JP7331749B2 - Medical image generation device and medical image generation program - Google Patents

Description

The present disclosure relates to a medical image generation apparatus and a medical image generation program for generating a schematic front view of the breast that a medical professional refers to when performing combined ultrasound examination for breast cancer examination.

In the diagnosis of breast cancer tumors, clues such as intraductal calcification, ductal dilatation, cysts and masses have been conventionally based on breast X-ray examination (mammography) in screening tests to search for tumors suspected of being cancerous. Screening tests have been performed to detect structures in the subject's tissue that will become

In recent years, mammography has made MLO (Medio-Lateral Oblique)
and CC (Cranio-Caudal (Cranio-Caudal direction)) of the breast is acquired and interpreted by a specialist, and includes tissue structures in the subject's tissue that may be breast cancer tumors extracted based on mammography images A combined ultrasound examination, in which an ultrasound examination is further performed on the region of interest and comprehensive judgment is made on the attribute of the tissue structure based on both images, is the mainstream. At that time, since the doctor who interprets the mammography and the medical staff who performs the ultrasound examination (hereinafter referred to as the "ultrasound operator") are usually different, the region of interest in the mammography image is transmitted to the ultrasound operator. We need a means to do so.

Under these circumstances, a medical information system has been proposed that roughly displays the position of the region of interest in the mammography image in a schematic diagram of the breast front (schema) for ultrasonography, based on the position of the region of interest in the mammography image (MLO and CC). (For example, Patent Documents 1, 2, and 3). As a result, the ultrasound operator can roughly grasp the position of the region of interest in the mammography image from the schema image for ultrasound examination and perform ultrasound scanning.

JP 2015-104464 A JP 2015-104465 A JP 2015-27450 A

However, according to the conventional medical information system described above, it is difficult for the ultrasound operator to grasp the position of the region of interest in the depth direction of the breast from the schema image. Therefore, the ultrasound operator needs to read out the mammography images (MLO and CC) at the time of ultrasound examination and directly confirm the position of the region of interest in the depth direction from the mammography images, which is complicated and time-consuming. However, the work burden on the operator increased, and it was difficult to improve the efficiency of the inspection.

The present disclosure has been made in view of the above problems, and is for medical image generation that enables an ultrasound operator to grasp the position of a region of interest in a mammography image in the depth direction of the breast from a schema image in combined ultrasound examination. An object of the present invention is to provide an apparatus and a medical image generation program.

A medical image generation device according to an aspect of the present disclosure is a medical image generation device that generates a schema image based on a mammography image and information on a region of interest, wherein an image for acquiring the information on the mammography image and the region of interest, and Front information representing a range of the attention area in the front direction is calculated for the attention area in the breast image included in the mammography image, based on the information acquisition unit and the acquired mammography image and information of the attention area. Depth information representing the position of the attention area in the depth direction for the attention area in the breast image included in the mammography image is calculated based on the front information calculation unit and the acquired mammography image and information on the attention area. a depth information calculation unit for generating the schema image; a schema image generation unit for generating a schema image by reflecting the front information and the depth information; and a display control unit for displaying the schema image on a display device. The unit generates a schema image in which the aspect of the identification unit reflecting the front information in the schema image is changed according to the content of the depth information.

According to the medical image generation apparatus and the medical image generation program according to one aspect of the present disclosure, the ultrasound operator can grasp the position of the region of interest in the depth direction of the breast from the schema image during ultrasound examination. can. Therefore, it is possible to eliminate the work of the ultrasound operator to read out the mammography image and confirm the position of the region of interest in the depth direction during the ultrasound examination, thereby reducing the work burden on the ultrasound operator and improving the examination efficiency. can.

1 is a system configuration diagram of a medical image system including a medical image archiving communication system 10 according to an embodiment; FIG. 3 is a functional block diagram showing a functional configuration of a diagnostic support device (CAD) 3; FIG. 2 is a functional block diagram showing the functional configuration of a server device 4 that constitutes the medical image generating apparatus; FIG. 4 is a functional block diagram showing a functional configuration of a control section 41 of the server device 4; FIG. It is an example of a process chart showing the flow of combined examination in breast cancer examination. (a) to (c) are examples of process diagrams showing the details of the process of image interpretation and scanning position confirmation in combined examination. 4 is a flowchart of medical image generation processing in the medical image generation apparatus; FIG. 8 is a flow chart showing details of calculation processing of the front information of the attention area in step S7 of FIG. 7. FIG. FIG. 5 is a schematic diagram for explaining a method of calculating front information representing the range of an attention area in a schema image in the front direction from information about an attention area in an MLO image and an attention area in a CC image; FIG. 8 is a flowchart showing details of a process of calculating depth information regarding depth in the depth direction of the attention area in step S8 of FIG. 7; FIG. 4(a) to 4(d) are schematic diagrams for explaining a method of calculating depth information regarding the depth in the depth direction of a region of interest from information of the region of interest in an MLO image or a CC image. FIG. FIG. 8 is a flow chart showing the details of the depth information calculation process regarding the depth direction thickness of the attention area in step S9 of FIG. 7 ; FIG. 8A to 8C are schematic diagrams for explaining a method of calculating depth information regarding the thickness of a region of interest in the depth direction from information of the region of interest in an MLO image or a CC image; FIG. (a) to (c) are examples of mammography images in which the depth of the region of interest differs in the depth direction. 14A to 14C are examples of schema images reflecting depth information in which depths in the depth direction of the attention areas corresponding to FIGS. 14A to 14C are different. 14A to 14C are other examples of schema images reflecting depth information in which depths in the depth direction of the regions of interest corresponding to FIGS. 14A to 14C are different. The upper diagrams of (a) to (c) are examples of mammography images (MLO, CC) in which the attention area RI has different depths in the depth direction. The lower diagram is an example showing another aspect of a schema image reflecting depth information with different depths in the depth direction of the attention area RI. The upper diagram shows mammography images (MLO, CC) with different depths of the attention area RI, and the lower diagrams show another mode of schema images reflecting depth information with different depths of the attention area RI. An example. (a) to (c) are examples of mammography images in which the thickness of the region of interest in the depth direction is different. 19A to 19C are another examples of schema images reflecting depth information in which depths and thicknesses in the depth direction of the regions of interest corresponding to FIGS. 19A to 19C are different. FIG. 4 is a diagram showing classification of methods for specifying a region of interest in an MLO image (left side in the column) and a CC image (right side in the column). It is an example of a schema image in which the method of reflecting depth information is changed according to the classification of the designation method of the attention area in FIG. 21 . 10 is a flowchart of medical image generation processing in a medical image generation apparatus according to a modified example; The upper diagram of (a) is a mammography image (MLO, CC) in the medical image generating apparatus according to Embodiment 2, the lower diagram is an example of a schema image, and the upper diagram of (b) is an image generated by an operator. The same mammography image (MLO, CC) in the medical image generating apparatus of Embodiment 2 after correction, and the lower figure are an example showing the aspect of the schema image as well. The upper diagram is a mammography image (MLO, CC) in the medical image generating apparatus of Embodiment 2 after correction by the operator, and the lower diagram is another example of a schema image. FIG. 11 is a functional block diagram showing the functional configuration of a control unit 41A of a server device in a medical image generating device according to Embodiment 3; (a) is a schematic diagram for explaining a method for identifying a reference candidate range in a breast image included in a CC image based on a region of interest in a breast image included in the MLO image; (b) is included in the MLO image; FIG. 4 is a schematic diagram showing a display mode of a region of interest RI in a breast image and a reference candidate range in a breast image included in a CC image; (a) to (c) are schematic diagrams showing an example of specifying a reference candidate range in a breast image included in a CC image for regions of interest existing at different positions in a breast image included in an MLO image. 8(a) and 8(b) are schematic diagrams showing an example of specifying reference candidate ranges in a breast image included in a CC image with respect to regions of interest existing at different positions in a breast image included in an MLO image. FIG. (a) is a schematic diagram for explaining a method of specifying a reference candidate range in an MLO image based on information of the attention area in the attention area in the CC image, and (b) is the attention area in the breast image included in the CC image. FIG. 4 is a schematic diagram showing a display mode of a reference candidate range in a breast image included in an RI and an MLO image; (a) to (c) are schematic diagrams showing an example of specifying a reference candidate range in a breast image included in an MLO image for regions of interest existing at different positions in a breast image included in a CC image. 4(a) and 4(b) are schematic diagrams showing an example of specifying a reference candidate range in a breast image included in an MLO image for regions of interest existing at different positions in a breast image included in a CC image. FIG. (a) is an example of a CC image in which a region of interest is displayed on a breast image, and (b) is an example of an MLO image in which a reference candidate range is displayed on a breast image.

