Analysis device and program

Abstract

To evaluate changes in signal strength over time.SOLUTION: An analysis device according to the embodiment includes an image acquisition unit, an index value acquisition unit, and an output unit. The image acquisition unit acquires a plurality of time-series ultrasound images. The index value acquisition unit acquires index values at multiple time phases which indicate a correlation between a first position in the ultrasound image in one time phase and a second position corresponding to the first position in the ultrasound image in another time phase. The output unit outputs identification information that identifies the degree of tissue variation, on the basis of the index values acquired in the plurality of time phases.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] An embodiment of the present invention relates to an analysis device and an analysis program. [Background technology]

[0002] Conventionally, ultrasound imaging has been used for the differential diagnosis of hemangiomas. However, ultrasound imaging places a burden on the subject and is time-consuming, making it difficult to introduce into routine examinations. To perform differential diagnosis of hemangiomas in routine examinations, a technology that can differentiate using imaging methods that do not use contrast agents, such as B-mode imaging or Doppler mode imaging, is required.

[0003] A characteristic finding of hemangiomas is that the echo intensity changes (fluctuations) over time, and is called the "thread worm" sign or the "wax and wane" sign. Capturing this finding may be helpful in the differentiation of hemangiomas. In addition, the above finding is not only seen in hemangiomas, but also in other pathological conditions, such as pus accumulated in a cyst, so the technology that captures this finding may be applicable to the diagnosis of other pathological conditions. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 4744833 specification Summary of the Invention [Problem to be solved by the invention]

[0005] An object of the present invention is to provide an analysis device and an analysis program capable of evaluating changes in signal intensity over time. [Means for solving the problem]

[0006] An analysis device according to an embodiment includes an image acquisition unit, an index value acquisition unit, and an output unit. The image acquisition unit acquires a plurality of ultrasound images in a time series. The index value acquisition unit acquires, at a plurality of time phases, index values ​​indicating a correlation between a first position in the ultrasound image at one time phase and a second position corresponding to the first position in the ultrasound image at another time phase. The output unit outputs identification information for identifying a degree of tissue fluctuation based on the index values ​​acquired at the plurality of time phases. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of an ultrasonic diagnostic apparatus according to the first embodiment. [Diagram 2] FIG. 2 is a diagram for explaining a point of interest according to the first embodiment. [Diagram 3] FIG. 3 is a diagram for explaining a point of interest according to the first embodiment. [Figure 4] FIG. 4 is a flowchart showing a processing procedure of the ultrasonic diagnostic apparatus according to the first embodiment. [Diagram 5] FIG. 5 is a flowchart showing the procedure of the index value calculation process according to the first embodiment. [Figure 6] FIG. 6 is a diagram for explaining the motion stabilizer process according to the first embodiment. [Figure 7] FIG. 7 is a diagram for explaining the process of calculating the first index value according to the first embodiment. [Figure 8] FIG. 8 is a diagram for explaining the process of calculating the second index value according to the first embodiment. [Figure 9] FIG. 9 is a diagram showing an example of a parametric image according to the first embodiment. [Figure 10] FIG. 10 is a diagram showing an example of a parametric image when residual tissue motion is included. [Figure 11] FIG. 11 is a flowchart showing the procedure of the index value calculation process according to the second embodiment. [Figure 12]FIG. 12 is a diagram for explaining the process of setting the reference region according to the second embodiment. [Figure 13] FIG. 13 is a diagram illustrating an example of a time series change in the third index value according to the second embodiment. [Figure 14] FIG. 14 is a diagram showing an example of a display image according to the second embodiment. [Figure 15] FIG. 15 is a diagram for explaining the processing of the calculation function according to the modified example of the second embodiment. [Figure 16] FIG. 16 is a flowchart showing the procedure of the index value calculation process according to the third embodiment. [Figure 17] FIG. 17 is a diagram for explaining a process of generating a two-dimensional correlation coefficient distribution according to the third embodiment. [Figure 18] FIG. 18 is a diagram for explaining a process of generating a two-dimensional correlation coefficient distribution according to the third embodiment. [Figure 19] FIG. 19 is a diagram for explaining a process of calculating a correlation coefficient of a two-dimensional correlation coefficient distribution according to the third embodiment. [Figure 20] FIG. 20 is a diagram for explaining a process of calculating a correlation coefficient of a two-dimensional correlation coefficient distribution according to the third embodiment. [Figure 21] FIG. 21 is a block diagram showing an example of the configuration of a medical image processing apparatus according to another embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, an analysis device and an analysis program according to an embodiment will be described with reference to the drawings. Note that the following embodiment is not limited to the following description. In addition, the embodiment can be combined with other embodiments or conventional techniques as long as there is no contradiction in the processing content.

[0009] The analysis device includes, for example, a medical diagnostic device and a medical information processing device. Here, the medical diagnostic device includes, for example, an ultrasound diagnostic device, a photoultrasound diagnostic device (photoacoustic imaging device), an X-ray diagnostic device, an X-ray CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, a SPECT (Single Photon Emission Computed Tomography) device, a PET (Positron Emission computed Tomography) device, a SPECT-CT device in which a SPECT device and an X-ray CT device are integrated, a PET-CT device in which a PET device and an X-ray CT device are integrated, or a group of these devices. Also, the medical information processing device includes, for example, a workstation, a PACS (Picture Archiving Communication System) viewer, etc.

[0010] (First embodiment) Fig. 1 is a block diagram showing an example of the configuration of an ultrasound diagnostic apparatus 1 according to the first embodiment. As shown in Fig. 1, the ultrasound diagnostic apparatus 1 according to the first embodiment includes an apparatus main body 100, an ultrasound probe 101, an input interface 102, and a display 103. The ultrasound probe 101, the input interface 102, and the display 103 are communicatively connected to the apparatus main body 100. Note that a subject P is not included in the configuration of the ultrasound diagnostic apparatus 1. Also, the ultrasound diagnostic apparatus 1 is an example of an analysis device.

[0011] The ultrasonic probe 101 has a plurality of transducers (e.g., piezoelectric transducers), which generate ultrasonic waves based on a drive signal supplied from a transmission / reception circuit 110 included in the device main body 100, which will be described later. The plurality of transducers included in the ultrasonic probe 101 receive reflected waves from the subject P and convert them into electrical signals. The ultrasonic probe 101 also has a matching layer provided on the transducers, a backing material that prevents ultrasonic waves from propagating backward from the transducers, and the like.

[0012] When ultrasonic waves are transmitted from the ultrasonic probe 101 to the subject P, the transmitted ultrasonic waves are successively reflected by discontinuous surfaces of acoustic impedance in the internal tissues of the subject P, and are received as reflected wave signals (echo signals) by a plurality of transducers of the ultrasonic probe 101. The amplitude of the received reflected wave signal depends on the difference in acoustic impedance at the discontinuous surfaces where the ultrasonic waves are reflected. When the transmitted ultrasonic pulse is reflected by the surface of a moving blood flow, a heart wall, or the like, the reflected wave signal undergoes a frequency shift due to the Doppler effect depending on the velocity component of the moving body in the ultrasonic transmission direction.

[0013] The first embodiment is applicable to any of the following cases: the ultrasonic probe 101 shown in FIG. 1 is a one-dimensional ultrasonic probe in which a plurality of piezoelectric transducers are arranged in a row; the one-dimensional ultrasonic probe in which a plurality of piezoelectric transducers arranged in a row are mechanically oscillated; and the two-dimensional ultrasonic probe in which a plurality of piezoelectric transducers are arranged two-dimensionally in a lattice pattern.

[0014] The input interface 102 corresponds to a mouse, a keyboard, a button, a panel switch, a touch command screen, a foot switch, a trackball, a joystick, etc. For example, the input interface 102 accepts various setting requests from the operator of the ultrasound diagnostic apparatus 1 and transfers the accepted various setting requests to the apparatus main body 100.

[0015] The display 103 displays a GUI (Graphical User Interface) that enables the operator of the ultrasound diagnostic device 1 to input various setting requests using the input interface 102, and also displays ultrasound image data generated in the device main body 100, etc.

[0016] The device main body 100 is a device that generates ultrasound image data based on a reflected wave signal received by an ultrasound probe 101, and as shown in Fig. 1, has a transmission / reception circuit 110, a signal processing circuit 120, an image processing circuit 130, an image memory 140, a storage circuit 150, and a processing circuit 160. The transmission / reception circuit 110, the signal processing circuit 120, the image processing circuit 130, the image memory 140, the storage circuit 150, and the processing circuit 160 are connected to each other so as to be able to communicate with each other.

[0017] The transmission / reception circuit 110 has a pulse generator, a transmission delay unit, a pulser, etc., and supplies a drive signal to the ultrasonic probe 101. The pulse generator repeatedly generates a rate pulse for forming a transmission ultrasonic wave at a predetermined rate frequency. The transmission delay unit focuses the ultrasonic waves generated from the ultrasonic probe 101 into a beam shape and provides a delay time for each piezoelectric transducer required to determine the transmission directivity to each rate pulse generated by the pulse generator. The pulser applies a drive signal (drive pulse) to the ultrasonic probe 101 at a timing based on the rate pulse. That is, the transmission delay unit changes the delay time provided to each rate pulse to arbitrarily adjust the transmission direction of the ultrasonic waves transmitted from the piezoelectric transducer surface.

[0018] The transmission / reception circuit 110 has a function of instantaneously changing the transmission frequency, transmission drive voltage, etc., in order to execute a predetermined scan sequence based on instructions from the processing circuit 160, which will be described later. In particular, the change in the transmission drive voltage is realized by a linear amplifier type oscillation circuit that can instantaneously switch its value, or a mechanism that electrically switches between multiple power supply units.

[0019] The transmission / reception circuit 110 also has a preamplifier, an A / D (Analog / Digital) converter, a reception delay unit, an adder, etc., and performs various processes on the reflected wave signal received by the ultrasonic probe 101 to generate reflected wave data. The preamplifier amplifies the reflected wave signal for each channel. The A / D converter performs A / D conversion on the amplified reflected wave signal. The reception delay unit provides a delay time required to determine the reception directivity. The adder performs addition processing on the reflected wave signal processed by the reception delay unit to generate reflected wave data. The addition processing by the adder emphasizes the reflected component from a direction according to the reception directivity of the reflected wave signal, and an overall beam for ultrasonic transmission and reception is formed by the reception directivity and transmission directivity.

[0020] When scanning a two-dimensional region of the subject P, the transmission and reception circuit 110 causes the ultrasonic probe 101 to transmit ultrasonic beams in two-dimensional directions. Then, the transmission and reception circuit 110 generates two-dimensional reflected wave data from the reflected wave signals received by the ultrasonic probe 101. When scanning a three-dimensional region of the subject P, the transmission and reception circuit 110 causes the ultrasonic probe 101 to transmit ultrasonic beams in three-dimensional directions. Then, the transmission and reception circuit 110 generates three-dimensional reflected wave data from the reflected wave signals received by the ultrasonic probe 101.

