JP7290766B2 - Information processing device and program - Google Patents

Description

The present invention relates to an information processing apparatus using a mammography image obtained by imaging a subject's breast using radiation such as X-rays, and a program.

2. Description of the Related Art Conventionally, in the medical field, mammography apparatuses that image a subject's breast using radiation such as X-rays have been widely used. When a mammography device is used to check for breast cancer or the like, images of both the left and right breasts of the subject are usually taken. Also, a mammography apparatus normally performs imaging by compressing the breast flat. This is for the purpose of fixing the breast, preventing overlapping of mammary gland tissue to image the entire breast, improving contrast, and reducing exposure dose. Conventionally, there is known a breast tomography apparatus (a so-called mammography CT (computed tomography) apparatus) that automatically stops imaging when there is an abnormality in the pressure and its distribution during compression (Patent Document 1). .

JP 2010-069241 A

If the breast is not properly compressed, the mammary gland tissue will overlap a lot, which may cause a problem that makes it impossible to make a proper diagnosis. In this case, a heavy burden is imposed on the subject, such as the necessity of re-imaging.

The only way to prevent such imaging failures is to compress the breast appropriately. It is not easy to find the appropriate compression conditions according to the characteristics and position the breast properly. Moreover, even when focusing on a specific subject, there are many cases where there is a difference in the above characteristics between the left and right breasts. Therefore, it is not easy to position the breast with appropriate compression conditions. Unless you are a skilled radiological technologist or doctor (hereinafter referred to as a radiological technologist, etc.), it is difficult. In particular, when radiologists, etc., who operate mammography apparatuses have low proficiency, it is difficult to find appropriate compression conditions according to the characteristics of the breast, and imaging failures tend to occur frequently due to breast positioning errors.

SUMMARY OF THE INVENTION An object of the present invention is to provide an information processing apparatus and a program that assist in reducing imaging failures caused by breast positioning failures.

Another information processing apparatus of the present invention is a compression thickness, which is a compression force and a thickness after compression on the right breast measured for imaging the right breast, which is the right breast of the subject, and a compression thickness of the left breast of the subject. A compression condition acquisition unit that acquires the compression force and the compression thickness, which is the thickness after compression, on the left breast measured for imaging of the left breast, the compression force and compression thickness of the right breast, and the compression force of the left breast and compression thickness, and a determination unit that determines whether positioning is possible or not, and a display unit that displays at least the determination result of the determination unit whether or not positioning is possible.

The display unit preferably displays the compression force and compression thickness for the right breast and the compression force and compression thickness for the left breast.

When the determination result of the determination unit indicates that positioning is impossible, the display unit preferably displays a message prompting re-imaging.

The determination unit preferably determines whether positioning is possible using a trained model that outputs whether positioning is possible based on the input of the right breast compression force and compression thickness and the left breast compression force and compression thickness.

The program of the present invention is the compression thickness, which is the measured compression force and the thickness after compression on the right breast, which is the right breast of the subject, and the left breast, which is the left breast of the subject. A compression condition acquisition unit that acquires the compression force and the compression thickness, which is the thickness after compression, on the left breast measured for imaging, a function that displays whether positioning is possible, the compression force and compression thickness of the right breast, and a function of judging whether or not positioning is possible using a learned model that outputs whether or not positioning is possible based on the input of the compression force and thickness of the left breast.

According to the information processing apparatus and program of the present invention, it is possible to reduce imaging failures due to breast positioning failures.

1 is an explanatory diagram showing the configuration of a mammography apparatus; FIG. 3 is a block diagram showing the configuration of a console; FIG. FIG. 4 is an explanatory diagram of a compression condition estimator; 4 is a flow chart showing the operation of the mammography apparatus; It is an example of display on the display unit. FIG. 3 is a block diagram of a learning device for learning a compression condition estimating unit; FIG. 7 is a block diagram showing the configuration of a console according to the second embodiment; FIG. FIG. 4 is an explanatory diagram of a determination unit; 9 is a flow chart showing the action of the mammography apparatus of the second embodiment; It is an example of display on the display unit. FIG. 3 is a block diagram of a learning device that learns a determination unit;

[First embodiment]
As shown in FIG. 1, a mammography apparatus 10, which is an example of a radiation imaging apparatus, includes an apparatus main body 11 for imaging the breast of a subject using X-rays, which are radiation, a console 12 for controlling the apparatus main body 11, Prepare.