<<Embodiment>>
<Overall Configuration of Medical Imaging System 100>
The medical imaging system 100 according to Embodiment 1 will be described in detail below with reference to the drawings. FIG. 1 is a system configuration diagram of a medical imaging system 100 according to an embodiment. As shown in FIG. 1, the medical imaging system 100 is used for diagnosing breast cancer tumors, and includes a mammography device 1, an ultrasonic diagnostic device 2, a diagnostic support device 3 (CAD: Computer-Aided Diagnosis), a server device 4, and medical images. It has a database 5 and a display device 6 (display). Furthermore, a gateway terminal (not shown) may be provided as a dedicated terminal for executing the position estimation function. These devices are interconnected via a network NT. The DICOM (Digital Imaging and Communication in Medicine) standard is applied to the network NT.

In the present embodiment, the medical imaging system 100 configured as described above captures an image of a subject region using the mammography apparatus 1 and stores and manages the obtained mammography images in the medical image database 5 . At the same time, receiving an operation input related to the position designation of the region of interest on the mammography image based on the designation of the interpreting doctor, the diagnosis support device 3 generates information about the candidate of the region of interest, and based on the information about the region of interest or the candidate, The position of the target area in the front direction and the depth direction of the subject is estimated, and a schema image reflecting them is generated. Then, the schema image is displayed on the display device 6 in order to support the ultrasonic examination in the ultrasonic diagnostic apparatus 2 .

Here, the "position of the attention area" is the position of an arbitrary reference point selected from the attention area. Alternatively, a plurality of points within the attention area may be used as reference points, and the position of each point may be used as the position of the attention area. Alternatively, when the attention area is a range of size equal to or larger than a predetermined value, a plurality of points that can cover the attention area may be set as reference points, and the position of each point may be the position of the attention area. It is possible to specify the position of the attention area in consideration of the size, and in subsequent processing, it is possible to calculate the position of the attention area in consideration of the size of the attention area.

<Configuration of Each Part of Medical Imaging System 100>
Each component of the medical imaging system 100 will be described below.

The mammography apparatus 1 scans the entire breast with a plate-like body from two predetermined directions (MLO: Medio-Lateral Oblique (medio-lateral oblique direction) and CC (Cranio-Caudal direction)) on the breast of the subject. In the sandwiched and compressed state, X-rays are irradiated along the MLO and CC, the X-rays transmitted through the breast are detected, and a mammography image (hereinafter referred to as "MLO image") of the entire breast taken from the MLO and the breast This is an X-ray imaging apparatus that generates a mammography image (hereinafter referred to as a "CC image") captured entirely from a CC.The mammography apparatus 1 is an MLO image and a CC image that are generated in conformity with the DICOM standard. Subject information, examination information, etc. are appended and output to the medical image database 5 for storage.

The ultrasonic diagnostic apparatus 2 transmits ultrasonic waves into the subject from an ultrasonic probe equipped with a piezoelectric element array, and detects reflected waves of ultrasonic waves caused by differences in acoustic impedance in the tissue of the subject. This device generates and displays an ultrasonic tomographic image showing the shape of the internal tissue of the subject based on the electrical signals received by the transducer and obtained. The ultrasonic diagnostic apparatus 2 appends subject information, examination information, clinical information, etc. regarding the image to the ultrasonic tomographic image generated in conformity with the DICOM standard, and outputs the image to the medical image database 5 for storage.

The diagnosis support apparatus 3 performs image analysis of the MLO image and the CC image supplied from the mammography apparatus 1, and performs detection processing of a region of interest candidate including a tissue structure in the subject suspected of being a breast cancer tumor. A candidate for the region of interest is a region that becomes a candidate when an image interpreting doctor extracts a region of interest from a mammography image. be.

Here, the "tissue structure" is a characteristic structure of the internal tissue of the subject. Refers to the breast milk ducts, etc. Also, the term “region of interest” refers to an image region showing tissue structures included in the MLO image and the CC image representing the entire breast of the subject. As mentioned above, the presence or absence of these structures is a major clue in screening tests for suspected breast cancer tumors. Therefore, screening tests for diagnosing breast cancer tumors involve detection of regions of interest contained in MLO and CC images.

FIG. 2 is a functional block diagram showing the functional configuration of the diagnostic support device 3. As shown in FIG. The diagnosis support device 3 has a CPU (Central Processing Unit) 31, a memory 32, a display device 33 (display), and an input/output unit 34 (interface), as shown in FIG. For example, an interface compatible with 10GBASE-T can be used for the input/output unit 34 to acquire images and information from the medical image database 5 . The CPU 31 stores part or all of the MLO image and the CC image in the memory 32, executes programs such as filtering processing and template matching, and detects tissue structures in the subject that are candidates for breast cancer tumors. For detection, for example, known detection algorithms described in JP-A-2000-126163, JP-A-2005-095501, etc. can be used. The CPU 31 can cause the display device 33 to superimpose the information about the detected attention area candidates on the reference source MLO image and CC image.

Information about the detected attention area candidate is added to the reference source MLO image and CC image and output to the input/output unit 34. Information about the attention area candidate is output from the input/output unit 34 to the medical image database 5 and stored. be done. Further, a configuration may be adopted in which related information detected from the MLO image and the CC image is also added. The information added to each of the MLO image and the CC image includes the position of the region of interest, the nipple position, the position of the breast base line, the position of the breast skin line, the upper and lower ends of the breast skin line (the left end and the right end in CC), the nipple Examples include the position of the axis.

The server device 4 (medical image generation device), the medical image database 5, and the display device 6 constitute a medical image archiving and communication system 10 (PACS: Picture Archiving and Communication System, hereinafter referred to as "PACS"). The PACS 10 stores and manages the medical image data acquired by the mammography apparatus 1 and the ultrasonic diagnostic apparatus 2 and the diagnosis support information from the diagnosis support apparatus 3 in the medical image database 5 . The medical images stored in the medical image database 5 are searchable based on subject information, examination information, clinical information, etc., and the selected medical image can be displayed on the display device 6 . In addition, a schema image is generated based on an instruction by an operation input from an interpreting doctor, and the schema image and the interpretation report created by the interpreting doctor are stored in the medical image database 5 in association with the reference source MLO images and CC images. to manage.