[0021] The signal processing circuit 120 performs, for example, logarithmic amplification, envelope detection processing, etc., on the reflected wave data received from the transmission / reception circuit 110 to generate data (B-mode data) in which the signal intensity of each sample point is expressed as luminance brightness. The B-mode data generated by the signal processing circuit 120 is output to the image processing circuit 130.

[0022] Furthermore, the signal processing circuit 120 generates data (Doppler data) in which motion information based on the Doppler effect of a moving object is extracted at each sample point in the scanning region from the reflected wave data received from the transmission / reception circuit 110. Specifically, the signal processing circuit 120 performs frequency analysis on the velocity information from the reflected wave data, extracts blood flow, tissue, and contrast agent echo components due to the Doppler effect, and generates data (Doppler data) in which moving object information such as average velocity, variance, and power is extracted for multiple points. Here, the moving object is, for example, blood flow, tissue such as the heart wall, and contrast agent. The motion information (blood flow information) obtained by the signal processing circuit 120 is sent to the image processing circuit 130 and displayed in color on the display 103 as an average velocity image, variance image, power image, or a combination of these images.

[0023] The image processing circuit 130 generates ultrasound image data from the data generated by the signal processing circuit 120. The image processing circuit 130 generates B-mode image data in which the intensity of the reflected wave is expressed as brightness from the B-mode data generated by the signal processing circuit 120. The image processing circuit 130 also generates Doppler image data representing moving object information from the Doppler data generated by the signal processing circuit 120. The Doppler image data is velocity image data, variance image data, power image data, or image data that is a combination of these.

[0024] Here, the image processing circuit 130 generally converts (scan converts) a scan line signal sequence of ultrasonic scanning into a scan line signal sequence of a video format represented by a television or the like, and generates ultrasonic image data for display. Specifically, the image processing circuit 130 generates ultrasonic image data for display by performing coordinate conversion according to the ultrasonic scanning form of the ultrasonic probe 101. In addition to scan conversion, the image processing circuit 130 also performs various image processing, such as image processing (smoothing processing) for regenerating an average brightness image using multiple image frames after scan conversion, and image processing (edge ​​enhancement processing) using a differential filter in an image. In addition, the image processing circuit 130 combines supplementary information (text information of various parameters, scales, body marks, etc.) with the ultrasonic image data.

[0025] That is, the B-mode data and the Doppler data are ultrasonic image data before the scan conversion process, and the data generated by the image processing circuit 130 is ultrasonic image data for display after the scan conversion process. When the signal processing circuit 120 generates three-dimensional data (three-dimensional B-mode data and three-dimensional Doppler data), the image processing circuit 130 generates volume data by performing coordinate conversion according to the ultrasonic scanning form of the ultrasonic probe 101. Then, the image processing circuit 130 performs various rendering processes on the volume data to generate two-dimensional image data for display.

[0026] The image memory 140 is a memory that stores an image for display generated by the image processing circuit 130. The image memory 140 can also store data generated by the signal processing circuit 120. The B-mode data and Doppler data stored in the image memory 140 can be called by an operator after diagnosis, for example, and become ultrasound image data for display via the image processing circuit 130. Note that when simply referring to an "image" in this embodiment, it includes not only an image for display in which a color is assigned to each pixel, but also a data string in which the coordinates of each pixel correspond to a pixel value (signal value) (also referred to as "image data").

[0027] The memory circuitry 150 stores various data such as control programs for transmitting and receiving ultrasound, image processing, and display processing, diagnostic information (e.g., patient ID, doctor's findings, etc.), diagnostic protocols, and various body marks. The memory circuitry 150 is also used for storing image data stored in the image memory 140 as necessary. The data stored in the memory circuitry 150 can be transferred to an external device via a communication interface (not shown).

[0028] The processing circuitry 160 controls the overall processing of the ultrasound diagnostic apparatus 1. Specifically, the processing circuitry 160 controls the processing of the transmission / reception circuitry 110, the signal processing circuitry 120, and the image processing circuitry 130 based on various setting requests input by the operator via the input interface 102 and various control programs and various data read from the storage circuitry 150. The processing circuitry 160 also controls the display 103 to display ultrasound image data for display stored in the image memory 140.

[0029] 1, the processing circuit 160 executes an acquisition function 161, a calculation function 162, a generation function 163, and an output control function 164. Here, the calculation function 162 is an example of a calculation unit. The generation function 163 is an example of a generation unit. The output control function 164 is an example of an output control unit.

[0030] Here, for example, the processing functions executed by the acquisition function 161, the calculation function 162, the generation function 163, and the output control function 164, which are components of the processing circuitry 160 shown in Fig. 1, are recorded in a storage device (e.g., the storage circuitry 150) of the ultrasound diagnostic device 1 in the form of a program executable by a computer. The processing circuitry 160 is a processor that realizes a function corresponding to each program by reading each program from the storage device and executing it. In other words, the processing circuitry 160 in a state in which each program has been read has each function shown in the processing circuitry 160 in Fig. 1. The processing functions executed by the acquisition function 161, the calculation function 162, the generation function 163, and the output control function 164 will be described later.

[0031] Here, the "change (fluctuation) of signal intensity over time," which is known as a characteristic finding of hemangioma, can be expressed, for example, as the variation (standard deviation) of signal intensity in the frame direction (time direction). In this case, since the variation increases as the signal intensity increases, normalization is required using the signal intensity itself. However, normalization (i.e., using a rate of change) results in a larger change in signal intensity at low luminance, which makes it more likely to vary. In other words, the standard deviation results in a value that depends on the signal intensity. For this reason, it is considered important to define an index value that represents the change in signal intensity over time, independent of the signal intensity.

[0032] Therefore, in this embodiment, in defining an index value that indicates a change in signal strength over time, attention is paid to two points: "the strength of correlation between frames" and "the relationship in the frame direction of correlation."

[0033] The focus point according to the first embodiment will be described with reference to Fig. 2 and Fig. 3. Fig. 2 and Fig. 3 are diagrams for explaining the focus point according to the first embodiment. Fig. 2 and Fig. 3 show the correlation of pixel value distribution between adjacent frames for a plurality of medical images arranged in time series. Note that in Fig. 2 and Fig. 3, it is assumed that there is no overall movement (positional deviation) of tissue between each medical image or that the movement is corrected.

[0034] FIG. 2 illustrates an example of a correlation when there is a change in signal intensity (pixel value) over time. The scatter plots SP11, SP12, and SP13 shown in FIG. 2 are plots of points corresponding to each pixel position in a kernel (analysis window) set in a region R11 where there is fluctuation in the frame direction. Specifically, the horizontal axis of the scatter plot SP11 corresponds to the pixel value of the nth frame, and the vertical axis corresponds to the pixel value of the n+1th frame. The horizontal axis of the scatter plot SP12 corresponds to the pixel value of the n+1th frame, and the vertical axis corresponds to the pixel value of the n+2th frame. The horizontal axis of the scatter plot SP13 corresponds to the pixel value of the n+2th frame, and the vertical axis corresponds to the pixel value of the n+3th frame.

[0035] FIG. 3 illustrates an example of correlation when there is no change in signal intensity over time. The scatter diagrams SP21, SP22, and SP23 shown in FIG. 3 are plots of points corresponding to each pixel position in the kernel set in the region R12 where there is no fluctuation in the frame direction. Specifically, the horizontal axis of the scatter diagram SP21 corresponds to the pixel value of the nth frame, and the vertical axis corresponds to the pixel value of the n+1th frame. The horizontal axis of the scatter diagram SP22 corresponds to the pixel value of the n+1th frame, and the vertical axis corresponds to the pixel value of the n+2th frame. The horizontal axis of the scatter diagram SP23 corresponds to the pixel value of the n+2th frame, and the vertical axis corresponds to the pixel value of the n+3th frame.

[0036] For example, in an area where a hemangioma exists, the signal intensity changes over time, and thus a correlation as shown in FIG. 2 is observed. Since pixel values ​​change between adjacent frames even at the same pixel position, the scatter plots SP11, SP12, and SP13 in FIG. 2 are distributed over a wider area (area with a shape close to a perfect circle) compared to the scatter plots SP21, SP22, and SP23 in FIG. 3. In other words, in an area where a hemangioma exists, it can be said that the correlation between frames is weak. Depending on the degree of change in pixel values, the correlation between frames may temporarily become stronger. For example, the scatter plot SP13 is concentrated in a narrower area (an elongated oval area) compared to the scatter plot SP12, and thus the correlation is stronger. In other words, in an area where a hemangioma exists, it can be said that the correlation changes in the frame direction.

[0037] On the other hand, in areas where hemangiomas are not present, that is, areas where background liver or tumors other than hemangiomas are present, there is no change in signal intensity over time, and therefore a correlation as shown in FIG. 3 is observed. Since there is little change in pixel value between adjacent frames at the same pixel position, the scatter plots SP21, SP22, and SP23 in FIG. 3 are concentrated in a narrower area (a long and thin oval area) compared to the scatter plots SP11, SP12, and SP13 in FIG. 2. In other words, in areas where hemangiomas are not present, it can be said that "the correlation between frames is strong." In addition, the scatter plots SP21, SP22, and SP23 in FIG. 3 all maintain a strong correlation. In other words, in areas where hemangiomas are not present, it can be said that "the correlation is constant in the frame direction."

[0038] To summarise the two points mentioned above, it can be said that there are the following characteristics depending on the presence or absence of hemangioma. That is, areas with hemangioma are characterized by "weak correlation between frames" and "correlation changes in the frame direction". On the other hand, areas without hemangioma are characterized by "strong correlation between frames" and "correlation is constant in the frame direction".

[0039] Therefore, the ultrasound diagnostic apparatus 1 according to the first embodiment executes the following processing functions in order to output an index value using these two points of interest.

[0040] In the following embodiment, the case of evaluating the fluctuation of hemangioma depicted in a B-mode image will be described, but the present invention is not limited thereto. For example, the disclosed technology is not limited to hemangioma, and can quantitatively evaluate any tissue change that exhibits fluctuation in the frame direction in the image. In addition, the disclosed technology is not limited to B-mode images, and can also evaluate fluctuation in other ultrasound images such as Doppler images, or medical images captured by other medical image diagnostic devices.

[0041] The processing procedure of the ultrasound diagnostic device 1 according to the first embodiment will be described with reference to Fig. 4 and Fig. 5. Fig. 4 is a flowchart showing the processing procedure of the ultrasound diagnostic device 1 according to the first embodiment. Fig. 5 is a flowchart showing the processing procedure of the index value calculation process according to the first embodiment. The processing procedure shown in Fig. 4 is started, for example, when an instruction to display a parametric image is received from an operator. Note that the processing procedure shown in Fig. 5 corresponds to the processing content of step S104 shown in Fig. 4.

[0042] In the following description, the description will be made with reference to Fig. 6 to Fig. 9. Fig. 6 is a diagram for explaining a motion stabilizer process according to the first embodiment. Fig. 7 is a diagram for explaining a process for calculating a first index value according to the first embodiment. Fig. 8 is a diagram for explaining a process for calculating a second index value according to the first embodiment. Fig. 9 is a diagram showing an example of a parametric image according to the first embodiment.

