index for each frame of the plurality of ultrasound data, and detects the subject's motion based on the calculated image quality evaluation index. The signal processing function 112 also performs principal component analysis on each frame of the plurality of ultrasound data, and detects the subject's motion based on the analysis results. The signal processing function 112 also detects the subject's motion by inputting each frame of the plurality of ultrasound data into a trained model trained using a dataset including training ultrasound data and information regarding the presence or absence of subject motion in the training ultrasound data. Therefore, the ultrasound diagnostic apparatus 10 according to the first embodiment can detect the subject's motion using various techniques.

[0059] Furthermore, according to the first embodiment, the selection function 113 selects, from the plurality of ultrasound data, a plurality of frame data in which the subject's movement is relatively small, as frame data representing frames to be synthesized. Therefore, the ultrasound diagnostic apparatus 10 according to the first embodiment can select frame data that does not include the subject's movement.

[0060] Furthermore, according to the first embodiment, the selection function 113 selects, from among the set number of frames of ultrasound data, frame data with a larger number of frames, with the boundary defined by the time point at which the subject's motion is detected by the signal processing function 112, as frame data representing the frame to be synthesized. Furthermore, the selection function 113 selects, from among multiple sets of ultrasound data acquired over time, frame data acquired before the time point at which the subject's motion is detected by the signal processing function 112, as frame data representing the frame to be synthesized. Furthermore, the selection function 113 selects, as frame data representing the frame to be synthesized, multiple sets of frame data whose frames represented by the multiple sets of ultrasound data have a similarity equal to or greater than a reference level. Therefore, the ultrasound diagnostic apparatus 10 according to the first embodiment can perform synthesis processing using as much frame data as possible, and can generate a higher-quality image even when the subject moves.

[0061] (Second embodiment) In the first embodiment described above, a case has been described in which a signal component representing an object is extracted from added data obtained by performing an addition process on a plurality of frame data to be combined. In the second embodiment, a case will be described in which signal components representing an object are extracted from each of a plurality of frame data to be combined, and a combination process is performed on each extracted signal component. Note that in the second embodiment, the processing content by the combination processing function 114 and the processing content by the extraction function 115 are different from those in the first embodiment. The following description will focus on these.

[0062] The extraction function 115 according to the second embodiment extracts signal components representing an object from each of a plurality of frames represented by frame data. Specifically, the extraction function 115 executes a signal component extraction process for each of a plurality of frame data to be synthesized selected by the selection function 113. FIG. 8 is a diagram for explaining an example of data processing according to the second embodiment. Here, FIG. 8 shows an example in which subject movement is detected between frame f5 and frame f6 in frame data 20, and frames f1 to f5 are selected as frame data to be synthesized.

[0063] As shown in FIG. 8, the extraction function 115 executes extraction processing on the synthesis target frames f1, f2, f3, f4, and f5, thereby extracting signal components 24 corresponding to each frame.

[0064] The synthesis processing function 114 according to the second embodiment performs synthesis processing to generate integrated data by integrating signal components in multiple frames. Specifically, the synthesis processing function 114 generates integrated data by integrating multiple signal components extracted from the frame data to be synthesized by the extraction function 115. For example, as shown in FIG. 8, the synthesis processing function 114 integrates five signal components 24 extracted from frames f1 to f5 to generate integrated data 25.

[0065] In the above example, a case has been described in which signal components are extracted from frame data to be synthesized that have been selected based on the detection result of the subject's movement. However, the embodiment is not limited to this, and the subject's movement may be detected after the signal components have been extracted. In such a case, for example, the extraction function 115 extracts signal components representing the object from all of frames f1 to f8 shown in FIG. 8. The signal processing function 112 detects the subject's movement based on the extracted signal components, and the selection function 113 selects multiple signal components to be synthesized based on the movement detection result. The synthesis processing function 114 generates integrated data by integrating the multiple signal components to be synthesized.

[0066] As described above, according to the second embodiment, the extraction function 115 extracts signal components representing an object from each of the multiple frames represented by the frame data. The synthesis processing function 114 performs synthesis processing to generate integrated data by integrating the signal components in the multiple frames. Therefore, the ultrasound diagnostic apparatus 10 according to the second embodiment can extract signal components from frame data that is not affected by body movement, making it possible to generate high-quality images that are free of artifacts.