The apparatus main body 11 includes a column 31, an X-ray generating section 32, an imaging table 33 containing an X-ray imaging section, a compression plate 36, an elevating section 37, and the like. The X-ray generating unit 32 and the imaging table 33 are integrated to constitute a movable unit 40 that adjusts the position of the apparatus main body 11 according to the subject.

The X-ray generator 32 includes at least an X-ray tube that generates X-rays. Therefore, in the mammography apparatus 10, the X-ray generator 32 is a radiation generator that generates radiation (X-rays). Also, the X-ray generator 32 can include a high voltage circuit that supplies a high voltage for generating X-rays to the X-ray tube. That is, the X-ray generator 32 can be configured by a so-called monotank.

The imaging table 33 is a stage on which the breast is placed, and a compression plate 36 is used to clamp the breast during imaging. Further, the imaging table 33 constitutes a radiographic imaging unit for imaging the subject's breast using radiation (X-rays). The imaging table 33 includes, as a radiation imaging unit, for example, an FPD (Flat Panel Detector) for imaging a breast using radiation, and a grid for removing scattered radiation (stationary Lisholmbrande or mobile bookie brandy) and The mammography apparatus 10 has a plurality of types of grids that can be exchanged according to imaging conditions, and can also perform imaging without using grids. Further, the imaging table 33 is provided with a grip portion 34a that is gripped by the right hand of the subject and a grip portion 34b that is gripped by the left hand of the subject. The gripping portion 34a and the gripping portion 34b are so-called armrests. Since X-rays are used in this embodiment, the radiation imaging unit is specifically an X-ray imaging unit.

The compression plate 36 compresses and flattens the breast of the subject placed on the imaging table 33 during imaging. This is to reduce overlapping of normal mammary glands and facilitate detection of lesion candidates such as calcification. A compression force sensor 38 is attached to the compression plate 36 to detect and output the force with which the compression plate 36 compresses the breast of the subject (hereinafter referred to as compression force). Therefore, when imaging is performed and a mammography image is obtained, the compression force sensor 38 can be used by the mammography apparatus 10 to obtain the compression force in the imaging. A compression thickness sensor 39 is attached to the compression plate 36 to detect and output the distance between the compression plate 36 and the imaging table 33, that is, the thickness of the compressed breast (hereinafter referred to as compression thickness). Therefore, when imaging is performed to obtain a mammography image, the compression thickness sensor 39 can be used to obtain the compression thickness in that imaging.

The elevating unit 37 elevates the compression plate 36 with respect to the imaging table 33 . Thereby, the lifting section 37 supports the compression plate 36 substantially parallel to the imaging table 33 at a specific distance according to the thickness of the breast.

The movable part 40 is rotatable within a predetermined angular range while maintaining the relative position and orientation of the X-ray generating part 32 and the imaging table 33 . Therefore, the device main body 11 can perform imaging with the imaging stand 33 arranged horizontally or with the imaging stand 33 inclined from the horizontal. Specifically, the apparatus main body 11 can perform CC imaging (craniocaudal imaging) in which the imaging table 33 is horizontally arranged and the breast is imaged from the craniocaudal direction. In addition, the apparatus main body 11 can perform MLO imaging (mediolateral oblique imaging) in which the imaging table 33 is arranged at an angle and the breast is imaged from the medial-lateral oblique direction.

