Ultrasonic diagnostic apparatus and diagnostic assistance method

Abstract

The present application provides an ultrasonic diagnostic apparatus and a diagnostic assistance method. An image analysis section (28) has a reliability calculation section (29) that calculates a reliability that indicates a likelihood that a lesion candidate is a lesion. A marker display control section (30) displays a marker that notifies of the lesion candidate on an ultrasonic image. The marker is continuously displayed in a continuous detection state in which the reliability continuously satisfies a candidate detection condition. At this time, from a time point at which the candidate detection condition is first satisfied to a passage of a shape-invariant period, the shape of the marker is fixed, and thereafter, the shape of the marker is changed in correspondence with a change in the reliability.

Description

Technical Field

[0001] This disclosure relates to ultrasound diagnostic devices and diagnostic aids, and particularly to techniques for notifying examiners of lesion candidates. Background Technology

[0002] In ultrasound examination, a probe, placed against the surface of the patient, scans along that surface. During this scan, the examiner observes real-time tomographic images displayed on a monitor to determine the presence of lesions. If a lesion is found, it is examined in detail.

[0003] Visually identifying temporarily appearing lesions on dynamically changing tomographic images is not easy. As an aid in lesion identification, CADe (Computer Aided Detection) is employed. This technique, for example, detects potential lesions within a tomographic image and notifies the examiner of the detected lesion. For instance, it displays markers surrounding potential lesions on the tomographic image. CADe is used in conjunction with or incorporated into CAD (Computer Aided Diagnosis). CAD is also labeled CADx.

[0004] Document 1 (JP Patent No. 2004-159739) discloses an ultrasound diagnostic device with CAD functionality. However, Document 1 does not describe the display method for the change notification of lesion candidates. Document 2 (International Publication No. 2018 / 198327) describes a display method for the change notification of lesion candidates in an ultrasound diagnostic device with CAD functionality. However, this technique relies on the premise that the tendency to overlook lesion candidates varies from examiner to examiner, and aims to prevent the overlooking of lesion candidates on a per-examiner basis. That is, this technique is unrelated to the probability that the lesion candidate is indeed a lesion. Document 3 (JP Patent Application Publication No. 2012-249964) describes a method for displaying the diagnostic name in an ultrasound diagnostic device with CAD functionality when the reliability of lesion candidate detection exceeds a threshold.

[0005] In addition, in this application specification, "lesion" refers to a site that may be diseased or that requires detailed examination. "Supplementary lesion" refers to a site detected to assist the examiner in identifying and diagnosing the lesion.

[0006] When displaying a marker indicating a candidate lesion, it's considered to represent the reliability of lesion detection by changing the marker's shape. However, with this technique, the marker's shape is prone to change immediately after display begins. This is because the initial detection phase of a candidate lesion is unstable, and reliability is easily affected. Depending on the situation, changes in the marker's shape can hinder ultrasound image observation; specifically, it can be visually obstructive for the examiner. Summary of the Invention

[0007] The purpose of this disclosure is to limit undesirable changes in the morphology of markers while demonstrating the reliability of lesion candidate detection by notifying changes in the morphology of the markers.

[0008] The ultrasound diagnostic apparatus disclosed herein includes: a calculation unit that receives a frame data string obtained by repeatedly scanning an ultrasound beam, and calculates the reliability of the probability that a lesion candidate contained in the frame data is a lesion for each frame data; and a display control unit that displays a flag notifying the lesion candidate on an ultrasound image formed based on the frame data string, continuously displays the flag in a continuous detection state in which the detection of the lesion candidate continues, wherein in the continuous detection state, the display control unit fixes the shape of the flag from the time point when the lesion candidate is first detected until a morphological invariance period has elapsed, and changes the shape of the flag according to the reliability after the morphological invariance period has elapsed.

[0009] The diagnostic aid method disclosed herein includes the following steps: calculating the reliability of the probability that a lesion candidate contained in each frame of data constituting a frame data string obtained by repeatedly scanning an ultrasound beam; and displaying a flag notifying the lesion candidate on an ultrasound image formed based on the frame data string, continuously displaying the flag in a continuous detection state where the detection of the lesion candidate is continuous, fixing the shape of the flag from the time point when the lesion candidate is initially detected until a period of morphological invariance has elapsed while the flag is continuously displayed, and then changing the shape of the flag according to the reliability. Attached Figure Description

[0010] Figure 1 This is a block diagram illustrating the ultrasonic diagnostic apparatus involved in the implementation.