[0043] In step S101, the processing circuitry 160 determines whether or not to start processing. For example, when the processing circuitry 160 receives an instruction from the operator to display a parametric image, it determines to start processing (Yes in step S101), and starts processing from step S102 onwards. Note that, when the processing circuitry 160 does not start processing (No in step S101), the processing from step S102 onwards is not started, and each processing function is in a standby state.

[0044] If step S101 is positive, in step S102, the acquisition function 161 acquires a plurality of time-series medical images. For example, the acquisition function 161 reads out B-mode image data of a plurality of time phases arranged in time series from the image memory 140. As a specific example, the acquisition function 161 reads out N pieces of B-mode image data from the first frame to the Nth frame from the image memory 140. Then, the acquisition function 161 sends the read out N pieces of B-mode image data to the calculation function 162. Note that N is a natural number. In the following process, the B-mode image data corresponding to the entire scanning range where the ultrasonic scan was performed may be processed, or only the image data within the region of interest may be processed.

[0045] Here, a case will be described in which N frames of B-mode image data have been captured and stored in advance in the image memory 140, but the embodiment is not limited to this. For example, the acquisition function 161 may acquire N frames of B-mode image data generated in real time. In this case, for example, the operator performs ultrasonic scanning for N frames using the ultrasonic probe 101 after step S101. Then, the image processing circuit 130 generates N frames of B-mode image data based on the N frames of B-mode data collected by this ultrasonic scanning. The acquisition function 161 acquires the generated N frames of B-mode image data from the image processing circuit 130.

[0046] In step S103, the calculation function 162 executes a motion tracking process on a plurality of medical images. For example, the read N sheets of B-mode image data include movement (positional deviation) between images caused by camera shake or body movement (such as heartbeat). For this reason, the calculation function 162 executes a motion tracking process on the N sheets of B-mode image data to identify the positional deviation between each medical image. Then, the calculation function 162 corrects the identified positional deviation between each medical image to generate a series of B-mode image data that does not include a positional deviation in the frame direction. Note that a known motion tracking process can be appropriately applied as the motion tracking process.

[0047] In step S104, the calculation function 162 executes an index value calculation process. The processes of steps S201 to S208 executed as the index value calculation process will be described with reference to FIG.

[0048] In step S201, the calculation function 162 executes smoothing processing in the spatial direction and the time direction on a plurality of medical images. For example, the calculation function 162 performs low-pass filter (LPF) processing on N B-mode images in the spatial direction. Here, for example, a moving average filter or a median filter can be applied for the LPF processing. This allows the calculation function 162 to reduce spike noise and speckle noise in the spatial direction.

[0049] Furthermore, the calculation function 162 performs LPF processing on N pixel values ​​arranged in the time direction for each pixel of the N B-mode images. This allows the calculation function 162 to reduce high-frequency noise such as spike noise in the time direction from the image signal intensity in the time direction. Note that the calculation function 162 may first perform either the LPF processing in the spatial direction or the LPF processing in the time direction, or may perform only one of them.

[0050] In step S202, the calculation function 162 sets a region of interest (ROI). For example, the calculation function 162 displays one B-mode image (typically the first frame image) of the N B-mode images after the smoothing process on the display 103. The operator specifies an area of ​​any position and size on the displayed B-mode image. The calculation function 162 sets the area specified by the operator on the B-mode image as the ROI. The calculation function 162 also sets areas of the same position and size as the ROI of the first frame as ROIs for the remaining B-mode images (for example, images from the second to Nth frames).

[0051] Although the case where the ROI is set manually has been described here, it may be set automatically. For example, the calculation function 162 can extract the contour of a structure in an image by a known segmentation process, and set the ROI in the image based on the position and shape of the extracted contour.

[0052] In step S203, the calculation function 162 executes a motion stabilizer process. For example, even in the B-mode image in which the positional deviation has been corrected by the process of step S103, the positional deviation may remain. This occurs when the areas of interest in the motion stabilizer process are different between step S103 and step S203. For example, in step S103, an arbitrary area is set as the area of ​​interest for the positional relationship between the background liver and the tumor in order to correct the overall movement. In contrast, in step S203, it is recommended to focus only on the background liver side in order to correct and detect the movement such as pulsation remaining in the overall movement that has been removed. If the areas of interest in both are different, a difference (positional deviation) occurs in the position correction result. Therefore, the calculation function 162 executes a motion stabilizer process on the N B-mode images after the smoothing process, and corrects or detects the remaining positional deviation between each image. Note that the areas of interest in the above-mentioned steps S103 and S203 are merely examples, and can be selected arbitrarily.

[0053] The process of step S203 will be described with reference to Fig. 6. Fig. 6 illustrates a process for detecting a positional deviation of an ROI from B-mode images of the nth, n+1th, and n+2th frames. In Fig. 6, region R20 indicates the ROI. The dashed rectangular region indicates the position of region R20 in the previous frame. Furthermore, n, n+1, and n+2 are all numbers included in the range of 1 to N.

[0054] As shown in FIG. 6, for example, the calculation function 162 performs a tracking process including pattern matching on the region R20 to identify the position of the region R20 in each B-mode image. Then, the calculation function 162 detects a positional deviation based on the difference in the position of the region R20 in adjacent frames. For example, the calculation function 162 detects a horizontal positional deviation x1 and a vertical positional deviation y1 between the images of the nth frame and the n+1th frame. The calculation function 162 also detects a horizontal positional deviation x2 and a vertical positional deviation y2 between the images of the n+1th frame and the n+2th frame.

[0055] In this way, the calculation function 162 detects the positional deviation between the N B-mode images. Note that the motion stabilizer process is an example of a motion correction process that corrects the motion between images, and a known motion stabilizer process can be appropriately applied. In addition, the same process as the motion tracking process described above may be applied as the motion stabilizer process.

[0056] In step S204, the calculation function 162 selects an image pair to be processed. For example, the calculation function 162 selects two B-mode images, the nth frame and the n+1th frame, as an image pair to be processed from among the N B-mode images.

[0057] In step S205, the calculation function 162 corrects the motion of the image pair. For example, the calculation function 162 corrects the motion (misalignment) between two medical images included in the image pair based on the misalignment detected by the motion stabilizer process.

[0058] In step S206, the calculation function 162 calculates a first index value indicating the similarity between the image signals for each pixel position. For example, the calculation function 162 calculates a first index value indicating the similarity between the image signals in an image pair consisting of two medical images included in a plurality of medical images.

[0059] In step S207, the calculation function 162 determines whether or not there are any image pairs remaining to be processed. For example, when there is an unprocessed B-mode image among the N B-mode images, the calculation function 162 determines that there are any image pairs remaining to be processed (Yes in step S207), and proceeds to processing in step S204. On the other hand, when there is no unprocessed B-mode image among the N B-mode images, the calculation function 162 determines that there are no image pairs remaining to be processed (No in step S207), and proceeds to processing in step S208.

[0060] That is, calculation function 162 repeatedly executes the processes of steps S204 to S207 while changing the image pair to be processed. Then, calculation function 162 selects a plurality of image pairs in order, and calculates a first index value based on each image pair in order.

[0061] The processing of steps S204 to S207 will be described with reference to Fig. 7. Fig. 7 illustrates a process for calculating first index values ​​of the nth and n+1th frames from the nth, n+1th, and n+2th frame B-mode images. In Fig. 7, regions R21, R22, and R23 indicate kernels (analysis windows) set in the nth, n+1th, and n+2th frame B-mode images, respectively.

[0062] First, a case will be described where the nth and n+1th frame B-mode images are selected as an image pair to be processed. In this case, the calculation function 162 sets a kernel for each of the two B-mode images, the nth and n+1th frames. For example, the calculation function 162 calculates a pixel position (x k ,y k ) is set as the center of the kernel (region R21). Here, the positional shift between the nth frame image and the n+1th frame image is (x1, y1). Therefore, the calculation function 162 calculates the pixel position (x k +x1,y k +y1) is used as the center to set the kernel (region R22).

[0063] Then, the calculation function 162 calculates a first index value for the nth frame based on a correlation coefficient between an image signal in the kernel of the nth frame and an image signal in the kernel of the n+1th frame. For example, a pixel index for identifying the pixel position is assigned to each pixel position in the kernel. Then, the calculation function 162 calculates a correlation coefficient using a pixel value for each pixel index in the region R21 and a pixel value for each pixel index in the region R22 as two-variable data. Then, the calculation function 162 calculates a first index value for the nth frame by subtracting the calculated correlation coefficient from "1".

[0064] Next, a case will be described where the B-mode images of the n+1th and n+2nd frames are selected as the image pair to be processed. Here, for the B-mode image of the n+1th frame, a kernel (region R22) has already been set in the previous calculation process. Therefore, the calculation function 162 sets a kernel (region R23) corresponding to region R22 for the B-mode image of the n+2th frame. For example, the center position of the kernel (region R22) of the n+1th frame is (x k +x1,y k +y1), and the positional shift between the images of the n+1th and n+2th frames is (x2, y2). Therefore, the calculation function 162 calculates the pixel position (x k +x1+x2,y k The kernel (region R23) is set with the center at the point (y1+y2).

[0065] Then, the calculation function 162 calculates a first index value for the n+1th frame based on the correlation coefficient between the image signal in the kernel of the n+1th frame and the image signal in the kernel of the n+2th frame. For example, the calculation function 162 calculates the correlation coefficient using the pixel value for each pixel index in the region R22 and the pixel value for each pixel index in the region R23 as two-variable data. Then, the calculation function 162 calculates the first index value for the n+1th frame by subtracting the calculated correlation coefficient from "1".

[0066] Thus, the calculation function 162 calculates the pixel location (x k ,y k ), calculates first index values ​​for the nth frame and the (n+1)th frame. The calculation function 162 also performs similar processing for other pixel positions within the ROI, thereby calculating first index values ​​for all pixel positions within the ROI. The calculation function 162 then performs similar processing for the remaining B-mode images of the N B-mode images, thereby calculating multiple first index values ​​arranged in chronological order. The calculated multiple first index values ​​are associated with the frame numbers (time phases) and pixel positions and stored in the storage circuitry 150 or a memory (not shown).

[0067] In FIG. 7, the reason why "1-correlation coefficient" is defined as the first index value is to define an index value that increases as the fluctuation increases. With this definition, the magnitude relationship of the first index value and the magnitude relationship of the fluctuation are linked, so that when output as a parametric image described later, a region (pixel) with large fluctuation can be depicted with high brightness. However, the first index value is not necessarily limited to "1-correlation coefficient". For example, the calculation function 162 can also output the value of the correlation coefficient as the first index value as it is. In this case, although the magnitude relationship of the first index value and the magnitude relationship of the fluctuation are inversely related, it can be used as an index value that indicates the degree of fluctuation. In addition, the calculation function 162 may define the first index value using a statistical value that indicates the similarity (dissimilarity) between image signals. For example, the calculation function 162 can calculate the correlation coefficient, 1-correlation coefficient, absolute coefficient, 1-absolute coefficient, sum of absolute values ​​of difference (SAD), or sum of squared difference (SSD) between image signals in two medical images as the first index value. That is, in this embodiment, the first index value indicating similarity may be a value (similarity) that increases with increasing similarity, or a value (dissimilarity) that increases with increasing dissimilarity. However, when quantifying the similarity of regions with different signal intensities, such as hemangioma and background liver, it is preferable to use a correlation coefficient as a calculation method that is less dependent on signal intensity. The correlation coefficient may also be calculated based on zero-mean normalized cross-correlation.