Further, the X-ray generating section 32 of the movable section 40 is rotatable within a predetermined range while the positions of the imaging table 33 and the compression plate 36 are fixed. Thereby, the apparatus main body 11 can perform so-called stereo imaging and tomosynthesis imaging. In stereo imaging, the breast of a subject fixed in a specific position and orientation (for example, the position and orientation of CC imaging) is photographed from one or more oblique directions with different tilt angles, and a fluoroscopic image from the oblique direction ( hereinafter referred to as a stereo image). In addition, tomosynthesis imaging is a tomographic image (hereinafter referred to as a tomosynthesis image) of the subject's breast, which is fixed at a specific position and orientation, using images captured from a plurality of oblique directions. It is a shooting form that obtains Note that the compression force and compression thickness can be obtained regardless of the imaging mode, and can be obtained for each imaging.

The console 12 is a control device that controls the mammography apparatus 10 and an information processing device that performs information processing using various mammography images. As shown in FIG. 2 , the console 12 includes a compression condition acquisition section 51 , a compression condition estimation section 52 and a display section 53 .

The compression condition acquisition unit 51 acquires the compression force applied to the breast in mammography and the compression thickness, which is the thickness of the breast after compression. Therefore, the compression condition acquisition unit 51 includes a compression force measurement unit 61 and a compression thickness measurement unit 62 . The compression force measurement unit 61 can measure the compression force in real time at least when imaging is performed, using the signal output by the compression force sensor 38 . In addition, the compressive force measurement unit 61 can measure the compressive force ex post facto after imaging by using the record of the output of the compressive force sensor 38 that is stored in association with the mammography image. The compression thickness measurement unit 62 can measure the compression thickness in real time at least when imaging is performed. In addition, the compression thickness measuring unit 62 can measure the compression pressure ex post facto after imaging by using the record of the output of the compression thickness sensor 39 that is stored in association with the mammography image. In this embodiment, the compression force measuring section 61 continuously measures the compression force while the compression plate 36 is attached to the movable section 40 . Similarly, in this embodiment, the compression thickness measuring section 62 continuously measures the compression thickness while the compression plate 36 is attached to the movable section 40 . As a result, the compression condition acquisition unit 51 acquires the compression force and compression thickness imaging conditions in real time.

In the present embodiment, the compression condition acquiring unit 51 acquires the right breast that is the right breast of the subject (hereinafter simply referred to as the right breast) or the left breast that is the left breast of the subject (hereinafter simply the left breast). ), the compression force and compression thickness in imaging the first breast are acquired. That is, the compression condition acquiring unit 51 acquires the compression condition (compression force and compression thickness) of the previously imaged breast (first breast) when imaging the right breast and the left breast in the same imaging mode.

When the second breast, which is the other of the right breast and the left breast, is imaged in the same imaging mode as the first breast, the compression condition estimation unit 52 uses the compression force and compression thickness of the first breast to estimate the second breast. estimating a suitable compression force and compression thickness in the imaging of . For example, when performing MLO imaging of the right breast and then performing MLO imaging of the left breast, the compression condition estimating unit 52 uses the compression force and compression thickness in the MLO imaging of the right breast to perform MLO imaging of the left breast. Determine the compression force and compression thickness that is estimated to be suitable for The first breast may be either the right breast or the left breast, and the second breast may be either the left breast or the right breast, depending on whether the first breast is the right breast or the left breast.

For example, as shown in FIG. 3, the compression condition estimator 52 receives input of a first compression force 71 that is the compression force of the first breast and a first compression thickness 72 that is the compression thickness of the first breast. A second compression force 81 that is recommended (or allowed) when imaging the second breast in the same imaging mode as the first breast, and a recommended (or permissible) compression force when imaging the second breast in the same imaging mode as the first breast. Alternatively, a second compression thickness 82, which is a compression thickness that is acceptable), and a learned model (so-called AI (artificial intelligence) program) that outputs the compression thickness. In this embodiment, the first compression force 71 and the first compression thickness 72 are specific numerical values, whereas the second compression force 81 output by the compression condition estimation unit 52 is a numerical value indicating the range of compression force (for example, The second compression thickness 82 output by the compression condition estimator 52 is a numerical value (for example, the upper limit and the lower limit) that indicates the range of the compression thickness.