[0011] Figure 2 This is a diagram used to illustrate the method of generating the logo.

[0012] Figure 3 This is a flowchart representing an example of actions related to the display of a logo.

[0013] Figure 4 This is the first example of a diagram showing the control of the sign display.

[0014] Figure 5 This is a diagram used to illustrate Example 1 in more detail.

[0015] Figure 6 This is the second example of a diagram showing the control of the sign display.

[0016] Figure 7 This is a diagram showing a modified example of an ultrasonic diagnostic device.

[0017] Figure 8 This is the third example of a diagram representing the control of the sign display. Detailed Implementation

[0018] The following description of the implementation method is based on the accompanying drawings.

[0019] (1) Overview of the implementation method

[0020] The ultrasound diagnostic apparatus according to the embodiment includes a calculation unit and a display control unit. The calculation unit receives frame data strings obtained by repeatedly scanning an ultrasound beam, and calculates the reliability of the probability that a lesion candidate contained in that frame data is a lesion for each frame data. The display control unit displays a flag notifying a lesion candidate on the ultrasound image formed based on the frame data strings. The flag is continuously displayed during continuous detection, where the detection of lesion candidates continues. During continuous detection, the display control unit fixes the shape of the flag from the initial detection of a lesion candidate until a morphological invariance period has elapsed, and changes the shape of the flag according to the reliability after the morphological invariance period. The calculation unit is equivalent to a calculator. The display control unit is equivalent to a controller.

[0021] According to the above structure, when a lesion candidate is detected, the morphology of the marker is fixed until the morphological invariance period has passed. Since reliability is usually unstable in the initial stage of lesion candidate detection, reliability display is stopped during this period. This avoids problems caused by changes in the marker morphology during the initial detection phase, such as the marker morphology becoming obstructive to the examiner. Therefore, during the initial detection phase, attention can be focused on observing the area indicated by the marker. After the initial detection phase, i.e., after the morphological invariance period, even if the marker morphology changes, the change is usually slow, thus minimizing the aforementioned problems. Furthermore, typically, the initial detection phase involves relatively rapid probe movement, followed by detailed observation of the lesion candidate while slowly adjusting the probe's position and orientation, after the probe has been stopped.

[0022] In this implementation, the display control unit, while in continuous detection mode, changes the shape of the marker in stages as reliability increases after a period of unchanged shape. Although the shape of the marker could also change continuously, changing the shape in stages makes it easier to identify changes in reliability.

[0023] In this implementation, a non-display range and a display range are defined on the reliability axis. The display range is divided into multiple intervals, including the lowest-order interval. During the shape-invariant period, the flag's shape is the initial shape. After the shape-invariant period, the flag's shape is any one of multiple shapes corresponding to the multiple intervals. The multiple shapes include the initial shape corresponding to the lowest-order interval. If the flag shape displayed during the shape-invariant period is the same as the flag shape corresponding to the lowest-order interval, the frequency of shape changes at the elapsed time points during the shape-invariant period can be reduced.

[0024] In this implementation, the phased changes in the shape of the marker include at least one of the following: phased changes in the thickness of the marker, phased changes in the color of the marker, phased changes in the transparency of the marker, and phased changes in the type of lines constituting the marker. The marker may consist of four display elements defining the four corners of the area containing the candidate lesion site. The marker may be a colored area.

[0025] In this implementation, the display control unit displays a flag when the reliability becomes greater than a threshold, and removes the flag when the reliability becomes less than the threshold. If the reliability becomes less than the threshold before the morphological invariance period, the flag on the ultrasound image is removed. This structure informs the user of temporary detections while preventing the flag from obstructing the observation of the ultrasound image.

[0026] In this implementation, the arithmetic unit determines the attributes of candidate lesions contained in each frame of data. During continuous detection, the display control unit changes the shape of the flag according to the combination of reliability and attributes after a period of morphological invariance. Based on this structure, the reliability can be identified by the shape of the flag, and the attributes of candidate lesions can be identified. The concept of attributes includes, for example, disease name, severity, and malignancy.