The medical image database 5 is a computer-readable recording medium such as a flexible disk, hard disk, MO, DVD, DVD-RAM, BD, SSD, semiconductor memory, or the like. MLO images and CC images output from the mammography apparatus 1, and ultrasonic tomographic images, schema images, etc. output from the ultrasonic diagnostic apparatus 2 are saved.

The display device 6 is a display device for so-called image display, and based on the image output from the display control unit 43, which will be described later, an MLO image, a CC image, information regarding designation of a region of interest by a doctor, an ultrasonic tomographic image, a schema, and so on. A display image including images and other clinical information is displayed on the screen. A liquid crystal display, a CRT, an organic EL display, or the like can be used as the display device 6 . The display device 6 has, for example, DVI, DisplayPort, etc. as an interface for receiving images from the display control unit 43 of the server device 4 .

The server device 4 is a main device that constitutes the main part of the medical image archiving communication system 10. In the present embodiment, the server device 4 is an MLO image, a CC image, and a region of interest related to an operation input from a radiologist, or a diagnosis support device. 3 functions as a medical image generating apparatus that generates a schematic image by receiving information on the candidate of the attention area from 3 as an input. FIG. 3 is a functional block diagram showing the functional configuration of the server device 4 that constitutes the medical image generating device. The server device 4 has a control section 41, a memory 42, a display control section 43, an operation input section 44, and an input/output section 45 (interface), as shown in FIG.

The control unit 41 is implemented by a programmable device such as a CPU, a GPGPU (General-Purpose computing on Graphics Processing Unit), a processor, and software. Alternatively, the configuration may be realized by a hardware circuit such as FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit). Details of the control unit 41 will be described later.

The memory 42 may be a semiconductor memory configured within an FPGA, ASIC, CPU, or GPGPU. Alternatively, a storage device connected from the outside may be used, and a part of the medical image database 5 may be used.

The display control unit 43 is a circuit that creates a display screen including a schema image, outputs it to the display device 6, and causes the display device 6 to display the display screen.

The input/output unit 45 is an interface for inputting/outputting information to/from the medical image database 5. For example, an interface card compatible with 10GBASE-T can be used. Like the control unit 41 , the display control unit 43 can be configured within an FPGA, ASIC, CPU, or GPGPU, and may be configured integrally with the control unit 41 .

These components can be single circuit components or aggregates of multiple circuit components. Also, a plurality of components can be combined to form a single circuit component, or a plurality of circuit components can be aggregated.

The operation input unit 44 receives and outputs various operation inputs such as various settings and operations from the operator to the control unit 41 . In the present embodiment, an interpreting doctor receives an operation input related to designation of a region of interest in a mammography image, and outputs information about the designated region of interest to the control unit 41 . The operation input unit 44 may be, for example, a keyboard having keys for various operations, an operation panel, a mouse, etc. having buttons and levers for various operations. Alternatively, the operation input unit 44 may be a touch panel integrated with the display device 6, for example. In this case, the interpreting doctor is configured to be able to specify a region of interest by performing a touch operation or a drag operation on the mammography image displayed on the display device 6 .

The server device 4 generates a schema image by the control unit 41 executing a medical image generation processing program that generates a schema image by inputting information about the MLO image, the CC image, and the region of interest or the candidate of the region of interest. It functions as a medical image generation device. FIG. 4 is a functional block diagram showing a functional configuration related to schema image generation of the control unit 41. As shown in FIG. The control unit 41 has an image and information acquisition unit 411, a front information calculation unit 412, a depth information calculation unit 413, and a schema image generation unit 414, as shown in FIG.

Specifically, the image and information acquisition unit 411 acquires information on the region of interest from the medical image database 5 . The front information calculation unit 412 stores part or all of the MLO image and the CC image in the memory 42, and based on the acquired MLO image and CC image and the information on the region of interest, the breast image included in each image. Front information representing the range of the attention area in the front direction is calculated. The depth information calculation unit 413 stores part or all of the MLO image and the CC image in the memory 42, and based on the acquired MLO image and CC image and the information on the region of interest, the depth information in the breast image included in each image. Depth information representing the range of the attention area in the depth direction is calculated. A schema image generation unit 414 generates a schema image by reflecting the front information and the depth information. In this case, a schema image may be generated in which the aspect of the identification portion of the front information in the schema image is changed according to the content of the depth information.

Details of processing in each component will be described later.

Here, "contents of depth information" refers to both the type of parameter and the value of the parameter when the depth information consists of a plurality of types of parameters. In the present embodiment, for example, the position (depth) of the attention area in the depth direction and the thickness of the attention area in the depth direction are shown as the types of parameters. As the parameter values, the value of the position (depth) of the attention area in the depth direction and the value of the thickness of the attention area in the depth direction are shown.

The control unit 41 outputs the generated schema image to the input/output unit 45 in association with the reference source MLO image and CC image, and the schema image is output from the input/output unit 45 to the medical image database 5 and stored. The control unit 41 outputs the schema image and the reference source MLO image and CC image to the display control unit 43, and the display control unit 43 creates a display screen including the schema image and outputs the display screen to the display device 6. 6 is displayed. It may be displayed on the display device 6 together with the MLO image and/or the CC image.

A gateway (not shown) is a dedicated terminal for executing the position estimation function, acquires lesion position information from the diagnosis support device 3 (CAD), and retrieves the corresponding inspection image from the medical image archiving and communication system 10 (PACS). It acquires and estimates the position and transmits the result to the medical image archiving communication system 10 .

<Flow of breast cancer examination>
First, an outline of the flow of examination for breast cancer will be described. In breast cancer examinations, combined examinations are performed in which tissue structures within a subject suspected of having breast cancer tumors in the breast are screened using a mammography apparatus and an ultrasonic diagnostic apparatus. FIG. 5 is an example of a process chart showing the flow of combined examination in breast cancer examination.

First, X-rays are irradiated along the MLO and CC in a state in which the entire breast of the subject is sandwiched between the MLO and CC and compressed by plate-shaped bodies, and an MLO image and a CC image obtained by photographing the entire breast from the MLO are obtained. get.

Next, the MLO image and the CC image are interpreted to detect tissue structures, and the work of confirming the position to be scanned by the ultrasonic probe in ultrasonic diagnosis is performed. FIGS. 6A to 6C are examples of process diagrams showing aspects of the process of image interpretation and scanning position confirmation.

In the example shown in FIG. 6A, first, the interpretation of the MLO image and the CC image is performed by the interpretation doctor. The radiologist examines each of the MLO image and the CC image for the presence of tissue structures in the subject suspected of having a breast cancer tumor, and inputs the region in which the tissue structure exists as a region of interest to the investigation input unit 44. input to the server device 4 via Next, the server device 4 generates a schema image representing the position to be scanned by the ultrasonic probe corresponding to the position of the subject in the attention area based on the information about the attention area.

The example shown in FIG. 6B is a case where the diagnosis support device 3 supports interpretation. In this case, prior to image interpretation by an image interpreting doctor, the image analysis of the MLO image and the CC image is performed by the diagnosis support device 3 based on the operation of the examination technician, and the target image including the tissue structure in the subject suspected of being a breast cancer tumor is detected. Detecting processing for area candidates is performed. When interpreting images, an interpreting doctor refers to information about candidates for regions of interest, interprets MLO images and CC images, and specifies candidate regions.