The display unit 53 displays the compression force and compression thickness estimated by the compression condition estimation unit 52 at least in the imaging of the second breast. That is, when imaging the second breast in the same imaging manner as the first breast, the display unit 53 displays the second compression force 81 and the second compression thickness 82 . The mammography apparatus 10 thereby assists in positioning the second breast. ``Display the estimated compression force and compression thickness at least in the imaging of the second breast'' means the timing at which the second breast can be positioned with reference to the estimated compression force and compression thickness (when imaging the second breast). It refers to presenting the compression force and compression thickness before or after the timing (particularly preferably, the timing of positioning the second breast). Positioning of the breast refers to keeping the part of the breast to be diagnosed (usually the entire breast) within the imaging range and adjusting the compression force and thickness of the breast to an appropriate range (obtaining a mammographic image that can be used for diagnosis). It means to make adjustments to fit within the range that can be

The display unit 53 can display the compression force measured by the compression force measurement unit 61 and/or the compression thickness measured by the compression thickness measurement unit 62 . If the compression force and the compression thickness are displayed in real time together with the second compression force 81 and the second compression thickness 82 that are the recommended values (recommended range), the radiological technologist or the like can set the compression force and compression thickness to the recommended values (recommended range). Since it becomes easy to match, the effect of assisting positioning in imaging of the second breast is high.

The operation of the mammography apparatus 10 configured as described above will be described below. As shown in FIG. 4, for example, when MLO imaging is performed on the right breast of a subject (step S101), the compression force measurement unit 61 measures the compression force in this imaging, and the compression thickness measurement unit 62 measures the compression force in this imaging. A compression thickness is measured (step S102). Thereby, the compression condition acquiring unit 51 acquires the compression force and compression thickness (first compression force 71 and first compression thickness 72) in MLO imaging of the right breast, which is the first breast.

When MLO imaging of the right breast is completed and then MLO imaging of the left breast of the same subject is performed, the compression condition estimation unit 52 uses the first compression force 71 and the first compression thickness 72 to perform MLO imaging of the left breast. The compression force and compression thickness (second compression force 81 and second compression thickness 82) recommended for MLO imaging are automatically estimated (step S103), and the display unit 53 displays them (step S104). For example, as shown in FIG. 5, the display unit 53 displays the actual compression force and compression thickness during imaging of the right breast and the recommended range of compression force and compression thickness during imaging of the left breast in the form of a table 90. . In FIG. 5, the actual compression thickness when imaging the right breast is "45 mm", and the actual compression force when imaging the right breast is "100 N (Newton)". Further, in FIG. 5, the recommended range of compression thickness for imaging the left breast is "40 to 50 mm" (40 mm or more and 50 mm or less), and the recommended range of compression force is "80 to 120 N" (80 N or more and 120 N or less). is.

As described above, after imaging the right breast, when imaging the left breast in the same imaging mode (MLO imaging in this embodiment) as the right breast, the mammography apparatus 10 sets the recommended compression force for imaging the left breast. and estimate and display the compression thickness. Since this display serves as a clear guideline for positioning the left breast, a radiologist or the like using the mammography apparatus 10 can position the left breast so that the compression force and compression thickness are within the above display. can be taken. That is, according to the mammography apparatus 10, positioning failure can be prevented, and as a result, imaging failures due to positioning failure can be reduced.

In this embodiment, MLO imaging of the right breast and left breast is performed. can be estimated and displayed. Also, in the present embodiment, imaging is performed from the right breast, but the mammography apparatus 10 can perform similar estimation and display when imaging the left breast first.

In the first embodiment, the compression condition estimating unit 52 estimates the compression force and compression thickness recommended for imaging the second breast in respective ranges. Compression thickness can be estimated with pinpoint values (one number each for compression force and compression thickness). When estimating the range of recommended compression force and compression thickness for imaging the second breast, radiological technologists can easily grasp the permissible compression force and compression thickness, which has the advantage of facilitating breast positioning. . On the other hand, when estimating the recommended compression force and compression thickness for imaging the second breast with pinpoint values, radiological technologists, etc. can easily and clearly determine the target compression force and compression thickness for imaging the second breast. In terms of being able to grasp, there is an advantage that positioning of the breast becomes easier.