[0027] In this implementation, the arithmetic unit determines either a significant attribute or a non-significant attribute for each frame of data, using these as candidate attributes for lesions included in that frame. If a significant attribute is determined, the display control unit, during continuous detection, changes the flag's shape according to reliability after a morphology-invariant period. Conversely, if a non-significant attribute is determined, the display control unit, during continuous detection, removes the flag after a morphology-invariant period. This structure prevents excessive notification of specific tissues that do not require detailed examination.

[0028] The diagnostic assistance method according to the embodiment includes a calculation step and a display step. In the calculation step, the reliability of the probability that a lesion candidate contained in each frame of data constituting a frame data string obtained by repeatedly scanning an ultrasound beam is a lesion is calculated. In the display step, a flag notifying a lesion candidate is displayed on the ultrasound image formed based on the frame data string. The flag is continuously displayed in a continuous detection state where the detection of lesion candidates continues. While the flag is continuously displayed, the shape of the flag is fixed from the time point when the lesion candidate is initially detected until a period of morphological invariance has elapsed, and then the shape of the flag is changed according to the reliability.

[0029] The program involved in the implementation method is a program for implementing the above-described computational assistance method in an information processing device. The information processing device has a non-transitory storage medium for storing the program. Examples of information processing devices include ultrasound diagnostic devices, ultrasound diagnostic systems, and computers.

[0030] (2) Detailed description of the implementation method

[0031] exist Figure 1 The structure of the ultrasound diagnostic apparatus according to the embodiment is shown in a block diagram. The ultrasound diagnostic apparatus is a medical device installed in medical institutions such as hospitals, and it forms an ultrasound image based on the received signal obtained by sending and receiving ultrasound waves to a living organism (the subject being examined). An organ that is the subject of ultrasound diagnosis is, for example, the breast.

[0032] In group breast examinations, it is necessary to identify lesions quickly and without omission. The ultrasound diagnostic device described in this embodiment has a CADe function that automatically detects potential lesion candidates (e.g., low-brightness regions that could be identified as tumors) within the ultrasound image to assist the examiner in identifying lesions. This will be described in detail later.

[0033] The probe 10 functions as a transceiver unit for ultrasound. Specifically, the probe 10 is a movable transceiver that is held and operated by the examiner (physician, examination technician, etc.). During breast ultrasound diagnosis, the transceiver surface (specifically, the acoustic lens surface) of the probe 10 is in contact with the surface of the patient's chest. While observing the real-time displayed tomographic images, the probe 10 is manually scanned along the chest surface. Once a potential lesion is identified on the tomographic image, the position and orientation of the probe 10 are slowly adjusted. Then, the position and orientation of the probe 10 are fixed, and the tomographic images are carefully observed.

[0034] In the illustrated structural example, probe 10 comprises an array of one-dimensionally arranged transducers. An ultrasonic beam (transmitting and receiving beams) 12 is formed by the transducer array, and a scanning surface 14 is formed by electronic scanning of the ultrasonic beam 12. The scanning surface 14 is the observation surface, i.e., the two-dimensional data acquisition area. Known electronic scanning methods for the ultrasonic beam 12 include electronic sector scanning and electronic linear scanning. Convex scanning of the ultrasonic beam 12 is also possible. Alternatively, a 2D transducer array can be provided within probe 10, allowing for the acquisition of volumetric data from within a biological organism through two-dimensional scanning of the ultrasonic beam.

[0035] Alternatively, a positioning system can be set up to obtain the position information of probe 10. The positioning system may consist of, for example, a magnetic sensor and a magnetic field generator. A magnetic sensor is mounted on probe 10 (more precisely, the probe head within probe 10). The magnetic sensor detects the magnetic field generated by the magnetic field generator. This allows the acquisition of three-dimensional coordinate information from the magnetic sensor. The position and orientation of probe 10 can be determined based on this three-dimensional coordinate information. Alternatively, motion information of probe 10 can be obtained based on information output from the positioning system, and this motion information can be used in the sign display control described later.

[0036] The transmitting circuit 22 functions as a transmitting beamformer. Specifically, during transmission, the transmitting circuit 22 supplies multiple transmitting signals in parallel to the vibrating element array, thereby forming a transmitting beam. During reception, if reflected waves from within the biological body reach the vibrating element array, multiple receiving signals are output in parallel from the multiple vibrating elements. The receiving circuit 24 functions as a receiving beamformer, generating beam data through phase-alignment and summing (also known as delay and summing) of the multiple received signals.