The example shown in FIG. 6(c) is a case where image interpretation and confirmation of the position to be scanned by the ultrasonic probe are performed in parallel. In this case, as in the case of FIG. 6(b), the diagnosis support device 3 detects a candidate region of interest based on the MLO image and the CC image. Based on the information, a schema image representing the position to be scanned by the ultrasonic probe corresponding to the position of the subject in the region of interest is generated.

Next, returning to FIG. 5 , an ultrasonic examination is performed by the ultrasonic diagnostic apparatus 2 . At this time, the doctor or laboratory technician who operates the ultrasonic diagnostic apparatus 2 applies the ultrasonic probe to the scanning position confirmed in the above process, scans it, and receives the reflected wave. Accordingly, by scanning the vicinity of the region corresponding to the region of interest in the subject's breast, it is possible to receive reflected waves from the region with high sensitivity and obtain an ultrasonic tomographic image with high resolution and high S/N.

Finally, comprehensive determination is performed based on the obtained MLO image, CC image, and ultrasonic tomographic image to determine the presence or absence of a tissue structure in the subject to be subjected to a detailed examination such as a biopsy.

<Operation of server device 4 (medical image generating device)>
Medical image generation processing and display operation in the medical image generation apparatus 4 configured as above will be described.

FIG. 7 is a flowchart of medical image generation processing in the server device 4 (medical image generation device).

First, the image and information acquisition unit 411 receives various operation inputs such as subject information, examination information, clinical information, etc. from the operator via the operation input unit 44 (step S1).

When the information is input, the image and information acquisition unit 411 selects mammography images (MLO images and CC images) based on the input subject information, examination information, clinical information, etc., and reads them out from the medical image database 5. It is displayed on the display device 6 (step S2). Further, it is determined whether or not the information about the candidate of the attention area is added to the mammography image (step S3), and if it is added, the information about the candidate of the attention area is added to the reference source MLO image and the CC image. The annotation is superimposed and displayed on the display device 6 (step S4), and if not added, the process proceeds to step S4.

Next, the image and information acquisition unit 411 receives an operation input for specifying a region of interest from the operator, who is an image interpreting doctor, via the operation input unit 44 (step S5). When the attention area is specified, it is checked whether or not to end the specification input of the attention area (step S6). goes to step S7.

Next, the front information calculation unit 412 calculates the front information of the attention area (step S7). FIG. 8 is a flowchart showing the details of the process of calculating the front information of the attention area in step S7. FIG. 9 is a schematic diagram for explaining a method of calculating the front information representing the range RI3 of the attention area in the schema image Im3 in the front direction from the information of the attention area RI1 in the MLO image Im1 and the attention area RI2 in the CC image Im2. be. In FIG. 9, the horizontal and vertical directions of the schema image Im3 in the front direction are the X direction and the Y direction, the horizontal and vertical directions of the MLO image Im1 are the X direction and the Z direction, and the horizontal and vertical directions of the CC image Im2 are the X direction. Let the direction be the Z direction. Further, as a method for calculating the front information, for example, a known method described in JP-A-2015-104464, JP-A-2015-104465, etc. can be used.

First, the front information calculation unit 412 performs image analysis on mammography images (MLO, CC) to detect skin lines, nipple positions, and breast base lines (step S71).

Specifically, as shown in FIG. 9, the skin line Sl1, the nipple position Ni1, and the breast base line Cw1 are detected from the MLO image Im1, and the skin line Sl2, the nipple position Ni2, and the breast base line Cw2 are detected from the CC image Im2. Here, in this specification, the nipple positions in the MLO image Im1 and the CC image Im2 are indicated by attaching numbers (1, 2) to the nipple position Ni, the skin line Sl, and the breast base line Sw, respectively. The numbers are omitted when the two images are not distinguished from each other.

Here, the "breast base line" is a line indicating the outer edge of the base portion of the breast in the breast image in the mammography image. , or "a straight line parallel to the Y-axis passing through the end point in the positive direction of the X-axis in the mammography image" can be used.

As a method for extracting the skin line and the chest wall line or the straight line connecting the upper end and the lower end of the breast image, a known technique can be used. A method can be used in which the luminance value of the skin line, which is the boundary between the . Alternatively, for example, a Sobel filter is used to acquire contour data corresponding to a change in brightness between two pixels, the contour data is binarized, and the binarized contour data is subjected to convex hull processing to obtain skin lines and A chest wall line or a straight line connecting the upper end and the lower end of the breast image is extracted. More specifically, a Sobel filter is used to extract regions where the luminance value (X-ray intensity) changes, and binarization is used to extract only regions where the degree of change in luminance value is equal to or greater than a threshold value, and contour extraction is performed. Alternatively, as disclosed in Japanese Patent Application Laid-Open No. 2011-125363, after the tomographic image is wavelet-transformed, only high-frequency components are inversely transformed, and a search is performed based on pixel values to extract skin lines. Further, as a method for calculating the nipple position, for example, a known method described in Japanese Patent Application Laid-Open No. 2015-104464, Japanese Patent Application Laid-Open No. 2015-104465, etc. can be used.

Next, the front information calculation unit 412 tilts the mammography image (MLO, CC) with respect to the schema image at the imaging angle and arranges the nipple axis (step S72).

Specifically, as shown in FIG. 9, a nipple axis Nl1 passing through the nipple position Ni1 of the MLO image Im1 and perpendicular to the breast baseline Cw1 passes through the nipple position Ni3 of the schema image Im3, and The nipple axis Nl1 is arranged so as to be inclined by the photographing angle θ (the vertical direction is 0°) of the MLO image Im1 with respect to the Y direction of the schema image Im3.

Further, a nipple axis Nl2 passing through the nipple position Ni2 of the CC image Im2 and perpendicular to the breast base line Cw2 passes through the nipple position Ni3 of the schema image Im3, and the nipple axis Nl2 is the Y direction of the schema image Im3. , the CC image Im2 is arranged so that the photographing angle is 0°.

In addition, in FIG. 9, the breast baselines Cw1 and 2 in the MLO image Im1 and CC image Im2 are illustrated as the X direction of the MLO image Im1 and CC image Im2, respectively.

Next, the front information calculation unit 412 enlarges or reduces the mammography image (MLO, CC) to match the outer edge of the schema image with the end of the mammography image (MLO, CC) (step S73).

Specifically, as shown in FIG. 9, the CC image Im2 is scaled based on the nipple position Ni2 so that the X-direction end point Xmax2 of the CC image Im2 coincides with the X-direction end point Xmax3 of the schema image Im3. . At the same time, the CC image Im2 may be scaled so that the X-direction end point Xmin2 of the CC image Im2 coincides with the X-direction end point Xmin3 of the schema image Im3.

In addition, the MLO image Im1 is positioned at the nipple position Ni1 so that the X-direction end point Xmax1 of the MLO image Im1 is in contact with the schema image Im3 and a straight line Nl1′ passing through the X-direction end point Xmax3 of the schema image Im3 and parallel to the nipple axis Nl1 is in contact with the schema image Im3. Scale to standard.

Next, the front information calculation unit 412 projects the attention area of each mammography image (MLO, CC) onto the schema image, calculates the overlapping area as the range of the attention area (front information) in the schema image (step S74), and calculates the front information. The information is output to the schema image generation unit 414 (step S75).