In the first embodiment, the compression condition estimator 52 is the AI program itself, but the compression condition estimator 52 may include the AI program instead of the AI program itself. For example, the compression condition estimating unit 52 uses a learned model that outputs the compression force and compression thickness for imaging the second breast based on the input of the compression force and compression thickness for imaging the first breast. Compression force and compression thickness can be estimated at . For this reason, in the mammography apparatus 10, for example, the compression condition estimating unit 52 communicates with an information processing apparatus, an analysis apparatus, or the like on which the learned model that operates in a device or the like other than the console 12 is installed, and uses this to perform the second Breast compression force and compression thickness can be estimated. Since the AI program is still included, the case in which the compression condition estimation unit 52 is an AI program is one mode of the case in which the compression condition estimation unit 52 includes an AI program. The determination accuracy is particularly high when the compression condition estimator 52 is an AI program and when the compression condition estimator 52 includes an AI program.

The learning of the compression condition estimation unit 52, which is an AI program, can be performed using the learning device 101 shown in FIG. The learning device 101 is a device for learning (optimizing) the compression condition estimation unit 52 and includes a learning data acquisition unit 111 and an evaluation model update unit 112 .

The learning data acquisition unit 111 acquires data used for learning by the compression condition estimation unit 52 . Specifically, the learning data acquisition unit 111 obtains the compression force and compression thickness at the time of imaging mammography images that could be used for diagnosis, and the right breast and left breast of the same subject were imaged in the same imaging mode. Acquire the compression force and compression thickness at the time. Then, the compression force and compression thickness when one breast is imaged are used as explanatory variables, and the compression force and compression thickness when the other breast is imaged are input to the learning compression condition estimation unit 52 as correct labels.

The compression condition estimator 52 uses the explanatory variables, compression force and compression thickness, and an evaluation model (e.g., an evaluation function used in a specific regression analysis model) that obtains the objective variable from the explanatory variables, to obtain an estimated value of the objective variable (so-called evaluation value), and the error between the estimated value of the objective variable and the correct label. Objective variables in this learning are the range of compression force and compression thickness.

The evaluation model updating unit 112 optimizes the parameter values used by the compression condition estimation unit 52 in the evaluation model, using the error between the estimated value of the objective variable output by the compression condition estimation unit 52 during learning and the correct label. do. Optimization means obtaining parameters that minimize the error for a set of multiple explanatory variables and correct labels, and updating the parameters used in the evaluation model to those parameters. The compression condition estimating unit 52 repeats the above-described learning by the learning device 101 to obtain an optimized evaluation model (learned model).

Although the learning device 101 is a device that performs so-called supervised learning, the compression condition estimator 52 can be trained by unsupervised learning or reinforcement learning. Also, the evaluation model and optimization method may be changed according to the specific learning mode. Further, the compression condition estimator 52 may be optimized by supervised learning using explanatory variables, objective variables and/or evaluation models different from those of the learning device 101 .

[Second embodiment]
In the above-described first embodiment and modified example, imaging failures are reduced by displaying the recommended compression force and compression thickness. Alternatively, imaging failures can be reduced by displaying whether or not positioning is possible. .

In this case, for example, as shown in FIG. The determination unit 201 determines whether or not positioning is possible using the compression force and compression thickness measured for imaging the right breast and the compression force and compression thickness measured for imaging the left breast. Then, the display unit 53 displays at least whether the positioning is possible or not, which is the determination result of the determination unit 201 .

For example, as shown in FIG. 8, the determination unit 201 determines a first compression force 211 that is the compression force for the right breast, a first compression thickness 212 that is the compression thickness for the right breast, and a first compression force that is the compression force for the left breast. 2. Receiving input of compression force 221 and second compression thickness 222 which is the compression thickness of the left breast, determination of "positioning possible" or "positioning impossible" in imaging of the subject's breast (right breast and left breast) It can consist of a trained model that outputs results 230 . In this case, the determination unit 201 is a classifier that classifies the combination of the first compression force 211, the first compression thickness 212, the second compression force 221, and the second compression thickness 222 into positioning enabled or positioning disabled.