[0037] Each electronic scan generates multiple beam data arranged in the scanning direction, which constitute the received frame data corresponding to the scanning plane 14. Each beam data consists of multiple echo data arranged in the depth direction. A beam data processing unit is provided in the subsequent stage of the receiving circuit 24, but its illustration is omitted.

[0038] The image forming unit 26 is an electronic circuit that generates tomographic images (B-mode tomographic images) based on received frame data. It has a DSC (Digital Scan Converter). The DSC has coordinate transformation, pixel interpolation, and frame rate conversion functions, etc. More specifically, the image forming unit 26 generates a display frame data string based on the received frame data string. Each display frame data constituting the display frame data string is tomographic image data. A real-time dynamic image is composed of multiple tomographic image data. Ultrasound images other than tomographic images can also be generated. For example, color Doppler flow imaging images or three-dimensional images representing tissues can be formed. In the illustrated structural example, the display frame data string is sent to the display processing unit 32 and the image resolution unit 28.

[0039] The image analysis unit 28 is the module that performs the CADe function. The image analysis unit 28 performs lesion candidate detection processing on a per-frame basis, that is, on a per-tomographic image. Specifically, through binarization processing of the tomographic image, edge detection, etc., lesion candidates are detected as low-brightness closed regions. When a lesion candidate is detected, lesion candidate information is output from the image analysis unit 28.

[0040] The lesion candidate information includes the location and size information of the candidate lesion. It also includes a reliability score, which is a numerical value representing the probability that the candidate lesion actually is a lesion. Figure 1 The reliability calculation unit 29 is shown in the diagram. The candidate lesion information can include the attributes of the candidate lesion (disease name, condition, malignancy).

[0041] The location information of a lesion candidate can be, for example, the coordinates of its own center point, or the coordinates of the center point of a shape tangent to and enclosing the lesion candidate. The center point is a representative point. The geometric center point or the centroid of the shape can be used as the center point. The size information of a lesion candidate can be, for example, the size of the lesion candidate itself, or the size of a shape tangent to and enclosing the lesion candidate. For example, the size of the lesion candidate can be determined based on the coordinates of the center point and the coordinates of the top-left corner of the shape. Given the coordinates of the center point, the coordinates of the top-left corner can be considered as the size information of the lesion candidate. The area of ​​the lesion candidate can also be calculated using this size information. Multiple lesion candidates can also be detected concurrently.

[0042] The marker display control unit 30 displays a marker superimposed on the ultrasound image to indicate a detected lesion candidate. If the reliability is greater than a given threshold, the marker display control unit 30 considers a lesion candidate detected and generates a marker surrounding the lesion candidate. The marker display control unit 30 represents the level of reliability through phased changes in the marker's shape. However, a certain period from the start of detection (initial detection period) is defined as a shape-unchanged period, during which the marker's shape is fixed in its initial form. After the shape-unchanged period, the marker's shape is changed according to the reliability. Examples of changes in marker shape include changes in line thickness, line color, line transparency, line type, and marker shape.

[0043] The initial form is less noticeable than the subsequent emphasized form group; that is, it is not prominent. Conversely, the emphasized form group is more recognizable or visible than the initial form; that is, it is more noticeable.

[0044] Here are some specific examples of shape changes. For example, a sign can be displayed in cool colors when reliability is low, and in warm colors when reliability is high. Similarly, a sign can be displayed with low brightness when reliability is low, and with high brightness when reliability is high. A sign can also be displayed with high transparency when reliability is low, and with low transparency when reliability is high. Furthermore, a sign can be displayed with thin line widths when reliability is low, and with thick line widths when reliability is high. A sign can also be displayed with dashed lines when reliability is low, and with solid lines when reliability is high. The type of sign itself can also be switched. For example, the display of four corner elements can be switched from a rectangular graphic to a single element. Several techniques can also be used simultaneously.

[0045] In the early stages of detection, the detection of lesion candidates is unstable, and the reliability is also unstable. By limiting the changes in the morphology of the markers in the early stages of detection, the responsiveness of the marker display can be improved, and the markers can be prevented from becoming obtrusive due to excessive conspicuousness in the early stages of detection.