[0068] 7, as an example, a case where the first index value is calculated using two B-mode images of adjacent frames (i.e., one frame interval) has been described, but the embodiment is not limited to this. For example, the calculation function 162 may select two B-mode images of any frame interval, such as two frames interval, three frames interval, etc., as an image pair, and calculate the first index value using this image pair.

[0069] In addition, in Fig. 7, as an example, a case where the first index value is calculated using a kernel consisting of 9 pixels of 3 x 3 has been described, but the embodiment is not limited to this. The number of pixels included in the kernel can be set arbitrarily. For example, the calculation function 162 can calculate the first index value using a kernel of one pixel of 1 x 1 as the smallest unit kernel.

[0070] In step S208, the calculation function 162 calculates a second index value based on the first index values ​​at each pixel position. For example, the calculation function 162 calculates a second index value corresponding to a statistical value of the first index value in the time direction based on the first index values ​​calculated for each of the multiple image pairs. Specifically, the calculation function 162 calculates a second index value for each of the multiple positions using the first index value corresponding to each of the multiple positions in the region of interest.

[0071] The process of step S208 will be described with reference to Fig. 8. Fig. 8 illustrates a process for calculating a second index value based on a time series of first index values ​​at each pixel position in the ROI (region R20). The two graphs shown in Fig. 8 show the time series changes in the first index value at a representative point in the ROI (region R20). In these graphs, the horizontal axis corresponds to time (frames) and the vertical axis corresponds to the first index value (1-correlation coefficient).

[0072] As shown in FIG. 8, at the representative point of the background liver, the correlation coefficient is relatively high (close to 1) in every frame, so the first index value is low in every frame. On the other hand, at the representative point of the hemangioma, the correlation coefficient is often lower than that of the background liver, so the first index value is generally higher than that of the background liver. Therefore, the calculation function 162 calculates the statistical value of the first index value in the time direction as the second index value. In other words, it can be said that the second index value is an index value related to the state of the hemangioma.

[0073] For example, the calculation function 162 calculates the standard deviation of the first index values ​​in the time direction as the second index value. As a result, the calculation function 162 calculates one second index value for each pixel position in the ROI (region R20). The calculated second index value is associated with the pixel position and stored in the memory circuit 150 or a memory (not shown).

[0074] In FIG. 8, the case where the standard deviation is calculated as the second index value has been described, but the embodiment is not limited thereto. For example, the calculation function 162 may calculate the variance, average value, median value, correlation coefficient, absolute coefficient, or moving average value of the first index value in the time direction as the second index value. When calculating the moving average value, it is preferable to hold and output the maximum or minimum value of the moving average values ​​calculated in the time direction. In addition, the calculation function 162 does not necessarily have to calculate the statistical value of the first index value in the time direction as the second index value. For example, the calculation function 162 can also calculate the difference between the first index values ​​(for example, the difference between adjacent first index values ​​in the time direction) as the second index value.

[0075] Returning to the description of Fig. 4, in step S105, the generation function 163 generates a parametric image. For example, the generation function 163 generates a parametric image indicating the spatial distribution of the second index value. Note that the parametric image is an example of an index image.

[0076] The process of step S105 will be described with reference to Fig. 9. Fig. 9 illustrates an example of a parametric image generated by the generation function 163. Fig. 9 illustrates a case where a parametric image I10 is generated based on the second index value calculated for the ROI (region R20) in Fig. 8.

[0077] 9, the generating function 163 generates a parametric image I10 by assigning a pixel value (color) corresponding to the magnitude of the second index value of each pixel position included in the ROI (region R20) to each pixel position in the ROI. In other words, a high-luminance region R24 in the parametric image I10 suggests the presence of a hemangioma.

[0078] The parametric image I10 is preferably displayed superimposed on a B-mode image I20 as a background image. As the B-mode image I20, an image of any frame can be appropriately selected from among the N B-mode images.

[0079] In step S106, the output control function 164 displays the parametric image. For example, the output control function 164 causes the parametric image I10 generated by the generation function 163 to be displayed on the display 103. Then, the processing circuit 160 ends the process.

[0080] 4 and 5 are merely examples, and the present invention is not limited to the illustrated processing procedures. For example, the smoothing process in step S201 does not necessarily have to be performed.

[0081] Also, for example, in FIG. 4, a case where a parametric image is displayed as an example of an output form has been described, but the embodiment is not limited to this. For example, the output control function 164 may output a representative value of the second index value. In this case, the calculation function 162 calculates a representative value of the measurement region based on the second index value of each position in the measurement region corresponding to at least a part of the ROI. For example, when the operator sets the region R24 in FIG. 9 as the measurement region, the calculation function 162 calculates a representative value using the second index value of each pixel in the region R24. Here, the representative value is, for example, a statistical value such as an average value, a median value, a maximum value, or a minimum value. Then, the output control function 164 causes the display 103 to display the representative value of the region R24 calculated by the calculation function 162.

[0082] Also, for example, the output control function 164 may display a histogram of the second index values. This histogram is, for example, a graph with the frequency (number of pixels) on the vertical axis and the magnitude of the second index value on the horizontal axis.

[0083] That is, the output control function 164 can output in an output form appropriately selected from among a parametric image, a representative value, a histogram, etc. In addition, the output destination of the output by the output control function 164 is not limited to the display 103. For example, the output control function 164 may store the information to be output in the storage circuitry 150, or may transmit it to an external device.

[0084] As described above, in the ultrasound diagnostic device 1 according to the first embodiment, the acquisition function 161 acquires a plurality of time-series medical images. For each image pair consisting of two medical images included in the plurality of medical images, the calculation function 162 calculates a first index value indicating the similarity between the image signals in the two medical images. In addition, the calculation function 162 calculates a second index value based on the first index value calculated for each of the plurality of image pairs. This allows the ultrasound diagnostic device 1 to evaluate the change in signal intensity over time.

[0085] For example, the ultrasound diagnostic device 1 calculates a first index value based on the "strength of correlation (similarity) between frames." In addition, the ultrasound diagnostic device 1 calculates a second index value based on the "relationship in the frame direction of correlation." This allows the ultrasound diagnostic device 1 to quantitatively evaluate changes in signal intensity over time.

[0086] As an example, an operator (doctor) may find a region (shadow) suspected to be a hemangioma in a B-mode image. In such a case, the operator uses the ultrasound diagnostic device 1 according to the first embodiment to display a parametric image on the display 103. Here, as shown in the region R24 in FIG. 9, when a region is highlighted with a luminance value equal to or higher than a predetermined value, the operator can determine that a change in signal intensity over time is recognized in the region. As a result, the operator can determine that the shadow in the B-mode image is a hemangioma. This allows the operator to easily identify a hemangioma by ultrasonic scanning of the B-mode image without performing differentiation using ultrasonic contrast. Therefore, the ultrasound diagnostic device 1 can identify a hemangioma even in a routine examination.

[0087] (Modification 1 of the first embodiment) For example, in FIG. 9, a case where a parametric image I10 is generated based on the second index value is described, but the embodiment is not limited to this. For example, the generation function 163 can generate a plurality of parametric images showing the spatial distribution of the first index value. Here, this parametric image is an image (parametric image of similarity) showing the similarity (correlation) between frames. In addition, the output control function 164 can display the generated plurality of parametric images of similarity side by side or in chronological order.

[0088] For example, the generating function 163 generates a similarity parametric image by assigning a pixel value according to the magnitude of the first index value to each pixel position. Here, since the first index value is time-series information, the similarity parametric image becomes a plurality of images (moving image) arranged in time series. In the similarity parametric image, the area where the hemangioma exists is depicted with a relatively high brightness. Furthermore, when the moving image of the similarity parametric image is played, the brightness value changes from high to low in the area where the hemangioma exists.

[0089] The first index value can be output not only as a parametric image of similarity but also as a graph. For example, the output control function 164 causes the display 103 to display a graph (similar to the graph in FIG. 8) showing the time series change of the first index value at an arbitrary position designated by the operator.

[0090] In addition, the first index value can be output in the form of a representative value, histogram, etc., not limited to a parametric image or graph of similarity. The output form of the representative value or histogram is the same as the output form of the representative value or histogram of the second index value described above, so a description thereof will be omitted.

[0091] (Modification 2 of the first embodiment) Also, for example, the output control function 164 may display the scatter diagrams shown in FIG. 2 and FIG.

[0092] For example, when the operator specifies a point on a B-mode image, the generating function 163 sets a kernel centered on the specified point. The generating function 163 then generates multiple scatter plots arranged in chronological order by plotting points corresponding to each pixel position in the kernel between adjacent frames. When there are N B-mode images, N-1 scatter plots are generated. The output control function 164 then displays the multiple generated scatter plots side by side or in chronological order. For example, at a position (point) where a hemangioma exists, scatter plots such as SP11 to SP13 are displayed. On the other hand, at a position where no hemangioma exists, scatter plots such as SP21 to SP23 are displayed.

[0093] This allows the operator to check the strength of correlation on the scatter diagram. Also, by viewing the scatter diagrams in the frame direction, the operator can check whether the correlation in each scatter diagram changes in the frame direction. As a result, the operator can determine whether a hemangioma is present at the specified position.

[0094] (Modification 3 of the first embodiment) Also, for example, in order to ensure a sufficient number of frames of B-mode images used for calculating the second index value, an indicator showing the minimum scan time can be displayed when capturing a B-mode image.

[0095] For example, when the operator brings the ultrasonic probe 101 into contact with the body surface of the subject P and starts ultrasonic scanning, an indicator is displayed. This indicator may be, for example, in the form of a countdown corresponding to the minimum scan time, or in the form of a notification when the minimum scan time has elapsed.

[0096] Second Embodiment In the above-mentioned first embodiment, the first index values ​​of all frames are used as the calculation target of the second index value, but the embodiment is not limited to this. For example, the ultrasound diagnostic device 1 can calculate the second index value using the first index value of the frame designated as the calculation target among N B-mode image data. For example, when there is a motion of a nature that cannot be corrected by the motion stabilizer process, it is preferable to designate the frame to be the calculation target.

[0097] For example, when the entire tissue depicted in the image is moving in a uniform translational manner, the motion of the tissue can be detected as a positional deviation and corrected by the motion stabilizer process. However, biological tissue does not necessarily move in a uniform translational manner. For example, even when the tissue in the image is affected by pulsation, there may be areas that are easily affected by the pulsation and areas that are not easily affected by the pulsation. In such a case, since the tissue in the image does not move in a uniform translational manner, even if the motion stabilizer process is performed, the correction cannot be completed and the tissue remains. In this way, when there is a tissue motion that remains even after the motion stabilizer process (hereinafter referred to as "residual tissue motion"), it becomes difficult to determine the area where the hemangioma exists.