[0067] (Other embodiments) In addition to the above-described embodiments, the ultrasound diagnostic device 10 according to the present application can also acquire information on whether a signal component has been extracted at each position within a plurality of frames selected as frame data to be synthesized, and acquire filter information by converting the acquired information into weights. Specifically, the extraction function 115 generates filter information corresponding to each position within the frame data based on the extraction results of signal components representing an object in the plurality of frame data, and the synthesis processing function 114 generates integrated data (or sum data) using the filter information. For example, the extraction function 115 acquires the presence or absence of a signal component at each position within the image for each of the synthesis target frames (frames f1 to f5) shown in FIG. 4A. The extraction function 115 then generates filter information for a position within the image in which the weight increases as the number of frames from which a signal component has been extracted increases. When performing synthesis processing, the synthesis processing function 114 uses the above-described filter information to generate sum data (or integrated data) in which positions where signal components are extracted more frequently are displayed as pixels representing the object.

[0068] FIG. 9 is a diagram illustrating an example of data processing according to another embodiment. Here, FIG. 9 illustrates an example in which frames f1 to f5 are selected as the frame data to be synthesized in frame data 20, as described in FIG. 8. For example, as shown in FIG. 9, extraction function 115 executes the above-described extraction process on frames f1, f2, f3, f4, and f5 to be synthesized, thereby extracting signal components 24 corresponding to the object for each frame. Here, the extraction process executed by extraction function 115 also identifies whether the pixel value at each position in the frame is greater than the pixel values ​​corresponding to the surrounding pixels around that position.

[0069] An extraction function 115 according to another embodiment executes the extraction process described above to acquire, for each position in the frame data to be synthesized, identification information for identifying whether the signal value at that position is greater than or equal to the signal values ​​corresponding to the surrounding areas, and generates filter information based on the acquired identification information. For example, as shown in FIG. 9 , the extraction function 115 acquires filter information 26 from identification information n1 based on the signal component 24 extracted from frame f1, identification information n2 based on the signal component 24 extracted from frame f2, identification information n3 based on the signal component 24 extracted from frame f3, identification information n4 based on the signal component 24 extracted from frame f4, and identification information n5 based on the signal component 24 extracted from frame f5.

[0070] For example, when using the kernel shown in FIG. 6 (a kernel based on a ratio of 3 vertical by 1 horizontal), the extraction function 115 places the center (second square) of the kernel 40 as a position of interest, and acquires identification information that represents the position of interest as "1" if the pixel value corresponding to that position is greater than (or equal to) the pixel values ​​corresponding to the positions at both ends of the kernel 40 (the first square and the third square), and represents the position of interest as "0" if the pixel value corresponding to the position of interest is smaller than the pixel values ​​corresponding to the surrounding pixels of the position. The extraction function 115 performs this process for each position of frames f1 to f5, thereby acquiring identification information n1 to n5 for each frame data. That is, the extraction function 115 acquires a binary image for each frame data in which each pixel is converted to "0" or "1."

[0071] Furthermore, the extraction function 115 acquires filter information 26 corresponding to each position of the frames based on the acquired identification information (identification information n1 to n5) for the five frames. Here, the filter information 26 is calculated according to the value represented by each identification information (n1 to n5) at corresponding positions (coordinates (x, y)) between the frames. For example, if all five frames at corresponding positions in each identification information are "1," the filter information at that position will be "1." Also, if there are three "1"s and two "0"s among the five frames at corresponding positions, the filter information at that position will be "0.6." If all five frames at corresponding positions are "0," the filter information at that position will be "0."

[0072] In this way, the more times that the identification information in the frame data to be synthesized has a "1" at corresponding positions, the larger the value of the filter information 26 is set to. In other words, the more times that the identification information has a "1" at corresponding positions, the more likely it is that a signal value that is larger than the surroundings is continuously obtained at that position, and the more likely it is that a fluid is present, so a higher weighting coefficient is set to the filter information at such positions. On the other hand, if the identification information contains a "0" at corresponding positions, the more likely it is that noise (artifacts, etc.) that occurred instantaneously or intermittently on the time axis is present at that position, so the more "0" there is at such positions, the lower the weighting coefficient is set to the filter information at such positions.

[0073] The synthesis processing function 114 generates integrated data (or sum data) using the filter information 26. For example, as shown in FIG. 9, the synthesis processing function 114 generates integrated data 25 by integrating five signal components 24 extracted from frames f1 to f5, respectively. Then, the synthesis processing function 114 generates integrated data using the filter information 26 by multiplying pixel values ​​representing each position of the generated integrated data 25 by filter information (weighting coefficients) set for each corresponding position in the filter information 26. This integrated data is multiplied by a weighting coefficient with a larger value for positions in the frames to be synthesized where fluid is likely to exist, thereby emphasizing the position of the fluid. On the other hand, this integrated data is multiplied by a weighting coefficient with a smaller value for positions in the frames to be synthesized where instantaneous or intermittent noise (artifacts, etc.) is likely to exist, thereby reducing noise.