[0046] The image forming unit 26, the image resolving unit 28, and the sign display control unit 30 can each be configured by a processor. A single processor can function as the image forming unit 26, the image resolving unit 28, and the sign display control unit 30. Alternatively, a CPU (described later) can function as the image forming unit 26, the image resolving unit 28, and the sign display control unit 30.

[0047] The display processing unit 32 has functions such as graphic image generation, color calculation, and image synthesis. The display processing unit 32 receives outputs from the image forming unit 26 and the sign display control unit 30. A sign surrounding a lesion candidate is an element constituting a graphic image. In this embodiment, the sign display control unit 30 generates the sign, but the sign can also be generated by the main control unit 38, the display processing unit 32, etc.

[0048] The display 36 is composed of an LCD, an organic EL display device, or the like. The display 36 displays tomographic images as dynamic images in real time, and also displays symbols as part of graphic images. The display processing unit 32 is, for example, composed of a processor.

[0049] Main control unit 38 control Figure 1 The operation of each component shown is illustrated. In this embodiment, the main control unit 38 is composed of a CPU that executes the program. An operation panel 40 is connected to the main control unit 38. The operation panel 40 is an input device and includes multiple switches, multiple buttons, a trackball, a keyboard, etc.

[0050] The operation panel 40 can be used to set or change the flag display conditions. The flag display conditions include the morphology fixation period, multiple thresholds, etc. For example, the morphology fixation period can be adaptively determined based on the movement speed of the probe 10, the stability of the frame data, the reliability deviation, etc.

[0051] In one embodiment, a display frame data string is provided to the image parsing unit 28, but a receive frame data string may also be provided to the image parsing unit 28 (refer to reference numeral 42). In this case, another image forming unit is provided separately from the image forming unit 26 to perform image forming simply and quickly.

[0052] exist Figure 2 The diagram illustrates the marker generation method. A lesion candidate 46 is included in the tomographic image 44. A binarized image is generated by binarizing the tomographic image 44 47. The binarized lesion candidate 46A is extracted by edge detection or region detection of the binarized image. For example, a rectangle 52 tangent to the lesion candidate 46A is defined by the coordinates of its horizontal endpoints and vertical endpoints. Specifically, the coordinates of the center point 48 and the upper left corner point 50 of the rectangle 52 are determined.

[0053] A rectangle 54 is defined outside rectangle 52 to represent a shape with certain margins 56 and 58 in the horizontal and vertical directions. This rectangle 54 is displayed as marker 64 on the tomographic image 44. Marker 64 is a shape that surrounds the lesion candidate 46 and its surroundings. In the illustrated example, marker 64 is composed of dashed lines. A marker consisting of four elements representing only the four corner portions can be displayed. Circular or elliptical markers can also be displayed.

[0054] In this implementation, the detection of lesion candidates 46 is performed on a frame-by-frame basis. When a lesion candidate 46 is detected, a marker 64 is displayed on the tomographic image 44 corresponding to the frame data containing it. The display of the marker 64 allows the examiner to become aware of the presence of the lesion candidate 46, preventing it from being missed.

[0055] In continuous detection of lesion candidate 46, the shape of flag 64 is fixed from the start of lesion candidate detection until the morphology remains unchanged. After the morphology remains unchanged period, the shape of flag 64 is changed according to the reliability. That is, the reliability is represented by the shape of flag 64. When lesion candidate 46 becomes undetectable, flag 64 is removed. In fact, the reliability is compared with the first threshold described later. If the reliability is lower than the first threshold, it is considered that no lesion candidate has been detected; if the reliability is higher than the first threshold, it is considered that a lesion candidate has been detected. Alternatively, if the detection of lesion candidate 46 is interrupted for one or several frames, it can be considered that the detection of lesion candidate 46 is continuous.

[0056] exist Figure 3 Will Figure 1 The operation of the flag display control unit is shown in a flowchart. In S10, it is determined whether a lesion candidate has been detected. As described above, the presence or absence of a lesion candidate is checked for each frame of data, and the reliability is referenced for each frame of data. Additionally, in the illustrated example, in S10, it is also determined that the flag display is complete.