[0098] A parametric image in the case where residual tissue motion is included will be described with reference to Fig. 10. Fig. 10 is a diagram showing an example of a parametric image in the case where residual tissue motion is included. A parametric image I30 is displayed superimposed on a B-mode image I40 as a background image.

[0099] 10, when residual tissue motion is included, the parametric image I30 may include high-intensity areas in areas other than the area R30 where the hemangioma exists. This is thought to be because the residual tissue motion increases the value of the first index value (1-correlation coefficient).

[0100] Therefore, in the second embodiment, a case will be described in which, among the N pieces of B-mode image data, frames that are significantly affected by residual tissue motion are excluded to calculate the second index value. Note that specifying frames to be excluded (reject frames) among the N pieces of B-mode image data is essentially synonymous with specifying frames to be calculated.

[0101] The ultrasound diagnostic device 1 according to the second embodiment has a similar configuration to the ultrasound diagnostic device 1 illustrated in Fig. 1, and is different in part in the index value calculation process illustrated in Fig. 4. Therefore, in the second embodiment, the differences from the first embodiment will be mainly described, and the description of the configuration described in the first embodiment will be omitted.

[0102] The processing procedure of the ultrasound diagnostic apparatus 1 according to the second embodiment will be described with reference to Fig. 11. Fig. 11 is a flowchart showing the processing procedure of the index value calculation processing according to the second embodiment. The processing procedure shown in Fig. 11 corresponds to the processing content of step S104 shown in Fig. 4.

[0103] In the following description, the description will be made with reference to Fig. 12 to Fig. 14. Fig. 12 is a diagram for explaining a process of setting a reference region according to the second embodiment. Fig. 13 is a diagram showing an example of a time series change in the third index value according to the second embodiment.

[0104] In step S301, the calculation function 162 executes smoothing processing in the spatial direction and the time direction for a plurality of medical images. Note that the processing in step S301 is similar to the processing in step S201 shown in FIG. 5, and therefore the description thereof will be omitted.

[0105] In step S302, the calculation function 162 sets the ROI and the reference region. For example, as shown in FIG. 12, the calculation function 162 displays one B-mode image (typically the first frame image) of the N B-mode images after the smoothing process on the display 103. The operator designates an area including a suspected area of ​​hemangioma and an unsuspected area (background liver) on the displayed B-mode image as the ROI (region R40). The operator also designates the background liver area as the reference region (region R50). The calculation function 162 sets the area designated by the operator on the B-mode image as the ROI and the reference region. The calculation function 162 also sets the ROI and the reference region in the remaining B-mode images (for example, images from the second to the Nth frames) based on the ROI and the reference region of the first frame. It is preferable to set the reference region to a region several times larger than the size of the speckle.

[0106] Although the case where the ROI and the reference region are set manually has been described here, they may be set automatically. For example, the calculation function 162 can extract the contour of a structure in an image by a known segmentation process, and set the ROI and the reference region in the image based on the position and shape of the extracted contour.

[0107] In step S303, the calculation function 162 executes a motion stabilizer process. Note that the process of step S303 is similar to the process of step S203 shown in FIG. 5, and therefore a description thereof will be omitted.

[0108] Then, when step S303 ends, the calculation function 162 executes the processes of steps S308 to S311 in parallel with the processes of steps S304 to S307. The processes of steps S308 to S311 are processes for calculating a third index value indicating the similarity between the image signals. The processes of steps S304 to S307 are similar to the processes of steps S204 to S207 shown in FIG. 5, and therefore description thereof will be omitted.

[0109] In step S308, the calculation function 162 selects an image pair to be processed. This process is similar to the process in step S104 shown in FIG. 5, so a description thereof will be omitted.

[0110] In step S309, the calculation function 162 corrects the motion of the image pair. This process is similar to the process in step S105 shown in FIG. 5, so a description thereof will be omitted.

[0111] In step S310, the calculation function 162 calculates a third index value indicating the similarity between the image signals for the reference region.

[0112] For example, a case will be described in which the nth frame and the n+1th frame B-mode images are selected as the image pair to be processed. In this case, the calculation function 162 calculates the third index value of the nth frame based on the correlation coefficient between the image signal in the reference region of the nth frame and the image signal in the reference region of the n+1th frame.

[0113] For example, a pixel index for identifying a pixel position is assigned to each pixel position in the reference area. The calculation function 162 calculates a correlation coefficient by treating the pixel value for each pixel index in the reference area of ​​the nth frame and the pixel value for each pixel index in the reference area of ​​the n+1th frame as two-variable data. The calculation function 162 then calculates the third index value for the nth frame by subtracting the calculated correlation coefficient from "1."

[0114] In step S311, the calculation function 162 determines whether or not there are any image pairs remaining to be processed. For example, when there are any unprocessed B-mode images among the N B-mode images, the calculation function 162 determines that there are any image pairs remaining to be processed (Yes in step S311), and proceeds to processing in step S308. On the other hand, when there are no unprocessed B-mode images among the N B-mode images, the calculation function 162 determines that there are no image pairs remaining to be processed (No in step S311), and proceeds to processing in step S312.

[0115] That is, the calculation function 162 repeatedly executes the processes of steps S308 to S311 while changing the image pair to be processed, thereby sequentially calculating the third index value based on each image pair.

[0116] In this way, the calculation function 162 calculates a plurality of third index values ​​arranged in time series based on the reference regions of each of the N B-mode images. The calculated third index values ​​are associated with the frame numbers (time phases) and stored in the memory circuitry 150 or a memory (not shown).

[0117] 11, "1-correlation coefficient" is defined as the third index value, but the third index value is not limited to this. For example, the calculation function 162 may calculate the correlation coefficient between image signals in two medical images, 1-correlation coefficient, absolute coefficient, 1-absolute coefficient, SAD, or SSD as the third index value. However, it is preferable to use the same definition as the first index value for the definition of the third index value.

[0118] In step S312, the output control function 164 displays the time series change of the third index value. For example, the output control function 164 reads out from the memory circuitry 150 a plurality of third index values ​​arranged in time series calculated by the calculation function 162. Then, the output control function 164 generates a graph showing the time series change of the third index value based on the read out plurality of third index values ​​arranged in time series, and displays the graph on the display 103.

[0119] In step S313, the calculation function 162 accepts the designation of frames to be calculated. For example, the operator views a graph showing the time series change of the third index value displayed on the display 103, and inputs the designation of frames to be excluded from the calculation target (reject frames). The calculation function 162 accepts the input designating the reject frames.

[0120] Selection of frames to be calculated according to the second embodiment will be described with reference to Fig. 13. The upper part of Fig. 13 illustrates a graph showing the time series change in the third index value in the reference region. The middle part of Fig. 13 illustrates a graph showing the time series change in the first index value at a representative point of the hemangioma. The lower part of Fig. 13 illustrates a graph showing the time series change in the first index value at a representative point of the background liver. In the graphs of Fig. 13, the horizontal axis corresponds to time, and the vertical axis corresponds to the first index value or the third index value (i.e., the value of "1-correlation coefficient").

[0121] As shown in the upper part of Fig. 13, the time series change in the third index value in the reference region represents the movement of the entire tissue including the background liver. In other words, a frame with a large third index value indicates a large movement of the entire tissue, and a frame with a small third index value indicates a small movement of the entire tissue.

[0122] Here, when the residual tissue motion is small (when the motion of the entire tissue can be roughly corrected by the motion stabilizer process), the third index value of the reference region should be small. However, when there is a frame in which the third index value of the reference region is large, it is found that the residual tissue motion exists in that frame. For example, the third index values ​​are large in the periods T10, T11, and T12, so it is considered that the residual tissue motion exists (upper part of FIG. 13).

[0123] Residual tissue motion may also affect the signal intensity of nearby hemangiomas and background liver. For example, the first index values ​​at the representative points of hemangiomas and background liver are generally large in periods T10, T11, and T12 (middle and lower rows of FIG. 13). In addition, the curves of the first index values ​​included in periods T10, T11, and T12 are similar to the curve shapes of the third index values ​​in the same periods. This suggests that the areas where hemangiomas and background livers exist are affected by residual tissue motion. Therefore, it can be said that frames including residual tissue motion are undesirable frames to be calculated.

[0124] Therefore, the output control function 164 displays a graph showing the time series change of the third index value (the graph in the upper part of FIG. 13) on the display. The output control function 164 may generate and display not only the graph showing the time series change of the third index value, but also a graph showing the time series change of the first index value at a position specified by the operator. In this case, the operator can arbitrarily specify a position where a hemangioma is thought to exist or a position where it is thought not to exist.

[0125] Then, the operator performs an input to designate the frames included in the periods T10, T11, and T12 as reject frames. As a result, the calculation function 162 accepts the input to designate the frames included in the periods T10, T11, and T12 as reject frames. Note that, although a case has been described here in which the operator designates the reject frames, it is also possible to designate frames to be calculated.

[0126] In step S314, the calculation function 162 calculates the second index value using the first index value of the frame to be calculated. For example, the calculation function 162 sets the first index value of the frames included in the periods T10, T11, and T12 among the N B-mode images to "null". Then, the calculation function 162 calculates the second index value using the first index value of the frames other than the periods T10, T11, and T12.

[0127] In this way, the ultrasound diagnostic device 1 according to the second embodiment calculates the second index value by excluding the first index value of the frame having a large influence of residual tissue motion from the N B-mode image data from the calculation target, which allows the ultrasound diagnostic device 1 to calculate the second index value more accurately.

[0128] An example of a display image according to the second embodiment will be described with reference to FIG. 14. FIG. 14 is a diagram showing an example of a display image according to the second embodiment. FIG. 14 illustrates a display image in the case where residual tissue motion is included. In FIG. 14, region R31 is a region for cine-displaying a B-mode image. Region R32 is a region for displaying a parametric image, and a B-mode image of an arbitrary time phase is displayed as a background image of the parametric image. Region R33 is a region for displaying a bar indicating a cine memory.

[0129] 14, the output control function 164 displays a parametric image I31 on the display 103. A region R34 in the parametric image I31 has many high-luminance pixels, suggesting the presence of a hemangioma. In addition, the area outside the region R34 is depicted as low-luminance because no hemangioma is present.

[0130] In this way, the ultrasound diagnostic apparatus 1 according to the second embodiment can clearly distinguish and display the area R34 where the hemangioma is present from the area (outside the area R34) where the hemangioma is not present, even when residual tissue motion is present.

[0131] Furthermore, the output control function 164 displays information indicating the rejected frames. For example, the output control function 164 displays positions corresponding to the rejected frames in a bar indicating the cine memory in region R33 as black regions. These black regions indicate frames corresponding to periods T10, T11, and T12 in Fig. 13. This allows the operator to grasp the rejected frames on the bar indicating the cine memory.

[0132] The black area indicating the rejected frame does not necessarily have to be black. It can be displayed in a color or hatching that can be distinguished from other areas. Furthermore, the information indicating the rejected frame does not necessarily have to be displayed on the bar indicating the cine memory. For example, it can be displayed on the display 103 as a numerical value indicating the imaging time, the number of frames, or the like.