[0074] Furthermore, in the second embodiment, the signal components extracted from each of the plurality of frame data by the extraction function 115 may be assigned weight information determined according to each of the plurality of frame data. For example, for each of the plurality of frame data, the shorter the elapsed time of the frame data, the higher the weight information assigned to the signal component extracted from that frame data. This allows the signal components extracted from each of the plurality of frame data to be weighted according to the elapsed time.

[0075] In the first embodiment, the extraction function 115 uses a kernel 30 based on a 3x1 ratio, arranged in four directions: vertical, horizontal, diagonally upper right, and diagonally upper left, and extracts signal components for each position in the frame based on a comparison of pixel values ​​corresponding to that position with pixel values ​​corresponding to surrounding positions within each direction and distance. However, this is not limited to this. The surrounding positions whose pixel values ​​are compared by the extraction function 115 may be positions within a predetermined direction and distance from the corresponding position in the frame. The extraction function 115 may also assign weight information to the signal components that is determined according to at least one of the direction and distance.

[0076] In the above embodiment, the extraction function 115 performs extraction processing to extract signal components representing the object as a process for increasing the resolution of the object. However, the process for increasing the resolution of the object may be processing other than the above. For example, the extraction function 115 may increase the resolution of the object by performing peak sharpening, which applies a nonlinear function to pixel values ​​in a frame. In such a case, for example, the extraction function 115 first performs resampling on the sum data 21 shown in FIG. 5 to increase pixel density and improve effective resolution. Here, resampling may, for example, replace each pixel with multiple pixels of smaller size. Furthermore, the signal value of each pixel after replacement may be interpolated from the original signal value (the signal value before replacement) using, for example, bicubic interpolation.

[0077] Furthermore, the extraction function 115 sharpens pixel values ​​(signal components relatively larger than those of surrounding pixels) representing the object by raising each pixel value of the resampled sum data 21 to a power (e.g., 8th power, 12th power, etc.) (hereinafter, pixel values ​​representing the object may be referred to as local peaks). The extraction function 115 extracts the position of the local peak by, for example, performing threshold processing on the pixel values ​​after the power raising. Note that the target of the above-mentioned local peak sharpening process is not limited to sum data, but may also be frame data. That is, the extraction function 115 performs enhancement processing to enhance the object on multiple frame data, and the synthesis processing function 114 adds the multiple frame data after the enhancement processing to generate the sum data. For example, the extraction function 115 performs the above-mentioned local peak sharpening process on each of frames f1 to f5 shown in FIG. 8. The synthesis processing function 114 generates the sum data by adding the frame data after the local peak sharpening process.

[0078] In the above-described embodiment, the synthesis processing function 114 generates single added data with an improved S / N ratio by adding multiple frames of data to be synthesized. However, ultrasound data with an improved S / N ratio may be acquired without performing this processing. In such a case, for example, the synthesis processing function 114 may input multiple frames of data 20 to be synthesized into a trained model that inputs multiple frames of data consecutive in the time direction obtained by executing an ultrasound scan and outputs single ultrasound data with a higher S / N ratio than the frame data, and acquire the single ultrasound data output from the trained model as the added data 21. This trained model is trained using a dataset that uses multiple frames of data as input data and single ultrasound data with a higher S / N ratio than each frame of data as training data. The synthesis processing function 114 inputs multiple frames of data 20 into the trained model and acquires the single ultrasound data output from the trained model with a higher S / N ratio than the multiple frame data as the added data 21.

[0079] Furthermore, in the first embodiment described above, an example in which the present invention is applied to an ultrasound diagnostic apparatus has been described, but the present invention is not limited to this and may be applied to medical information processing devices other than ultrasound diagnostic apparatuses. For example, the present invention may be applied to a medical information processing device such as a workstation or server that acquires ultrasound data based on the results of an ultrasound scan of a subject. For example, a medical information processing device such as a workstation or server may perform the above-mentioned processing using multiple frame data collected in the past.

[0080] In addition, in the above description, the fluid represented by the ultrasound image is blood flow, and therefore, the control function 111 acquires the first ultrasound data obtained by an ultrasound scan in the absence of a contrast agent, but the fluid may be a contrast agent. In this case, the control function 111 may acquire multiple ultrasound data obtained by an ultrasound scan in the presence of a contrast agent.