Specifically, the region of interest RI1 of the MLO image Im1 is projected onto the schema image Im3, the region of interest RI2 of the CC image Im2 is projected onto the schema image Im3, and the overlapping region is the range of the region of interest in the schema image Im3 in the front direction. It is calculated and output as front information representing RI3.

Next, the depth information calculation unit 413 calculates depth information regarding the depth of the attention area (step S8). FIG. 10 is a flowchart showing the details of the depth information calculation process regarding the depth direction depth of the attention area in step S8. FIG. 11 shows a method of calculating depth information about the depth in the depth direction of a region of interest from information of the region of interest in an MLO image or a CC image, wherein (a) is the first method and (b) is the second method. and (c) are schematic diagrams for explaining the third method.

First, the depth information calculation unit 413 calculates the total breast thickness D1 in the depth direction in the mammography image (MLO, CC) (step S81). In the first method, as shown in FIG. 11A, in the mammography image (MLO, CC), the distance from the nipple position Ni to the breast base line Cw along the depth direction axis (Z direction) in the mammography image is Let D1 be the total breast thickness in the depth direction.

Next, the depth information calculator 413 calculates the distance (region of interest depth) D2 from the nipple position or skin line to the reference position of the region of interest in the mammography image (MLO, CC) (step S82). In the first method, as shown in FIG. 11A, a straight line passing through the reference point Pt in the region of interest RI and parallel to the depth direction axis (Z direction) in the mammography image and a straight line passing through the nipple position Ni along the depth direction axis Let the distance from the intersection with the vertical straight line L1 to the reference point Pt be the depth of interest D2.

Here, in this example, as shown in FIG. 11A, the shallowest point in the depth axis direction in the attention area RI is set as the reference point Pt. However, the reference point Pt may be an arbitrary point existing within the attention area RI. For example, the central point of the attention area RI or the deepest point in the depth axis direction may be selected as the reference point Pt. Alternatively, a point group consisting of a plurality of points may be used as the reference point Pt. For example, four vertices may be selected as reference points Pt for a rectangular attention area RI, and the attention area depth D2 may be calculated for each of them. When the attention area RI is in a range of a size equal to or greater than a predetermined value, a plurality of points capable of covering the attention area RI are selected as the reference points Pt, thereby determining the position of the attention area RI considering the size of the attention area RI. calculation becomes possible.

Next, the depth information calculation unit 413 calculates the ratio D2/D1 (relative depth of attention area) of the depth of attention area D2 to the total breast thickness D1 (step S83).

Next, the depth information calculator 413 calculates a representative value of the relative depths of the regions of interest in the mammography image from the relative depths of the regions of interest in the MLO image and the CC image (step S84), and converts the front information to the schema image generation unit 414. Then, the relative depth of the region of interest is output as depth information to the schema image generation unit 414 (step S85), and the process returns to FIG. When the relative depths of the regions of interest of the MLO image and the CC image are different, the calculation of the representative value may be performed by using a method such as averaging, selecting either one, or selecting the maximum value.

When adopting the second method, in step S81, as shown in FIG. 11B, in the mammography image (MLO, CC), the depth direction axis in the mammography image passes through the reference point Pt in the region of interest RI. The distance from the intersection of the straight line L2 parallel to (Z direction) and the breast skin line S1 to the intersection of the straight line L2 and the breast base line is defined as the total breast thickness D1 in the depth direction, and in step S82, the straight line L2 and the breast skin The distance from the intersection with the line Sl to the reference point Pt may be set as the attention area depth D2. In the second method, as shown in FIG. 11B, the shallowest point in the depth axis direction in the attention area RI is set as the reference point Pt. However, the reference point Pt may be any point existing within the attention area RI.

Further, when adopting the third method, in step S81, as shown in FIG. The distance from the intersection of the straight line L3 parallel to the nipple axis and the breast skin line S1 to the intersection of the straight line L3 and the breast base line Cw is defined as the total breast thickness D1 in the depth direction. As shown in (c), the distance from the intersection of the straight line L3 and the breast skin line Sl to the reference point Pt may be the depth of interest D2. With such a configuration, it is possible to calculate the position of the region of interest in the depth direction even when the depth direction in the shama image and the direction of the breast base line do not match. In the third method, as shown in FIG. 11(c), the shallowest point among the intersections of the straight line parallel to the nipple axis in the attention area RI and the attention area RI is set as the reference point Pt. However, the reference point Pt may be any point existing within the attention area RI.

Alternatively, when adopting the third method, as shown in FIG. 11(d), when a straight line L4 passing through the upper end and the lower end of the nipple Ni is taken, the nipple axis passes through the nipple position Ni and is perpendicular to the straight line L4. With the straight line Nil, the total breast thickness D1 and the region of interest depth D2 may be calculated using the same method as in FIG. 11(c).

Next, the depth information calculation unit 413 calculates depth information regarding the thickness of the attention area (step S9). FIG. 12 is a flowchart showing the details of the depth information calculation process regarding the thickness of the target area in the depth direction in step S9. FIG. 13 shows a method of calculating depth information regarding the depth direction thickness of a region of interest from information of the region of interest in an MLO image or a CC image, wherein (a) is the first method and (b) is the second method. It is a schematic diagram for description.

First, the depth information calculator 413 calculates the total breast thickness D3 in the depth direction in the mammography image (MLO, CC) (step S91). In the first method, as shown in FIG. 13A, from the intersection of the breast skin line Sl and a straight line L2 passing through the reference point Pt in the region of interest RI and parallel to the depth direction axis (Z direction) in the mammography image, The distance to the intersection of the straight line L2 and the breast base line Cw is defined as the total breast thickness D3 in the depth direction. In the first method, as shown in FIG. 13A, the shallowest point in the depth axis direction in the attention area RI is set as the reference point Pt. However, the reference point Pt may be any point existing within the attention area RI.

Next, the depth information calculator 413 calculates the region-of-interest thickness D4 in the mammography image (MLO, CC) (step S92). In the first method, as shown in FIG. 11A, the distance between the two intersections of the straight line L2 and the outer edge of the attention area is set as the attention area thickness D4.

Next, the depth information calculation unit 413 calculates the ratio D4/D3 (relative thickness of the region of interest) of the thickness of the region of interest D4 to the total breast thickness D3 (step S93).

Next, the depth information calculation unit 413 calculates a representative value of the relative thickness of the region of interest in the mammography image from the relative thickness of the region of interest in the MLO image and the CC image (step S94), and generates a schema image using the relative thickness of the region of interest as depth information. Output to the unit 414 (step S95) and return to FIG. The method of calculating the representative value is the same as in step S84.

In the above process, when the second method is adopted, in step S91, a straight line L3 that passes through the reference point Pt in the region of interest RI, passes through the nipple position Ni, and is parallel to the normal to the breast base line Cw. The distance from the intersection with the breast skin line Sl to the intersection of the straight line L3 and the breast base line Cw may be the total breast thickness D3 in the depth direction. The distance between the intersection points may be used as the attention area thickness D4. With this configuration, it is possible to calculate the thickness of the region of interest in the depth direction even when the depth direction in the shama image does not match the direction of the breast base line. In the second method, as shown in FIG. 13(b), the shallowest point among intersections of straight lines parallel to the nipple axis in the attention area RI and the attention area RI is set as the reference point Pt. However, the reference point Pt may be any point existing within the attention area RI.