[0133] Also, instead of displaying the rejected frames, the frames that were the subject of the calculation may be displayed. Information indicating the frames that were the subject of the calculation can be displayed in the same display format as information indicating the rejected frames.

[0134] In the second embodiment, the process of calculating the first index value (steps S304 to S307) and the process of calculating the third index value (steps S308 to S311) are performed in parallel, but the embodiment is not limited to this. For example, the ultrasound diagnostic device 1 may perform the process of calculating the first index value and then the process of calculating the third index value, or may perform the process of calculating the third index value and then the process of calculating the first index value.

[0135] In addition, when the process of calculating the first index value is executed after the process of calculating the third index value is executed, it is also possible to calculate the first index value by using the B-mode image group from which the rejected frame is excluded as the processing target. In this case, the calculation function 162 selects an image pair to be processed from the B-mode image group from which the rejected frame is excluded (processing of step S304). Then, the calculation function 162 executes the processes of steps S305 and S306 on the selected image pair, and determines whether an image pair to be processed remains (processing of step S307). In the process of step S307, when an unprocessed B-mode image exists in the B-mode image group from which the rejected frame is excluded, the calculation function 162 determines that an image pair to be processed remains. On the other hand, when an unprocessed B-mode image does not exist in the B-mode image group from which the rejected frame is excluded, the calculation function 162 determines that an image pair to be processed does not remain. As a result, the ultrasound diagnostic device 1 does not execute the process of calculating the first index value for frames that are not the calculation target, and therefore the load of the process of calculating the first index value can be reduced.

[0136] (Modification of the second embodiment) In the second embodiment, the case where the operator manually specifies the reject frames has been described, but the embodiment is not limited to this. For example, the ultrasound diagnostic device 1 can automatically determine the reject frames.

[0137] That is, in the ultrasound diagnostic device 1 according to the modified example of the second embodiment, the calculation function 162 sets a reference region in each of the two medical images. Then, the calculation function 162 calculates a third index value indicating a similarity between image signals in the two medical images for the reference region. Then, the calculation function 162 determines a frame to be calculated from among the multiple medical images based on the third index value. Then, the calculation function 162 calculates a second index value using the first index value of the frame to be calculated.

[0138] For example, the calculation function 162 according to the modified example of the second embodiment executes a process of determining frames to be calculated based on the third index value, instead of the process of step S313 in Fig. 11. Note that the calculation function 162 does not necessarily have to execute the process of displaying the time series change in the third index value (the process of step S312), but it is preferable to display the time series change in the third index value in order to present the operator with the basis for determining the rejected frames.

[0139] Here, if the movement of the entire tissue is properly corrected, the third index value is considered to have the following two characteristics.

[0140] The first characteristic is that the third index value does not exceed a certain level. This is based on the assumption that, if the movement of the entire tissue is corrected, there may be a certain degree of difference, but there will be no difference greater than a certain level. Therefore, frames with a third index value greater than the average value are excluded.

[0141] The second feature is that the third index value does not vary. In other words, the standard deviation of the third index value should be within a certain range. Therefore, frames whose standard deviation of the third index value is greater than or equal to the average value of the standard deviation are excluded.

[0142] Therefore, the calculation function 162 determines a frame that satisfies both the first feature and the second feature as a calculation target. For example, the calculation function 162 sets a first threshold based on the first feature "the third index value does not become higher than a certain value." The calculation function 162 also sets a second threshold based on the second feature "the third index value does not vary." Then, the calculation function 162 determines a frame to be calculated based on a comparison of the third index value with each of the first and second thresholds.

[0143] The processing of the calculation function 162 according to the modified example of the second embodiment will be described with reference to Fig. 15. Fig. 15 is a diagram for explaining the processing of the calculation function 162 according to the modified example of the second embodiment. Graph G10 is a graph showing the time series change of the third index value. The processing for generating this graph is similar to the processing of the second embodiment described above, and therefore the description will be omitted.

[0144] For example, the calculation function 162 calculates the average value of the third index values ​​of all frames as the first threshold value based on the first feature. Then, the calculation function 162 compares the third index value with the first threshold value. For example, the calculation function 162 generates graph G11 by subtracting the value of the first threshold value from the third index value of each frame in graph G10. In graph G11, the horizontal axis corresponds to time, and the vertical axis corresponds to the subtracted value. Also, the position of "0 (zero)" on the vertical axis corresponds to the first threshold value.

[0145] Then, the calculation function 162 excludes frames having a value greater than the first threshold value. That is, the calculation function 162 excludes frames having a value greater than "0" in the graph G11, and leaves frames included in the periods T20, T21, and T22.

[0146] Furthermore, the calculation function 162 calculates the average value of the standard deviations of the third index values ​​of all frames as the second threshold value based on the second feature. Then, the calculation function 162 compares the third index value with the second threshold value. For example, the calculation function 162 generates the graph G12 by subtracting the value of the second threshold value from the standard deviation of the third index values ​​of each frame. Here, the standard deviation of the third index values ​​of each frame is calculated as the standard deviation of the third index values ​​of several frames before and after each frame. In the graph G12, the horizontal axis corresponds to time, and the vertical axis corresponds to the subtracted value. Furthermore, the position of "0 (zero)" on the vertical axis corresponds to the second threshold value.

[0147] Then, the calculation function 162 excludes frames having a value greater than the second threshold value. That is, the calculation function 162 excludes frames having a value greater than "0" in the graph G12, and leaves frames included in the period T23.

[0148] Here, frames that satisfy both the first and second characteristics are frames that remain in both graph G11 and graph G12. Since period T23 is included in period T21, frames that satisfy both the first and second characteristics in FIG. 15 are frames that are included in period T23. For example, if frames that satisfy both the first and second characteristics are represented by "1" and frames that do not satisfy both by "0", graph G13 in FIG. 15 is obtained. In other words, calculation function 162 determines frames included in period T23 as the objects of calculation.

[0149] In this way, the ultrasound diagnostic device 1 according to the modification of the second embodiment can automatically determine the frame to be calculated. In the above description, a frame satisfying both the first feature and the second feature is used as the calculation target, but the embodiment is not limited to this. For example, the calculation function 162 may use a frame satisfying at least one of the first feature and the second feature as the calculation target. In addition, the first threshold and the second threshold described above are merely examples, and any threshold can be set. However, it is preferable that the threshold is a value based on the statistical value of the third index value in the time direction.

[0150] (Third embodiment) In the above-mentioned first and second embodiments, the similarity (first index value) of one point (center position) is calculated based on the image signal in the kernel, and the case where the fluctuation of the similarity in the frame direction is focused on is described. However, the half-width of the correlation coefficient distribution depends on the point spread function (PSF) of the imaging system. For example, when the PSF is wide, even if there is a fluctuation due to the fluctuation in the frame direction, the width of the correlation coefficient distribution is wide, so the correlation coefficient does not decrease as expected, and the ability to distinguish the fluctuation may decrease.

[0151] Therefore, in the third embodiment, rather than focusing on the correlation coefficient at a single point, a search area is set and attention is paid to the fluctuation in the correlation coefficient distribution (distribution of similarity) in two-dimensional space in the frame direction, thereby improving the ability to identify fluctuations in the frame direction.

[0152] An ultrasound diagnostic device 1 according to the third embodiment has a configuration similar to that of the ultrasound diagnostic device 1 illustrated in Fig. 1, and differs in part in the index value calculation process illustrated in Fig. 5. Therefore, in the third embodiment, differences from the first embodiment will be mainly described, and a description of the configuration described in the first embodiment will be omitted.

[0153] The processing procedure of the ultrasound diagnostic apparatus 1 according to the third embodiment will be described with reference to Fig. 16. Fig. 16 is a flowchart showing the processing procedure of the index value calculation processing according to the third embodiment. The processing procedure shown in Fig. 16 corresponds to the processing content of step S104 shown in Fig. 4 (i.e., the processing content of Fig. 5).

[0154] In the following description, reference will be made to Fig. 17 to Fig. 20. Fig. 17 and Fig. 18 are diagrams for explaining a process for generating a two-dimensional correlation coefficient distribution according to the third embodiment. Fig. 19 and Fig. 20 are diagrams for explaining a process for calculating correlation coefficients of a two-dimensional correlation coefficient distribution according to the third embodiment.

[0155] In steps S401 to S405, the calculation function 162 executes each process. Note that the processes in steps S401 to S405 are similar to the processes in steps S201 to S205 shown in FIG. 5, and therefore the description thereof will be omitted.

[0156] In step S406, the calculation function 162 generates a two-dimensional correlation coefficient distribution for each pixel position. That is, for each image pair consisting of two medical images included in the multiple medical images, the calculation function 162 calculates a group of index values ​​(fourth group of index values) indicating the similarity between the image signal at each position of one medical image and the image signal at each of multiple positions in the other image. The two-dimensional correlation coefficient distribution is an example of a group of index values.

[0157] In step S407, the calculation function 162 moves the center position of the next kernel to the pixel position with the highest correlation coefficient in the two-dimensional correlation coefficient distribution.

[0158] In step S408, the calculation function 162 determines whether or not there are any image pairs remaining to be processed. For example, when there is an unprocessed B-mode image among the N B-mode images, the calculation function 162 determines that there are any image pairs remaining to be processed (Yes in step S408), and proceeds to processing in step S404. On the other hand, when there is no unprocessed B-mode image among the N B-mode images, the calculation function 162 determines that there are no image pairs remaining to be processed (No in step S408), and proceeds to processing in step S409.

[0159] That is, calculation function 162 repeatedly executes the processes of steps S404 to S408 while changing the image pair to be processed. Then, calculation function 162 selects a plurality of image pairs in order, and calculates a first index value based on each image pair in order.

[0160] The process of steps S404 to S408 will be described with reference to Figs. 17 and 18. Fig. 17 illustrates a process for generating a two-dimensional correlation coefficient distribution for the nth frame from the B-mode images of the nth frame and the n+1th frame. In Fig. 17, a region R40 indicates a kernel set in the B-mode image of the nth frame. Moreover, a region R41 indicates a search region set in the B-mode image of the n+1th frame. Moreover, a region R42 indicates a kernel set in the B-mode image of the n+1th frame. Fig. 18 illustrates a process for generating a two-dimensional correlation coefficient distribution for the n+1th frame from the B-mode images of the n+1th frame and the n+2th frame. In Fig. 18, a region R43 indicates a kernel set in the B-mode image of the n+1th frame. Moreover, a region R44 indicates a search region set in the B-mode image of the n+2th frame. Moreover, a region R45 indicates a kernel set in the B-mode image of the n+2th frame.

[0161] First, a case where the n-th and (n+1)-th frame B-mode images are selected as an image pair to be processed will be described with reference to Fig. 17. In this case, the calculation function 162 calculates the pixel position (x p ,y p ) is set as the center of the kernel (region R40).

[0162] Then, the calculation function 162 calculates the pixel position (x p ,y p ) is set as the center of the search region (region R41). Here, when the positional shift between the nth frame image and the n+1th frame image is (x1, y1), the calculation function 162 calculates the pixel position (x p +x1,y p A search area (area R41) is set with the center at point +y1.