As described above, when the determination unit 201 is provided in the console 12, as shown in FIG. 9, for example, when MLO imaging is performed on the right breast of the subject (step S201), the compression force measurement unit 61 measures the compression force in this imaging. is measured, and the compression thickness measurement unit 62 measures the compression thickness in this imaging (step S202). As a result, the compression condition acquiring unit 51 acquires the compression force and compression thickness (first compression force 211 and first compression thickness 212) in imaging the right breast. After that, when the left breast of the subject is imaged by MLO (step S203), the compression force measurement unit 61 measures the compression force in this imaging, and the compression thickness measurement unit 62 measures the compression thickness in this imaging (step S203). S204). As a result, the compression condition acquiring unit 51 adds the previously acquired compression force and compression thickness for imaging the right breast to the compression force and compression thickness for imaging the left breast (the second compression force 221 and the second compression thickness 222). ).

When the compression condition acquisition unit 51 acquires the compression force and compression thickness for the left and right breasts as described above, the determination unit 201 uses these to determine whether positioning is possible (step S205), and the display unit 53 displays "Positioning possible". Alternatively, the judgment result of "Positioning impossible" is displayed (step S206). For example, as shown in FIG. 10, the display unit 53 displays the actual compression force and compression thickness during imaging of the right breast, the actual compression force and compression thickness during imaging of the left breast, and the determination result of positioning availability. , are displayed in the form of Table 235. In FIG. 10 , the actual compression thickness when imaging the right breast is “45 mm”, the actual compression force when imaging the right breast is “100 N”, and the actual compression thickness when imaging the left breast is “43 mm”. and the actual compression force "98 N" when imaging the left breast. Also, in FIG. 10, the determination result of the determination unit 201 is that positioning is possible.

As described above, the mammography apparatus 10 determines whether or not positioning is possible and displays the result, thereby preventing the end of imaging without failure, and as a result, the failure of imaging can be reduced. For example, if the determination result indicates that positioning is not possible, a radiologist or the like can see the display of the determination result and immediately redo imaging of the left breast and/or the right breast on the spot. As a result, substantially only mammography images for which positioning is possible are used for diagnosis. Therefore, at the stage of using mammography images for diagnosis, imaging failures caused by positioning of the breast can be reduced.

As shown in the second embodiment, the display unit 53 preferably displays the compression force and compression thickness of the right breast and the compression force and compression thickness of the left breast. This is because, when the determination result indicates that positioning is not possible, it is easy to grasp the cause. Specifically, the display of the compression force and compression thickness during imaging of the left and right breasts indicates whether the radiological technologist should redo the imaging of the right breast, redo the imaging of the left breast, or Since it is a material for determining whether to retake both breast imaging, if these are displayed, reimaging can be performed smoothly while suppressing the radiation dose even if positioning is impossible.

In the second embodiment described above, when the determination result of the determination unit 201 indicates that positioning is not possible, the display unit 53 displays a display prompting re-shooting (for example, a message or a warning color display). preferably. This is to more directly prompt re-shooting when positioning is impossible, and to reliably prevent shooting failure.

In the second embodiment described above, the determination unit 201 is the AI program itself, but the determination unit 201 may be configured to include the AI program instead of the AI program itself. For example, the determining unit 201 can determine whether or not positioning is possible using a trained model that outputs whether or not positioning is possible based on the input of the right breast compression force and compression thickness and the left breast compression force and compression thickness. . Therefore, in the mammography apparatus 10, for example, the determination unit 201 communicates with the learned model operating in a device other than the console 12, and can use this to determine whether positioning is possible. Note that the AI program is still included, so the case where the determination unit 201 is the AI program is one aspect of the case where the determination unit 201 includes the AI program. The determination accuracy is particularly high when the determination unit 201 is an AI program and when the determination unit 201 includes an AI program.