Alternatively, when adopting the second method, as shown in FIG. 13C, when a straight line L4 passing through the upper end and lower end of the nipple Ni is taken, the nipple axis is a straight line passing through the nipple position Ni and perpendicular to the straight line L4. Nil, the total breast thickness D3 and the region-of-interest thickness D4 may be calculated using the same method as in FIG. 13(b).

Next, the schema image generation unit 414 reflects the front information and the depth information to generate a schema image in which the aspect of the identification part of the front information in the schema image is changed according to the content of the depth information (step S10). ), the display control unit 43 creates a display screen including a schema image, outputs it to the display device 6, and displays it on the display device 6 together with the MLO image and/or the CC image (step S11).

Next, the image and information acquisition unit 411 confirms whether or not to end the specification input of a further attention area (step S12). , the process is terminated when it is terminated.

<Aspect of schema image>
An example of a schema image generated and displayed by the server device 4 (medical image generation device) will be described with reference to the drawings.

(Reflection of the depth in the depth direction of the attention area)
FIGS. 14(a) to 14(c) are examples of mammography images with different depths of attention regions in the depth direction. FIGS. 15(a) to (c) and FIGS. 16(a) to (c) are examples of schema images reflecting depth information with different depths in the depth direction of the attention area. FIGS. 15A to 15C are examples using a bar-shaped indicator bar At1 representing the position of the region of interest in the front direction, and the intersection of the indicator bar At1 and the diameter perpendicular to the schema image is the mammography image ( MLO, CC) corresponds to the position of the attention area RI in the X direction. Also, FIGS. 16A to 16C are examples in which a fan-shaped indicator At2 is used to represent the position of the region of interest in the front direction. MLO, CC) is the existence range of the attention area RI.

The schema image generator 414 further classifies the depth into "shallow, medium, deep" as shown in FIGS. As shown in (a) to (c), one or more selected from the color, density, brightness, transparency, and texture of indicators At1 and At2 representing the position of the attention area in the front direction corresponding to the classification depth. By using different elements, schema images reflecting depth information with different depths in the depth direction may be generated and displayed.

This allows the ultrasound operator to grasp the position of the region of interest in the depth direction of the breast from the schema image during ultrasound examination. It is common for an ultrasound operator to make a diagnosis while alternately looking at a target region of a subject and a display screen. At this time, according to the medical image generating apparatus according to the present embodiment, the form of the schema image itself is changed to transmit information about the position of the region of interest in the depth direction of the breast. The range of movement is suppressed within the range between the target site and the display screen. In addition, the ultrasound operator can visually recognize the position of the region of interest in the depth direction of the breast in a state that is closer to the operator's sense from the change in the mode of the indicator portion of the schema image itself.

Therefore, it is possible to eliminate the work of the ultrasound operator to read out the mammography image and confirm the position of the region of interest in the depth direction during the ultrasound examination, thereby reducing the work burden on the ultrasound operator and improving the examination efficiency. can.

(Reflection of Position Estimation Error Based on Depth Direction of Attention Area)
As the distance from the papilla to the attention area RI increases, the error in the position estimation result increases, and the reliability of the indicator position indicating the position of the attention area in the front direction decreases. On the other hand, in the schema image, a range in which the attention area RI may exist is displayed as an indicator range according to the depth information (distance) of the attention area RI, and a display function including the position estimation error is provided. can be done. The size of the indicator is the range in which the region of interest RI may exist in the mammography image (MLO, CC).

The upper diagrams of FIGS. 17A to 17C are examples of mammography images (MLO, CC) having different depths in the depth direction of the region of interest RI. The lower diagrams of FIGS. 17A to 17C are examples showing other forms of schema images reflecting depth information with different depths in the depth direction of the attention area RI.

As shown in the upper diagrams of FIGS. 17A to 17C, the depth of the attention area RI is classified into "shallow, medium, and deep", and as shown in the lower diagrams, , and by varying the size of the indicator At6 representing the position of the attention area in the front direction, a schema image reflecting the possible existence range of the attention area RI is generated. For example, in FIG. 17A, the depth of the attention area RI is shallow and the error in the position estimation result is small, so the size of the indicator At6 representing the position of the attention area in the front direction is displayed small. On the other hand, in FIG. 17(c), the depth of the attention area RI is large, and the position estimation result has a large error, so the size of the indicator At6 is displayed large.

Also, the upper diagram in FIG. 18 shows mammography images (MLO, CC) with different depths of the attention area RI, and the lower diagram shows different schema images reflecting depth information with different depths of the attention area RI. It is another example showing a mode. In FIG. 18 as well, corresponding to the depth of the attention area RI, even if the indicator At7 representing the position of the attention area RI is not in the vicinity of the center of the breast in plan view, the size of the indicator At7 is varied so that the attention area A schema image is generated that reflects the range in which RI may exist.

Here, the form of the identifying portion reflecting the front information in the schema image may be changed according to the size of the range in which the attention area RI may exist in the front direction corresponding to the attention area RI depth. At this time, the relationship between the depth of the attention area RI and the size of the indicator At6 is determined based on the relationship between the depth and the position estimation result error. Specifically, the deeper the depth of the attention area RI, the larger the size of the range in which the attention area RI may exist in the front direction. A configuration may be adopted in which any element selected from the width, length, or area of the identification portion of information is increased. In this example, the diameter of the indicator At6 increases according to the depth of the attention area RI. As a result, it is possible to realize an error display function that accurately displays a range in which the attention area RI may exist, according to the magnitude of the error in the actual position estimation result.

As described above, in the schema image reflecting the position estimation error, the position estimation error can be reflected according to the depth information (distance), and the schema image showing the possible existence range of the attention area RI can be displayed. . As a result, the schema image can be displayed using an indicator that allows the operator to more intuitively recognize the possible existence range of the attention area RI.

(Reflecting the depth and thickness of the region of interest in the depth direction)
FIGS. 19A to 19C are examples of mammography images in which the depth direction thickness of the attention area is different. FIGS. 20A to 20C are another examples of schema images reflecting depth information in which the depth and thickness of the regions of interest corresponding to FIGS. 19A to 19C differ in the depth direction. The display method of the position of the attention area RI in is the same as in FIGS. 15(a) to (c).

The schema image generation unit 414 classifies the thickness of the attention region R1 in the depth direction into “small, medium, and large” as shown in FIGS. , one or more elements selected from color, density, brightness, transparency, and texture of indicators At1 and At2 representing the position of the attention area in the front direction are made different corresponding to the classification depth, and classification By varying one or more elements selected from the width, length, or area of the indicator At3 representing the position of the attention area in the front direction corresponding to the thickness, both depth and thickness in the depth direction are different. A schema image reflecting the information may be generated or displayed.

Thus, the ultrasound operator can grasp the thickness of the breast in the region of interest in the depth direction from the schema image in addition to the position in the depth direction of the breast in the region of interest during ultrasonic examination. Here, too, according to the present embodiment, since information is transmitted about the thickness of the breast in the region of interest in the depth direction by changing the mode of the schema image itself, movement of the line of sight of the ultrasound operator is displayed as the target region. Suppressed within range with the screen. In addition, the ultrasound operator can visually recognize the thickness of the breast in the depth direction of the region of interest in a state that is closer to the operator's sense through vision, from the change in the mode of the indicator portion of the schema image itself. Therefore, it is possible to eliminate the work of the ultrasound operator to read out the mammography image and confirm the thickness of the region of interest in the depth direction during the ultrasound examination, thereby reducing the work burden on the ultrasound operator and improving the examination efficiency. can.