[0081] Note that the components of each device illustrated in the above description of the embodiments are conceptual functional units and do not necessarily have to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to that illustrated, and all or part of the devices can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

[0082] The methods described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer or a workstation. This program can be distributed via a network such as the Internet. This program can also be recorded on a non-transitory computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, a DVD, a USB memory, or a flash memory such as an SD card memory, and can be executed by being read from the non-transitory recording medium by a computer.

[0083] As described above, according to the embodiment, it is possible to improve the image quality of an ultrasound image while avoiding the occurrence of artifacts due to body movement.

[0084] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0085] 10 Ultrasound diagnostic equipment 100 Medical information processing device 111 Control Functions 112 Signal Processing Functions 113 Selection Function 114 Composition Processing Function 115 Extraction Function 116 Output Function

Claims

1. an acquisition unit that acquires a plurality of ultrasound data representing frames of a subject that are consecutive in the time direction; a detection unit that detects the movement of the subject; a selection unit that selects a plurality of frame data representing a synthesis target from the plurality of ultrasound data based on the motion detection result; a synthesis processing unit that performs synthesis processing on the selected plurality of frame data; A medical information processing device comprising:

2. The medical image processing apparatus according to claim 1 , wherein the synthesis processing unit adds the plurality of frame data to generate added data.

3. The medical image processing apparatus according to claim 2 , further comprising an extraction unit that performs high definition processing of the object represented by the added data.

4. an extracting unit that extracts signal components representing an object from each of the plurality of frame data; The medical image processing apparatus according to claim 1 , wherein the synthesis processing unit performs synthesis processing to generate integrated data by integrating each signal component in the plurality of frame data.

5. the extraction unit generates filter information corresponding to each position of the frame data based on an extraction result of the signal component representing the object in the plurality of frame data; The medical image processing apparatus according to claim 4 , wherein the synthesis processing unit generates the integrated data using the filter information.

6. an extraction unit that executes enhancement processing for enhancing an object on each of the plurality of frame data; The medical image processing apparatus according to claim 1 , wherein the synthesis processing unit generates added data by adding the plurality of frames of data after the enhancement processing.

7. The medical image processing apparatus according to claim 1 , wherein the detector calculates an image quality evaluation index for each frame of the plurality of ultrasound data, and detects the movement of the subject based on the calculated image quality evaluation index.

8. The medical image processing apparatus according to claim 1 , wherein the detector performs a principal component analysis on each frame of the plurality of ultrasound data, and detects the movement of the subject based on the analysis result.

9. 2. The medical information processing device according to claim 1, wherein the detection unit detects the movement of the subject by inputting each frame of the plurality of ultrasound data to a trained model trained using a dataset including training ultrasound data and information regarding the presence or absence of movement of the subject in the training ultrasound data.

10. 10. The medical image processing device according to claim 1, wherein the selection unit selects, from the plurality of ultrasound data, a plurality of frame data in which the subject's movement is relatively small, as frame data representing the frames to be synthesized.

11. 11. The medical image processing device according to claim 10, wherein the selection unit selects, from the set number of frames of the ultrasound data, frame data with a larger number of frames, with a boundary defined by a point in time when the motion of the subject is detected by the detection unit, as frame data representing the frame to be synthesized.

12. 11. The medical information processing device according to claim 10, wherein the selection unit selects, from the plurality of ultrasound data acquired over time, frame data acquired before the time point at which the motion of the subject is detected by the detection unit, as frame data representing the frame to be synthesized.

13. The medical image processing apparatus according to claim 10 , wherein the selection unit selects a plurality of frame data having a similarity equal to or greater than a reference level between frames represented by the plurality of ultrasound data as frame data representing the frames to be synthesized.

14. an execution unit that causes the ultrasound probe to perform an ultrasound scan; an acquisition unit that acquires a plurality of ultrasound data representing successive frames of the subject in a time axis direction; a detection unit that detects the movement of the subject; a selection unit that selects a plurality of frame data representing a synthesis target from the plurality of ultrasound data based on the motion detection result; a synthesis processing unit that performs synthesis processing on the selected plurality of frame data; An ultrasound diagnostic device comprising:

15. an acquiring step of acquiring a plurality of ultrasound data representing successive frames of the subject in a time axis direction; a detecting step of detecting a movement of the subject; a selection step of selecting a plurality of frame data representing a synthesis target from the plurality of ultrasound data based on the motion detection result; a synthesis processing step of performing synthesis processing on the selected plurality of frame data; A medical information processing method comprising:

Images