The learning of the determination unit 201, which is an AI program, can be performed using a learning device 250 shown in FIG. The learning device 250 is a device for learning (optimizing) the determination unit 201 and includes a learning data acquisition unit 251 and an evaluation model update unit 252 .

The learning data acquisition unit 251 acquires data used for learning by the determination unit 201 . Specifically, the learning data acquisition unit 251 determines the compression force and compression thickness when the right breast and left breast of the same subject are imaged in the same imaging mode, and whether positioning is possible (i.e., whether mammography images can be used for diagnosis). (whether or not), and Then, the compression force and compression thickness when imaging the left and right breasts are used as explanatory variables, and whether or not positioning is possible is input as a correct label to the determination unit 201 for learning.

The determination unit 201 uses an evaluation model (e.g., an evaluation function used in a specific regression analysis model) that obtains the objective variable from the explanatory variables, compression force and compression thickness, and an evaluation value for estimating the objective variable, Output the evaluation value error (allowable error range). In this learning, the objective variable is the classification of positioning possible or positioning impossible.

The evaluation model updating unit 252 optimizes the values of the parameters used by the determining unit 201 in the evaluation model, using the evaluation values output by the determining unit 201 during learning, their errors, and correct labels. Therefore, the determination unit 201 becomes an optimized evaluation model (learned model) by repeating the learning by the learning device 250 .

Although the learning device 250 is a device that performs so-called supervised learning, the determination unit 201 can learn by unsupervised learning or reinforcement learning. Also, the evaluation model and optimization method may be changed according to the specific learning mode. Also, the determination unit 201 may be optimized by supervised learning using explanatory variables, objective variables, and/or evaluation models different from those of the learning device 250 .

In addition, although the compression force is used in each of the above-described embodiments and modifications, data of the pressure (compression pressure) with which the compression plate 36 compresses the breast is used instead of or in conjunction with the compression force. can be used, estimated, and/or displayed.

In addition, the compression condition estimating unit 52 calculates the size, shape, and/or hardness (elasticity) of the breast, etc. instead of or in addition to the compression force and/or compression thickness. With such information input, recommended compression conditions can be output. Similarly, the determination unit 201 may determine breast-related parameters such as breast size, shape, and/or hardness (elasticity) instead of, or in addition to, the compression force and/or compression thickness. Based on the input of information, it is possible to determine whether or not positioning is possible. In this way, the accuracy of estimation and determination can be improved by considering information related to the breast.

In addition, the compression condition estimator 52 obtains input of biological information in place of or in addition to the breast compression force and/or compression thickness, and outputs recommended compression conditions. can do. Biological information includes, for example, the subject's height, weight, age, and whether or not there is a disease or treatment that causes changes in breast characteristics. The biometric information can also include information about the race, or the region (country, prefecture, state, region, etc.) in which the subject resides. Since breast characteristics may differ depending on race and the like, estimation accuracy is improved by taking biometric information into consideration as described above. Similarly, the determination unit 201 can determine whether or not positioning is possible by obtaining the biometric information input instead of or in addition to the breast compression force and/or compression thickness. . By taking biometric information into consideration in this way, the determination accuracy of the determining unit 201 is improved.

In each of the above-described embodiments and modifications, etc., the console 12 functions as an information processing device. It may have a function as a device. Further, the function of the console 12 as an information processing device in each of the above-described embodiments and modifications may be provided by the device main body 11 . When the function of the console 12 as an information processing apparatus in each of the above-described embodiments and modifications is provided in the console 12 or the apparatus main body 11, the entire mammography apparatus 10 functions as an information processing apparatus. In addition, the functions of the console 12 as an information processing device in each of the above-described embodiments and modifications can be distributed and implemented in a plurality of devices (a plurality of computers, etc.).