Note that one or more elements selected from the width, length, or area of the indicator At3 representing the position of the attention area in the front direction are made different in accordance with the size of the range of the attention area R1 in the front direction. Alternatively, according to the operator's selection, a schema image reflecting the size of the range in the front direction instead of the thickness in the depth direction may be generated or displayed. Accordingly, the operator can grasp the size of the range of the attention area in the front direction from the schema image in addition to the position of the attention area in the depth direction of the breast during ultrasonic examination.

(When the method of specifying the attention area is different)
FIG. 21 is a diagram showing classification of methods for specifying regions of interest in an MLO image (left side in the column) and a CC image (right side in the column) by a radiologist. The upper column in FIG. 21 shows when both the attention area R1 is specified for the MLO image and the CC image, the middle column shows when the attention area R1 is selected and specified for either the MLO image or the CC image, and the lower column shows the MLO. A case is shown in which a plurality of regions of interest R1 are designated for each of the image and the CC image. In addition, the left column of FIG. 21 shows that, for example, when the radiologist designates a point-like region of interest R1 consisting of a narrow region for calcification, etc., the right column shows that the radiologist specifies, for example, a tumor, breast This represents a case where a rectangular region of interest R1 having a width and height equal to or greater than predetermined values is specified for an expansion, cyst, mammary duct, or the like. Each image is displayed with the left and right breasts facing sideways.

FIG. 22 shows an example of a schema image in which the depth information is reflected in different ways according to the classification of the attention area designation method in FIG. 21 .

As shown in the upper column of FIG. 22, when both the attention area R1 is specified for the MLO image and the CC image, the intersection of the corresponding indicator bars At4 indicates the position of the attention area in the schema image in the front direction. represent.

The left orchid is a schema image when a point-like attention area R1 is specified, and displays a bar-like indicator bar At4 indicating the position of the attention area in the front direction. In this case, the intersection of the indicator bar At4 and the diameter perpendicular to the schema image corresponds to the X-direction position of the region of interest RI in the mammography image (MLO, CC).

The right side is a schema image when a rectangular attention area R1 having a width and height equal to or greater than a predetermined value is specified, and displays a strip-shaped indicator band At5 representing the position of the attention area in the front direction. At this time, the center of the rectangular region of interest R1 may be aligned with the center line of the indicator band At5. Alternatively, the width of the indicator band At5 may correspond to the larger one of the width and height of the region of interest R1. At this time, the width of the region of interest R1 is normalized with respect to the full width of the breast image in the X direction, the height of the region of interest R1 is normalized with respect to the full width of the breast in the Z direction of the breast image, and the width of the indicator band At5 and You can make it correspond.

As a result, the size of the region of interest in the mammography image and the size of the indicator band At5 in the schema image can be quantitatively approximated for display, making it easy to confirm the range to be scanned in ultrasound examination. .

Further, as shown in the upper column of FIG. 22, when the attention area R1 is selected and specified for the MLO image or the CC image, one displayed indicator bar At4 indicates the attention area in the schema image in the front direction. Represents a possible range.

Further, as shown in the lower column of FIG. 22, when a plurality of attention areas R1 are specified for each of the MLO image and CC image, two indicator bars At4 are displayed for each attention area. The intersection point represents the position of the region of interest in the schema image in the front direction. At this time, one or more elements selected from the color, density, brightness, transparency, and texture of the indicator bar At4 may differ for each attention area.

This makes it easy to distinguish between indicator bars At4 corresponding to different attention areas, and to easily recognize the intersection of two indicator bars At4 corresponding to the same attention area.

In addition, when the information on the region of interest is accompanied by findings related to the identification of tumors, duct dilatation, cysts, calcifications, ducts, etc. by the radiologist, indicator bar At4 or indicator band At5 in the schema image is You may display in a different aspect according to findings. Alternatively, the information about the finding may be script-displayed in or near the indicator bar At4 or the indicator band At5.

This allows the inspector to grasp the finding information from the schema image.

<<Embodiment 2>>
Although the medical image generating apparatus 4 according to Embodiment 1 has been described above, the present disclosure is not limited to the above embodiment except for its essential characteristic components. For example, a form obtained by applying various modifications to the embodiment that a person skilled in the art can think of, or a form realized by arbitrarily combining the constituent elements and functions of each embodiment without departing from the scope of the present invention. are also included in this disclosure. Below, the above-described second embodiment will be described as an example of such a form.

Embodiment 1 above shows that the nipple position can be calculated by the known method described in, for example, Japanese Patent Application Laid-Open No. 2015-104464, as described above. However, if the detection of the nipple position and nipple axis fails due to factors such as the nipple not protruding in the mammography image, an error may occur in the estimation result and an erroneous position may be presented to the operator.

On the other hand, in the medical image generating apparatus according to the second embodiment, in the medical image generating apparatus 4, the operator is presented with the detection results of the nipple position, the breast end, and the nipple axis. Adopt a modifiable configuration. As a result, the position and thickness of the attention area RI estimated with higher accuracy based on the detection result corrected by the operator can be reflected in the schema image and displayed.

As a specific aspect, in the medical image generating apparatus according to Embodiment 2, the display control unit 43 causes the display device 6 to display a mammography image, and furthermore, in a state in which the mammography image is displayed on the display device 6, the mammography is performed. An operation input unit 44 for receiving an operation input from the operator to change any of the nipple position, the breast end, and the inclination of the nipple axis detected from the mammography image in the breast image included in the image, and front information The calculation unit 412 calculates corrected front information representing the range of the attention area RI in the front direction based on the mammography image, information on the attention area RI, and operation input. Based on the information and the operation input, corrected depth information representing the position of the attention area RI in the depth direction is calculated, and the schema image generation unit 414 generates a corrected schema image reflecting the corrected front information and the corrected depth information, The display control unit 43 employs a configuration that causes the display device 6 to display the corrected schema image.

23 is a flowchart of medical image generation processing in the medical image generation apparatus according to Embodiment 2. FIG. In FIG. 23, the medical image generating process in the medical image generating apparatus according to the second embodiment includes the process of steps S21 to S23 after the process of displaying the schema image in step S11. different from 1. The processing in steps S1 to S12 is the same as the processing in the first embodiment shown in FIG. 7, so the same numbers are given and the description is omitted.

The processing of steps S21 to S23 will be described below with reference to the drawings.

The upper diagram in FIG. 24A shows mammography images (MLO, CC) displayed in the medical image generating apparatus of Embodiment 2, and the lower diagram shows an example of schema images.

First, in step S11, a display screen including a schema image is created, output to the display device 6, and displayed on the display device 6 together with the MLO image and/or the CC image.

Next, in step S21A, an operation input as to whether or not the schema image needs to be modified is accepted (step S21A). If not (No), the process proceeds to step S12 to continue the process.

On the other hand, if correction is required (Yes), the process proceeds to step S22A, where the nipple position Ni1 and the breast end Br1 (corresponding to the end point Xmax1 in the X direction of the MLO image Im1 in FIG. 9) in the MLO image (left diagram). , the nipple reference line (nipple axis Nl1). In addition, the nipple position Ni2, the breast ends Br21 and Br22 (corresponding to the X-direction end points Xmin2 and Xmax2 of the CC image Im2 in FIG. 9, respectively), and the nipple reference line (nipple axis Nl2) in the CC image (right-hand diagram) are indicate.