[0057] In S12, the start flag is displayed. The initial form is selected as the flag's form. In S14, it is determined whether a period of morphological invariance has elapsed since the initial detection of the lesion candidate. In S16, it is determined whether the detection of the lesion candidate is still continuous at the current time point. If the detection of the lesion candidate is continuous, i.e., if continuous detection is determined, S14 is repeated. During this process, if a period of morphological invariance has elapsed, S18 is executed. If it is determined in S16 that the detection of the lesion candidate is interrupted, i.e., if continuous detection is completed, the steps after S10 are executed, based on the flag being removed in S17. Additionally, in the illustrated example, in S16, it is also determined that the flag display is complete.

[0058] In S18, the flag shape is changed according to reliability. In the first example described later, the flag display range determined on the reliability axis is divided into three intervals, and three flag shapes are assigned to the three intervals. In S18, a specific shape is selected from the three shapes according to the reliability level, and the flag with that shape is displayed. Among the three shapes is an initial shape corresponding to the lowest partition. If the reliability belongs to the lowest partition immediately after the shape-invariant period, the flag shape is maintained as a result. This reduces the frequency of flag shape changes, thus balancing the provision of reliability information with the convenience of image observation.

[0059] In S20, similar to S16 above, it is determined whether the detection of the lesion candidate is continuous. If the detection of the lesion candidate is continuous, that is, in continuous detection state, S18 is repeated. If S20 determines that the detection of the lesion candidate is interrupted, that is, in continuous detection state, after the flag is cleared in S17, each step after S10 is performed. In addition, in the illustrated example, it is also determined in S20 that the flag indicates completion.

[0060] exist Figure 4 The first example of a sign display control is shown. Figure 4 The lower part shows the reliability axis, which characterizes the magnitude of the reliability R. The indicator display range 200 and the indicator non-display range 202 below it are defined on the reliability axis. The lower limit of the indicator display range 200 is defined by a first threshold th1. The indicator display range 200 is divided into a low interval 200A, a middle interval 200B, and a high interval 200C. The low interval 200A is the lowest interval. Furthermore, the reliability is a value between 0 and 100. The reliability at a certain time t is represented as Rt.

[0061] In the initial stage of testing (refer to reference numeral 72), a marker indicating the initial form (form A) is displayed. This initial stage corresponds to the form-invariant period, during which the marker indicating form A is displayed regardless of the reliability Rt. After the initial stage, i.e., after the form-invariant period (refer to reference numeral 74), as shown by reference numerals 76, 78, and 80, a form corresponding to the interval of reliability Rt is selected from forms A, B, and C, and a marker indicating that form is displayed.

[0062] Form A is the initial form described above. If inequality is used to represent the manifestation or salience (i.e., the degree of prominence) of each form, then the relationship A < B < C holds. Form A is selected when the reliability Rt is above the first threshold th1 but below the second threshold th2. Form B is selected when the reliability Rt is above the second threshold th2 but below the third threshold th3. Form C is selected when the reliability Rt is above the third threshold th3. th1 is, for example, 60, th2 is, for example, 75, and th3 is, for example, 90.

[0063] As described above, in the continuous detection state of lesion candidates, from the start of lesion candidate detection until a certain period has elapsed, the shape of the marker does not change even if the reliability changes. During this period, the reliability is mostly unstable, and if the shape of the marker changes according to the reliability, the marker can easily obstruct the observation of the ultrasound image. According to the embodiment, such a problem can be avoided. Furthermore, according to the embodiment, after a certain period, the change in reliability can be grasped or identified by the change in the shape of the marker. Thus, the lesion candidates indicated by the marker can be examined in detail while considering the reliability. After a certain period, the movement of the probe usually slows down, or the probe actually becomes stationary, so the shape of the marker does not change drastically.

[0064] use Figure 5 To explain in more detail Figure 4 The first example is shown. Reference numeral 82 indicates a graph representing the time variation of reliability. The horizontal axis is the time axis, and the vertical axis is the reliability axis. Three intervals are defined by setting a first threshold th1, a second threshold th2, and a third threshold th3 on the reliability axis. Reference numeral 84 indicates the markers displayed at each time point.

[0065] At time t1, the reliability Rt exceeds the first threshold th1, thus indicating the presence of the initial form A. At time t2, the reliability Rt falls below the first threshold th1. The period from time t1 to time t2 does not reach the form-invariant period 86.

[0066] At time t3, the reliability Rt exceeds the first threshold th1 again, and then also exceeds the second threshold th2. In this case, the flag with the initial form A is displayed. At time t4, after a form-invariant period 88, changes in the flag form are allowed from time t4 onwards. The form-invariant period 88 can be determined in seconds.