In the above-described embodiments, etc., the compression condition acquisition unit 51, the compression force measurement unit 61, the compression thickness measurement unit 62, the compression condition estimation unit 52, the learning data acquisition unit 111, the evaluation model update unit 112, the determination unit 201, and the learning data acquisition unit 251 and the evaluation model updating unit 252, the hardware structure of the processing unit that executes various processes is various processors as shown below. Various processors include CPU (Central Processing Unit), GPU (Graphical Processing Unit), FPGA (Field Programmable Gate Array), etc., which are general-purpose processors that run software (programs) and function as various processing units. Programmable Logic Devices (PLDs), which are processors whose circuit configuration can be changed after manufacturing, and dedicated electric circuits, which are processors with circuit configurations specially designed to perform various types of processing, etc. .

One processing unit may be composed of one of these various processors, or a combination of two or more processors of the same or different type (for example, a plurality of FPGAs, a combination of CPU and FPGA, or a combination of CPU and A combination of GPUs, etc.). Also, a plurality of processing units may be configured by one processor. As an example of configuring a plurality of processing units in one processor, first, as represented by computers such as clients and servers, one processor is configured by combining one or more CPUs and software, There is a form in which this processor functions as a plurality of processing units. Secondly, as typified by System On Chip (SoC), etc., there is a form of using a processor that realizes the functions of the entire system including multiple processing units with a single IC (Integrated Circuit) chip. be. In this way, the various processing units are configured using one or more of the above various processors as a hardware structure.

Further, the hardware structure of these various processors is, more specifically, an electrical circuit in the form of a combination of circuit elements such as semiconductor elements.

REFERENCE SIGNS LIST 10 Mammography Apparatus 11 Apparatus Main Body 12 Console 31 Post 32 X-ray Generator 33 Imaging Table 34a Grasping Part 34b Grasping Part 36 Compression Plate 37 Lifting Part 38 Compression Force Sensor 39 Compression Thickness Sensor 40 Movable Section 51 Compression Condition Acquisition Section 52 Compression Condition Estimation Section 53 Display Section 61 Compression Force Measurement Section 62 Compression Thickness Measurement Section 71 First Compression Force 72 First Compression Thickness 81 Second Compression Force 82 Second Compression Thickness 90 Table 101 Learning Device 111 Learning Data Acquisition Section 112 Evaluation Model Update Section 201 Determination unit 211 First compression force 212 First compression thickness 221 Second compression force 222 Second compression thickness 230 Determination result 235 Table 250 Learning device 251 Learning data acquisition unit 252 Evaluation model update unit S101 to S206 Operation steps

Claims (5)

Compressive force on the right breast and compression thickness, which is the thickness after compression, measured for imaging of the right breast, which is the right breast of the subject, and measurement of imaging of the left breast, which is the left breast of the subject. a compression condition acquiring unit that acquires a compression force and a compression thickness, which is the thickness after compression, on the left breast;
a determination unit that determines whether or not positioning is possible using the compression force and compression thickness of the right breast and the compression force and compression thickness of the left breast;
a display unit that displays at least whether positioning is possible or not, which is the determination result of the determination unit;
Information processing device. The information processing apparatus according to claim 1, wherein the display unit displays the compression force and compression thickness of the right breast and the compression force and compression thickness of the left breast. 3. The information processing apparatus according to claim 1, wherein when the determination result of the determination unit indicates that the positioning is impossible, the display unit displays a display prompting re-imaging. 3. The determining unit determines whether the positioning is possible or not by using a learned model that outputs whether or not the positioning is possible based on the inputs of the right breast compression force and compression thickness and the left breast compression force and compression thickness. 4. The information processing device according to any one of 1 to 3. Compressive force on the right breast and compression thickness, which is the thickness after compression, measured for imaging of the right breast, which is the right breast of the subject, and measurement of imaging of the left breast, which is the left breast of the subject. a function of driving an information processing device having a compression condition acquisition unit that acquires a compression condition acquisition unit that acquires a compression force and a compression thickness that is a thickness after compression to the left breast, and a display unit that displays whether positioning is possible;
A function to determine whether the positioning is possible using a trained model that outputs whether the positioning is possible or not based on the input of the right breast compression force and compression thickness and the left breast compression force and compression thickness, and a function to determine whether the positioning is possible or not. program to implement.

Images