[0163] Then, the calculation function 162 calculates the pixel position (x p ,y p) and the image signal of each pixel position in the search region in the B-mode image of the n+1th frame. For example, the calculation function 162 generates a two-dimensional correlation coefficient distribution between the image signal of the pixel position (x q ,y q ) is set as the center of the kernel (region R42). Then, the calculation function 162 calculates the pixel position (x q ,y q For example, the calculation function 162 calculates the correlation coefficient by treating the pixel value for each pixel index in the region R40 and the pixel value for each pixel index in the region R42 as two-variable data.

[0164] Then, the calculation function 162 calculates the pixel position (x q ,y q ) is changed, and the image signal at each changed pixel position and the pixel position (x p ,y p ) image signal. In this way, the calculation function 162 calculates the correlation coefficient between the image signal at each pixel position in the search region (region R41) and the image signal at the pixel position (x p ,y p 17 is generated as a two-dimensional correlation coefficient distribution for the nth frame by assigning a luminance value to each pixel position in the search region (region R41) according to the magnitude of the correlation coefficient at each pixel position.

[0165] Then, the calculation function 162 moves the center position of the kernel in the next frame to the pixel position with the highest correlation coefficient in the two-dimensional correlation coefficient distribution. For example, the calculation function 162 moves the center position of the kernel in the next frame to the pixel position with the highest brightness P11 (x r ,y r ), the center position of the kernel in the next frame is moved to the coordinate system in the two-dimensional correlation coefficient distribution corresponding to the coordinate system in the B-mode image of the n+1th frame. r ,yr ) is the pixel position (x r ,y r ) corresponding to the pixel position (x p ,y p ) can be moved (tracked) to the position with the highest similarity within the search area in the n+1th frame.

[0166] Next, a case where the B-mode images of the n+1th and n+2th frames are selected as an image pair to be processed will be described with reference to FIG. 18. Here, for the B-mode image of the n+1th frame, the central position of the kernel (x r ,y r ) has already been set. Therefore, the calculation function 162 calculates the pixel position (x r ,y r ) is set as the center of the kernel (region R43).

[0167] Then, the calculation function 162 calculates the pixel position (x r ,y r ) is set as the center of the search region (region R44). Here, if the positional shift between the images of the n+1th and n+2th frames is (x2, y2), the calculation function 162 calculates the pixel position (x r +x2,y r A search area (area R44) is set with the center at y=0.01 + y2.

[0168] Then, the calculation function 162 calculates the pixel position (x r ,y r ) and the image signal of each pixel position in the search region in the B-mode image of the n+2th frame. For example, the calculation function 162 generates a two-dimensional correlation coefficient distribution between the image signal of the pixel position (x s ,y sThe calculation function 162 then calculates the pixel position (x s ,y s For example, the calculation function 162 calculates the correlation coefficient by treating the pixel value for each pixel index in the region R43 and the pixel value for each pixel index in the region R45 as two-variable data.

[0169] Then, the calculation function 162 calculates the pixel position (x s ,y s ) is changed, and the image signal at each changed pixel position and the pixel position (x r ,y r ) image signal. In this way, the calculation function 162 calculates the correlation coefficient between the image signal at each pixel position in the search region (region R44) and the image signal at the pixel position (x r ,y r ) is generated as a two-dimensional correlation coefficient distribution for the n+1th frame. Note that the two-dimensional correlation coefficient distribution shown in Fig. 18 is shown by assigning a luminance value to each pixel position in the search region (region R44) according to the magnitude of the correlation coefficient at each pixel position.

[0170] Then, the calculation function 162 moves the center position of the kernel in the next frame to the pixel position with the highest correlation coefficient in the two-dimensional correlation coefficient distribution. For example, the calculation function 162 moves the center position of the kernel in the next frame to the pixel position with the highest brightness P12 (x t ,y t ), where the coordinate system in the two-dimensional correlation coefficient distribution corresponds to the coordinate system in the B-mode image of the (n+2)th frame. t ,y t ) is the pixel position (x t ,y t )

[0171] Thus, the calculation function 162 calculates the pixel location (x p ,y p ), and generates two-dimensional correlation coefficient distributions for the nth and n+1th frames. The calculation function 162 also performs similar processing on other pixel positions within the ROI to generate two-dimensional correlation coefficient distributions for all pixel positions within the ROI. The calculation function 162 then performs similar processing on the remaining B-mode images of the N B-mode images to generate multiple two-dimensional correlation coefficient distributions arranged in time series. The multiple generated two-dimensional correlation coefficient distributions are stored in the storage circuitry 150 or a memory (not shown) in association with the frame number (time phase) and the pixel position.

[0172] In step S409, the calculation function 162 calculates the correlation coefficient of the two-dimensional correlation coefficient distribution for each corresponding pixel position. That is, the calculation function 162 calculates a value indicating the similarity between the two index value groups calculated for the two image pairs as a first index value.

[0173] The process of calculating the correlation coefficient of the two-dimensional correlation coefficient distribution will be described with reference to Fig. 19 and Fig. 20. Fig. 19 illustrates a process of calculating the correlation coefficient using the time series two-dimensional correlation coefficient distribution at the representative point of the background liver. Fig. 20 illustrates a process of calculating the correlation coefficient using the time series two-dimensional correlation coefficient distribution at the representative point of the hemangioma.

[0174] 19 and 20, two-dimensional correlation coefficient distributions for the nth frame, the n+1th frame, and the n+2th frame are generated for each pixel position. Here, the calculation function 162 calculates the correlation coefficient between the correlation coefficient in the two-dimensional correlation coefficient distribution for the nth frame and the correlation coefficient in the two-dimensional correlation coefficient distribution for the n+1th frame. For example, the calculation function 162 calculates the correlation coefficient for the nth frame using the correlation coefficient for each pixel index in the two-dimensional correlation coefficient distribution for the nth frame and the correlation coefficient for each pixel index in the two-dimensional correlation coefficient distribution for the n+1th frame as two-variable data.

[0175] Furthermore, the calculation function 162 calculates the correlation coefficient between the correlation coefficient in the two-dimensional correlation coefficient distribution of the n+1th frame and the correlation coefficient in the two-dimensional correlation coefficient distribution of the n+2th frame. For example, the calculation function 162 calculates the correlation coefficient of the n+1th frame using the correlation coefficient for each pixel index in the two-dimensional correlation coefficient distribution of the n+1th frame and the correlation coefficient for each pixel index in the two-dimensional correlation coefficient distribution of the n+2th frame as two-variable data.

[0176] In this way, the calculation function 162 calculates the correlation coefficient between the pair while changing the pair of the two-dimensional correlation coefficient distribution to be processed. In this way, the calculation function 162 generates time-series information of the correlation coefficient for each pixel position.

[0177] For example, Fig. 19 illustrates a case where time series information of correlation coefficients is generated for the representative point of the background liver. In this case, since the fluctuation is small in the surrounding area including the representative point of the background liver, the two-dimensional correlation coefficient distributions of the nth frame, the n+1th frame, and the n+2th frame are similar to each other. Therefore, as shown in graph G20, the correlation coefficient at the representative point of the background liver is constant at a relatively high value.

[0178] On the other hand, Fig. 20 illustrates an example of generating time series information of correlation coefficients for representative points of hemangioma. In this case, since fluctuations are large in the surrounding area including the representative point of hemangioma, the two-dimensional correlation coefficient distributions of the nth frame, the n+1th frame, and the n+2th frame are often different from each other. Therefore, as shown in graph G21, the correlation coefficient at the representative point of hemangioma tends to be low compared to the background liver, and changes in the frame direction.

[0179] In this way, the calculation function 162 calculates a plurality of correlation coefficients arranged in time series for all pixel positions within the ROI. Note that, in Fig. 19, for the sake of clarity, the same distribution (image) is illustrated as the two-dimensional correlation coefficient distribution for the nth frame, the n+1th frame, and the n+2th frame, but in reality, it is rare for the same distributions to be arranged side by side.

[0180] 19 and 20, the case of calculating the correlation coefficient has been described, but the present invention is not limited to this. The correlation coefficient calculated in the process of step S409 corresponds to the "first index value indicating the similarity between the image signals in the two medical images" described in the first embodiment. Therefore, the calculation function 162 can calculate the 1-correlation coefficient, the absolute coefficient, the 1-absolute coefficient, the SAD, or the SSD instead of the correlation coefficient.

[0181] In step S410, the calculation function 162 calculates the standard deviation in the frame direction. For example, the calculation function 162 calculates the standard deviation in the frame direction using a plurality of correlation coefficients in time series calculated for each pixel position in the ROI.

[0182] The standard deviation calculated in the process of step S410 corresponds to the "second index value corresponding to the statistical value in the time direction of the first index value (similarity)" described in the first embodiment. Therefore, the calculation function 162 can calculate the variance, the average value, the median value, the correlation coefficient, the absolute coefficient, or the moving average value instead of the standard deviation.

[0183] In this way, in the ultrasound diagnostic device 1 according to the third embodiment, the calculation function 162 generates a two-dimensional correlation coefficient distribution for each pixel position in the ROI. Then, the calculation function 162 calculates a first index value indicating the similarity between the two-dimensional correlation coefficient distributions. Then, the calculation function 162 calculates a second index value in the frame direction using a plurality of first index values ​​arranged in time series. This allows the ultrasound diagnostic device 1 to improve the discrimination ability of fluctuations in the frame direction.

[0184] (Modification of the third embodiment) For example, the output control function 164 may further cause a two-dimensional correlation coefficient distribution to be displayed.

[0185] For example, when an operator specifies a point on a B-mode image, the calculation function 162 generates a plurality of two-dimensional correlation coefficient distributions arranged in time series for the specified point. Then, the output control function 164 displays the generated plurality of two-dimensional correlation coefficient distributions arranged in time series order. For example, at a position (point) where a hemangioma exists, the two-dimensional correlation coefficient distribution shown in FIG. 20 is displayed. On the other hand, at a position where a hemangioma does not exist, the two-dimensional correlation coefficient distribution shown in FIG. 19 is displayed.

[0186] This allows the operator to check whether the correlation in each two-dimensional correlation coefficient distribution changes in the frame direction by viewing the two-dimensional correlation coefficient distribution in the frame direction. As a result, the operator can determine whether a hemangioma is present at a specified position.

[0187] (Other embodiments) In addition to the above-described embodiment, the present invention may be embodied in various different forms.

[0188] (Using signals other than B-mode images) For example, in the above-described embodiment, the first index value and the second index value are calculated using a plurality of time-series medical images, but the embodiment is not limited to this. For example, the first index value and the second index value can be calculated using a signal group before being generated as a medical image.

[0189] That is, the acquisition function 161 acquires a plurality of signal groups in a time series. Furthermore, the calculation function 162 calculates a first index value indicating a similarity between signals in each signal group pair consisting of two signal groups included in the plurality of signal groups. Furthermore, the calculation function 162 calculates a second index value corresponding to a statistical value of the first index value in the time direction based on the first index value calculated for each of the plurality of signal group pairs.

[0190] For example, the acquisition function 161 may acquire, as a medical image, B-mode data before being converted into a B-mode image. In other words, the medical image may be information in which the signal values ​​at each position included in the scannable area corresponding to the medical image are collected for each scannable area.