[0067] At time t4, reliability Rt belongs to the first interval, maintaining the initial state A. Then, at time t5, reliability Rt exceeds the second threshold, changing to a state B flag. Subsequently, at time t6, reliability Rt exceeds the third threshold, changing to a state C flag. Then, at time t7, reliability Rt returns to the first interval, changing to the initial state A flag. At time t8, reliability Rt falls below the first threshold, and the flag is removed.

[0068] Subsequently, at time t9, the reliability Rt exceeds the first threshold th1, actually reaching the third threshold th3, but at this time point, the flag is displayed with the initial form A. At time t10, the reliability Rt falls below the first threshold th1. The flag is displayed momentarily or only for a very short period. At time t11, the reliability Rt exceeds the first threshold again, and at time t12, after a period of form invariance, changes in the form of the flag are permitted.

[0069] The period of morphological invariance can be specified by the user, or it can be automatically set adaptively according to the situation. For example, the period of morphological invariance can be determined based on the stability of frame data, the rate of change of the overall reliability distribution of the frame, etc.

[0070] exist Figure 6 The second example of sign display control is shown. Among the possible sign shapes 92-98, shape 92 is the initial shape, and the other three shapes 94, 96, and 98 represent reliability levels, respectively. After a period where the shape remains unchanged, shape 94 is selected if the reliability falls within the first interval, shape 96 is selected if the reliability falls within the second interval, and shape 98 is selected if the reliability falls within the third interval. The initial shape 92 includes a rectangular sign body 100. Bar graphs are not included in area 102.

[0071] Shapes 94, 96, and 98 each have a rectangular sign body 100, and further include bar charts 102A, 102B, and 102C. The length (horizontal dimension) of the bars in bar charts 102A, 102B, and 102C represents the degree of reliability. According to Example 2, the reliability can be visually identified by observing the shape of the sign displayed after a period of shape invariance. When displaying bar charts 102A, 102B, and 102C, they can be displayed semi-transparently so that they do not completely obscure the ultrasonic image.

[0072] exist Figure 7 The middle shows Figure 1The illustrated example shows a modified version of the structure. In addition to the reliability calculation unit 29A, the image analysis unit 28A also includes an evaluation unit 104. In the illustrated example, the evaluation unit 104 has the function of identifying three categories of attributes that constitute a tumor. Specifically, it has the function of identifying benign tumor 1 (equivalent to a cyst), other benign tumor 2, and malignant tumors. Among these, for example, benign tumor 1 is designated as a non-critical attribute, while benign tumor 2 and malignant tumors are designated as critical attributes.

[0073] In the candidate detection state, the marker display control unit 30A selects an initial shape as the marker shape from the start of detection until the morphology remains unchanged. Then, for benign tumor 1, the marker is not displayed; for benign tumor 2 and malignant tumors, the marker shape is changed according to reliability. That is, for candidate lesions with non-significant attributes, the marker is displayed for a very short time to inform the examiner of its presence, and then removed for the sake of simplifying the display content. For candidate lesions with significant attributes, the marker is displayed according to… Figure 3 The process shown is used to perform the flag display control.

[0074] exist Figure 8 The third example shown is the sign display control. It uses... Figure 7 The structure shown is a prerequisite. Additionally, in Figure 8 In the middle, to and Figure 4 The same elements shown are labeled with the same reference numerals and their descriptions are omitted.

[0075] exist Figure 8 In the initial detection phase (during the morphology-unchanged period) 106, the initial morphology 120 is selected as the marker regardless of the attribute (category). After the initial detection period, the marker morphology is changed for benign tumors 116 and malignant tumors 118, but the marker is removed for benign tumors 114 (see attached reference numeral 134). Reference numeral 115 indicates an important attribute, corresponding to tumors 116 and malignant tumors 118. Benign tumor 114 corresponds to a non-important attribute. For the important attribute 115, after the initial detection period, the marker morphology is changed according to the reliability Rt.

[0076] Specifically, in the illustrated example, benign tumor 116 is shown with a blue marker, the shape (line type) of which changes according to the reliability Rt as indicated by reference numerals 122, 124, and 126. Malignant tumor 118 is shown with a red marker, the shape (line type) of which changes according to the reliability Rt as indicated by reference numerals 128, 130, and 132. As the reliability Rt increases, the marker shape is selected in stages to make it more prominent. Furthermore, in... Figure 8In the third example shown, the forms 122 and 128 corresponding to the lowest interval are not the same as the initial form 120. On the other hand, regardless of the differences in attributes, the initial form 120 is the same.

[0077] In this implementation, when the ultrasound image contains multiple candidate lesions, the aforementioned flag display control is performed for each candidate lesion. The image analysis unit can be configured as a machine learning-type analyzer, such as a CNN (Convolutional Neural Network).

Claims

1. An ultrasound diagnostic device, comprising: The arithmetic unit (29) receives frame data strings obtained by repeatedly scanning the ultrasound beam, and calculates the reliability of the probability that a lesion candidate contained in that frame data is a lesion for each frame data; and The display control unit (30) displays a flag indicating a candidate lesion on the ultrasound image formed based on the frame data string, and continuously displays the flag during continuous detection of the candidate lesion. In the continuous detection state, the display control unit (30) fixes the shape of the marker from the time point when the candidate lesion is first detected until the morphology remains unchanged period, and changes the shape of the marker according to the reliability after the morphology remains unchanged period.

2. The ultrasonic diagnostic device according to claim 1, wherein, In the continuous detection state, the display control unit (30) changes the shape of the flag in stages as the reliability increases after the period of unchanged shape.

3. The ultrasonic diagnostic device according to claim 2, wherein, Determine the non-display range and the display range on the reliability axis. The display range is divided into multiple intervals, including the lowest interval. During the period when the form remains unchanged, the form of the symbol is the initial form. After the period of morphological invariance, the shape of the marker is any one of the multiple shapes corresponding to the multiple intervals. The plurality of forms includes the initial form corresponding to the lowest interval.

4. The ultrasonic diagnostic device according to claim 2, wherein, The phased change of the shape of the logo includes at least one of the following: a phased change of the thickness of the logo, a phased change of the color of the logo, a phased change of the transparency of the logo, and a phased change of the type of lines constituting the logo.

5. The ultrasonic diagnostic device according to claim 1, wherein, The display control unit (30) displays the flag when the reliability becomes greater than the threshold, and removes the flag when the reliability becomes less than the threshold. If the reliability becomes less than the threshold before the period of morphological invariance has elapsed, the mark on the ultrasound image is removed.

6. The ultrasonic diagnostic device according to claim 1, wherein, The arithmetic unit (29) determines the attributes of the lesion candidates contained in each frame of data. In the continuous detection state, the display control unit (30) changes the shape of the flag according to the combination of the reliability and the attribute after the period of shape invariance.

7. The ultrasonic diagnostic device according to claim 1, wherein, The arithmetic unit (29) determines, for each frame of data, either an important attribute or a non-important attribute, as a candidate attribute for the lesion contained in that frame of data. When the display control unit (30) determines the important attribute, in the continuous detection state, after the period of unchanged shape, it changes the shape of the flag according to the reliability. When the display control unit (30) determines that the non-important attribute has been determined, it removes the flag during the continuous detection state after the period of unchanged shape.

8. A diagnostic aid method, comprising: The process involves calculating the reliability of the probability that a lesion candidate contained in each frame of a frame data string obtained by repeatedly scanning an ultrasonic beam; and... The process involves displaying a marker indicating a potential replacement lesion on an ultrasound image formed based on the frame data string. The marker is continuously displayed during continuous detection of the candidate lesion. While the marker is continuously displayed, the shape of the marker is fixed from the time point when the candidate lesion is first detected until the period of morphological invariance. After that, the shape of the marker is changed according to the reliability.

9. A program product comprising a program that executes in an information processing apparatus, the program comprising: The function calculates the reliability of the probability that a lesion candidate contained in a frame of data is a lesion, based on each frame of data in a frame data string obtained by repeatedly scanning an ultrasound beam; and The function is to display a flag indicating a potential lesion on the ultrasound image formed based on the frame data string. The marker is continuously displayed during continuous detection of the candidate lesion. While the marker is continuously displayed, the shape of the marker is fixed from the time point when the candidate lesion is first detected until the period of morphological invariance. After that, the shape of the marker is changed according to the reliability.

Images