[0191] Moreover, the present invention is not limited to ultrasonic scanning performed while sliding the center of the aperture in the lateral direction for each ultrasonic transmission and reception, such as B mode. For example, ultrasonic scanning may be performed in which the center of the aperture is fixed during a certain packet, and the same region is transmitted and received a plurality of times, and then slid in the lateral direction. This ultrasonic scanning is similar to ultrasonic scanning in color Doppler, but the signal information used is not frequency but amplitude. In other words, the medical image may be information in which the signal value of each position included in the scannable region is collected for each scan line of the scannable region. In this case, since the signal value is acquired for each divided region, the frame interval when calculating the first index value can be narrowed to the PRF (Pulse Repetition Frequency) interval. Therefore, the calculation process of the first index value has a very high time resolution, which is useful when analyzing the correlation of the change in signal intensity at a high flow velocity.

[0192] Also, for example, before the above-mentioned ultrasonic scanning, a push pulse (acoustic radiation force) or external excitation may be applied, and the response of the tissue due to the external force itself may be analyzed. Also, the response caused by the shear wave generated by the external force may be analyzed. In other words, the medical image may be information on the signal values ​​at each position included in the scannable region, which are collected for each scan line after the push pulse or external excitation.

[0193] (Graph view) Furthermore, for example, the ultrasound diagnostic device 1 can display any of the graphs described in the above-mentioned embodiments on the display 103. For example, the generation function 163 generates a graph based on the first index value or the second index value in a predetermined position or area. Here, the position or area is specified by the operator. Then, the output control function 164 displays the generated graph.

[0194] (Medical information processing device) For example, in the above-described embodiment, the disclosed technology is applied to the ultrasound diagnostic device 1, but the embodiment is not limited to this. For example, the disclosed technology may be applied to a medical information processing device 200. The medical information processing device 200 corresponds to, for example, a workstation or a PACS (Picture Archiving Communication System) viewer.

[0195] Fig. 21 is a block diagram showing an example of the configuration of a medical information processing device 200 according to another embodiment. As shown in Fig. 21, the medical information processing device 200 includes an input interface 201, a display 202, a storage circuitry 210, and a processing circuitry 220. The input interface 201, the display 202, the storage circuitry 210, and the processing circuitry 220 are connected to each other so as to be able to communicate with each other.

[0196] The input interface 201 is an input device for receiving various instructions and setting requests from an operator, such as a mouse, a keyboard, a touch panel, etc. The display 202 is a display device for displaying medical images and a GUI for the operator to input various setting requests using the input interface 201.

[0197] The storage circuitry 210 is, for example, a NAND (Not AND) type flash memory or a HDD (Hard Disk Drive), and stores various programs for displaying medical image data and GUIs, and information used by the programs.

[0198] The processing circuitry 220 is an electronic device (processor) that controls the entire processing in the medical information processing device 200. The processing circuitry 220 executes an acquisition function 221, a calculation function 222, a generation function 223, and an output control function 224. The acquisition function 221, the calculation function 222, the generation function 223, and the output control function 224 are recorded in the storage circuitry 210 in the form of programs executable by a computer, for example. The processing circuitry 220 reads out and executes each program to realize a function corresponding to each read program (the acquisition function 221, the calculation function 222, the generation function 223, and the output control function 224).

[0199] The acquisition function 221 can execute basically the same process as the acquisition function 161 shown in Fig. 1. Furthermore, the calculation function 222 can execute basically the same process as the calculation function 162 shown in Fig. 1. Furthermore, the generation function 223 can execute basically the same process as the generation function 163 shown in Fig. 1. Furthermore, the output control function 224 can execute basically the same process as the output control function 164 shown in Fig. 1.

[0200] As a result, for example, in the medical image processing device 200, the acquisition function 221 acquires a plurality of medical images in time series. For each image pair consisting of two medical images included in the plurality of medical images, the calculation function 222 calculates a first index value indicating the similarity between the image signals in the two medical images. In addition, the calculation function 222 calculates a second index value based on the first index value calculated for each of the plurality of image pairs. As a result, the medical image processing device 200 can evaluate the change in signal intensity over time.

[0201] (Application to 3D images) Also, for example, in the above-mentioned embodiment, the processing is described in the case where a two-dimensional image is used, but the present invention is not limited to this and can also be applied to the case where a three-dimensional image (volume data) is used.

[0202] (Using normalized SAD or normalized SSD) In addition, although it has been described that the above-mentioned correlation coefficient, 1-correlation coefficient, absolute coefficient, 1-absolute coefficient, SAD, or SSD can be applied as the first index value, the present invention is not limited thereto. For example, the normalized SAD or normalized SSD can be further applied as the first index value.

[0203] For example, the calculation function 162 calculates the normalized SAD "S1" using the following formula (1). (n) (x i ,y j )" represents the amplitude (equivalent to the pixel value) of the image signal at each pixel position in the kernel of the nth frame. (n+1) (x i ,y j "i" represents an integer from 1 to k, and "j" represents an integer from 1 to l. For example, in the example of FIG. 7, "k" and "l" are both "3".

[0204]

number

[0205] In addition, the calculation function 162 calculates the normalized SSD "S2" by using the following formula (2). (n) (x i ,y j )," "Amp (n+1) (x i ,y j ), “i”, “j”, “k”, and “l” are the same as in formula (1), and therefore the explanation will be omitted.

[0206]

number

[0207] In this way, the calculation function 162 can calculate, as the first index value, a value obtained by normalizing the sum of the absolute values ​​of the differences between the image signals in two medical images, or a value obtained by normalizing the sum of the squares of the differences between the image signals in two medical images. Note that the processing after calculating the first index value is the same as in the above-mentioned embodiment. Also, the above "S1" and "S2" are also called relative errors.

[0208] Moreover, the above "S1" and "S2" can also be applied as the third index value described in the second embodiment. That is, the calculation function 162 can calculate, as the third index value, a value obtained by normalizing the sum of the absolute values ​​of the differences between the image signals in two medical images, or a value obtained by normalizing the sum of the squares of the differences between the image signals in two medical images.

[0209] The term "processor (circuit)" used in the above description means a circuit such as a CPU (Central Processing Unit), a GPU (Graphics 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 function by reading and executing a program stored in the memory circuit 150. Instead of storing a program in the memory circuit 150, the program may be directly built into the circuit of the processor. In this case, the processor realizes its function by reading and executing a program built into the circuit. Each processor in this embodiment is not limited to being configured as a single circuit for each processor, but may be configured as one processor by combining multiple independent circuits to realize its function. Furthermore, multiple components in each figure may be integrated into one processor to realize its function.

[0210] Moreover, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or a part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc. For example, the function of the image processing circuit 130 may be integrated into the processing circuit 160. Furthermore, all or any part of each processing function 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.

[0211] Furthermore, among the processes described in the above-mentioned embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by a known method. In addition, the information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified.

[0212] The image processing method described in the above embodiment can be realized by executing a prepared image processing program on a computer such as a personal computer or a workstation. This image processing program can be distributed via a network such as the Internet. This image processing program can also be recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, or a DVD, and executed by being read from the recording medium by the computer.

[0213] According to at least one of the embodiments described above, it is possible to evaluate the change in signal strength over time.

[0214] Although some embodiments of the present invention 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, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope of the invention and its equivalents described in the claims, as well as in the scope and spirit of the invention. [Explanation of symbols]

[0215] 1. Ultrasound diagnostic equipment 160 Processing Circuit 161 Acquisition Function 162 Calculation functions 163 Generation function 164 Output control function

Claims

1. an image acquisition unit for acquiring a plurality of time-series ultrasound images; an image generating unit that acquires, at a plurality of time phases, index values ​​indicating a correlation between a signal value at a first position in the ultrasound image at one time phase and a signal value at a second position corresponding to the first position in the ultrasound image at another time phase, and generates an index image in which pixel values ​​according to a change in the correlation in a time axis direction are assigned to each position based on the index values ​​acquired at the plurality of time phases; An analysis device equipped with:

2. an image acquisition unit for acquiring a plurality of time-series ultrasound images; an image generator that acquires, at a plurality of time phases, index values ​​indicating a correlation of a change in signal intensity between a first position in the ultrasound image at one time phase and a second position corresponding to the first position in the ultrasound image at another time phase, and generates an index image that indicates a change in the correlation in a time axis direction based on the index values ​​acquired at the plurality of time phases; An analysis device equipped with:

3. The image generator superimposes the index image on the ultrasound image. The analysis device according to claim 1 or 2.

4. The image generating unit includes: Setting a region of interest in the ultrasound image; obtaining the index value for each of a plurality of positions within the region of interest; The analysis device according to claim 1 or 2.

5. the image generating unit calculates, as the index value, a correlation coefficient between the image signals at the first position and the second position, a 1-correlation coefficient, an absolute coefficient, a 1-absolute coefficient, a sum of absolute values ​​of differences, a sum of squared differences, a normalized value of the sum of absolute values ​​of differences, or a normalized value of the sum of squared differences; The analysis device according to claim 1 or 2.

6. The image generating unit calculates a standard deviation, a variance, a mean value, a median value, a correlation coefficient, or an absolute coefficient as a change in the index value in a time axis direction acquired in a plurality of time phases. The analysis device according to claim 5.

7. The image generating unit includes: Accepting an input to designate a frame to be calculated among the plurality of ultrasound images; Calculating the index value for the frame to be calculated; The analysis device according to claim 1 or 2.

8. The image generating unit generates the plurality of ultrasound images. Information in which signal values ​​at each position included in a scannable area corresponding to the ultrasound image are collected for each of the scannable areas; The signal values ​​of each position included in the scannable area are collected for each scan line of the scannable area; or The signal value of each position included in the scannable area is acquired for each scan line after a push pulse or external excitation. The analysis device according to claim 1 or 2.

9. The ultrasound image is a B-mode image. The analysis device according to claim 1 or 2.

10. The image generating unit includes: performing a motion correction process for correcting motion between the ultrasound images; Calculating the index value using the plurality of ultrasound images after the motion correction process. The analysis device according to claim 1 or 2.

11. The motion compensation process includes a motion tracking process and a motion stabilization process. The analysis device according to claim 10.

12. Computer, an image acquisition unit that acquires a plurality of time-series ultrasound images; an image generating unit that acquires, at a plurality of time phases, index values ​​indicating a correlation between a signal value at a first position in the ultrasound image at one time phase and a signal value at a second position corresponding to the first position in the ultrasound image at another time phase, and generates an index image in which pixel values ​​according to a change in the correlation in a time axis direction are assigned to each position based on the index values ​​acquired at the plurality of time phases; A program that functions as a

13. Computer, an image acquisition unit that acquires a plurality of time-series ultrasound images; an image generating unit that acquires, at a plurality of time phases, index values ​​indicating a correlation of a change in signal intensity between a first position in the ultrasound image at one time phase and a second position corresponding to the first position in the ultrasound image at another time phase, and generates an index image that indicates a change in the correlation in a time axis direction based on the index values ​​acquired at the plurality of time phases; A program that functions as a

Images