Next, in step S23A, change input from the operator is received for one or more conditions selected from the inclinations of the nipple reference lines Nl1 and Nl2, the nipple positions Ni1 and Ni2, and the positions of the breast ends Br1, XBr21 and XBr22. . Changing the inclination of the nipple reference lines Nl1, Nl2 changes the nipple positions Ni1, Ni2, or changes any point between the nipple positions Ni1, Ni2 and the breast baseline on the nipple reference lines Nl1, Nl2. Defined as breast base points Nl1b, Nl2b, the positions of the breast base points Nl1b, Nl2b may be moved along the breast base line. In this example, as shown in FIG. 24(a), the breast base points Nl1b and Nl2b are the intersections of the nipple base lines Nl1 and Nl2 and the breast base line, and the positions of the breast base points Nl1b and Nl2b are set to the breast base line. It was configured to move along.

Here, as described above, the breast baseline line refers to the “chest wall line in the breast image”, the “straight line connecting the upper end and the lower end of the breast image”, or the “line along the Y axis passing through the end point in the positive direction of the X axis in the mammography image”. Either "parallel straight lines" can be used.

Also, the breast base point Nl2b may be the middle point between the breast ends Br21 and Br22.

The correction point CR in this example is the breast base point Nl1b of the MLO image.

FIG. 24(b) shows similarly mammography images (MLO, CC) in the medical image generating apparatus of Embodiment 2 after being corrected by the operator, and the lower figure similarly shows an example of a schema image. . As shown in FIG. 24(b), in the corrected portion CR′, the breast base point N11b in the MLO image of the breast on the L (left) side is moved upward (MV). , the position of the indicator At4 indicating the position of the region of interest RI in the MLO image of the L (left) breast in the schema image moves downward (MV).

Next, returning to step S7, based on the corrected information, the front information of the attention area RI, the depth information about the depth of the attention area (step S8), and the depth information about the thickness of the attention area (step S9) are obtained. Calculated and modified schema images are generated in different display modes according to the content of the modified depth information (step S10), and the modified schema images are displayed together with the MLO and/or CC images ( step S11).

As a result, with the MLO image and the CC image displayed, the operator can detect the nipple reference lines Nl1 and Nl2, the nipple positions Ni1 and Ni2, and the breast ends Br1, Br21, and Br22 in the MLO image and the CC image. The deviation can be corrected, and the accuracy of the position and thickness of the attention area RI displayed as a schema image can be improved.

In addition, as another aspect of step S22A, the nipple position Ni1 in the MLO image and the nipple position Ni2 in the CC image are displayed, and in step S23A, changes are input from the operator for the conditions of the nipple positions Ni1 and Ni2. may be configured. As a result, by hiding the nipple reference lines Nl1 and Nl2 and the breast ends Br1, Br21 and Br22, the visibility of the MLO image and the CC image is improved, and only the nipple positions Ni1 and Ni2 can be corrected. , operability is good, and simple and effective correction is possible.

In step S22A, as still another aspect, a configuration in which the inclinations of the nipple axes Nl1 and Nl2 are changed, or a configuration in which the positions of the breast ends Br1, Br21 and Br22 are changed may be employed.

<<Embodiment 3>>
In the medical image generating apparatus according to Embodiment 3, in the medical image generating apparatus 4 according to Embodiments 1 and 2, furthermore, in either the MLO image Im1 or the CC image Im2 in the mammography image, It is characterized in that a reference candidate range RL in the other image is calculated based on the attention area RI and displayed on the other image.

<Configuration>
The configuration of the medical image generation apparatus according to Embodiment 3 will be described below.

The medical image generation device according to Embodiment 3 is constructed by the server device 4 that constitutes the main part of the medical image archiving communication system 10 shown in FIG. It functions as a medical image generating apparatus that calculates a reference candidate range RL by inputting information about the attention area RI. As in the first embodiment, it may function as a medical image generating apparatus that generates a schema image.

The medical image generation apparatus according to Embodiment 3 employs the same configuration as the server device 4 according to Embodiment 1 shown in FIG. 41 and the control unit 41A according to the third embodiment shown in FIG.

FIG. 26 is a functional block diagram showing the functional configuration of the control section 41A of the server device in the medical image generating device of Embodiment 3. As shown in FIG. The control unit 41A receives the MLO image, the CC image, and the information about the region of interest or the candidate of the region of interest as input, and executes a program of medical image generation processing for calculating the reference candidate range.

As shown in FIG. 26, the control unit 41A has an image and information acquisition unit 411, a front information calculation unit 412, a depth information calculation unit 413, and a reference candidate range calculation unit 414A. Among these, the configuration and operation of the image and information acquisition unit 411, the front information calculation unit 412, and the depth information calculation unit 413 are the same as the respective elements in the control unit 41 according to the first embodiment, and thus the description is omitted. do.

The reference candidate range calculation unit 414A in the control unit 41A acquires the front information from the front information calculation unit 412 and the depth information from the depth information calculation unit 413, and designates it in either the MLO image Im1 or the CC image Im2. A reference candidate range RL in the other image is calculated based on the attention area RI thus obtained.

Specifically, the reference candidate range calculation unit 414A defines a virtual schema image Im3′ corresponding to the breast image included in the MLO image Im1 and the breast image included in the CC image Im2, and calculates the MLO image Im1 or CC image Im2. The relative position of the region of interest RI in the breast image included in one of the images in the breast baseline direction Cw1 or Cw2 of the breast image is placed on the breast image included in the other image via the virtual schema image Im3′. A reference candidate range RL is calculated as a range in the direction Cw2 or Cw1 of the breast base line of the breast image that is projected on the target.

<Operation>
Processing for calculating the reference candidate range RL in the medical image generating apparatus according to the third embodiment will be described below with reference to the drawings.

(Operation of displaying reference candidate range on MLO image based on attention area of CC image)
FIG. 27(a) is a schematic diagram for explaining a method of specifying a reference candidate range in a breast image included in a CC image based on a region of interest in a breast image included in an MLO image. In FIG. 27A, Im1 represents an MLO image, Im2 represents a CC image, and Im3' represents a virtual schema image, which is a virtual schema image. The imaging angle θ is the angle (tilt angle) formed by the radiation irradiation direction when acquiring the MLO image Im1 with respect to the vertical direction. Sl1 and Sl2 are skin lines of the breast image.

The reference candidate range RL in the breast image in the CC image Im2 based on the region of interest RI in the breast image in the MLO image Im1 is calculated as follows.

First, a breast image included in the MLO image Im1 and a virtual schema image Im3' corresponding to the breast image included in the CC image Im2 are set in the virtual space. The method of arranging the MLO image Im1, the CC image Im2, and the virtual schema image Im3′ in FIG. 27A is the same as the method of arranging the MLO image Im1, the CC image Im2, and the schema image Im3 in FIG.

Next, the relative position d1 in the breast base line Cw1 direction of the region of interest RI in the breast image included in the MLO image Im1 is calculated.

Next, corresponding to the relative position d ₁ in the MLO image Im1, the intersection points p ₁ and p ₂ where the straight line Nl1′ indicating the relative position d in the virtual schema image Im3′ intersects the outer circumference of the breast in the virtual schema image Im3′ are determined. calculate.

Here, the relative position d in the schema image Im3 is defined as the distance from the nipple axis Nl1 of the straight line Nl1′ parallel to the imaging angle θ, and the distance from the nipple axis Nl1 to the breast end in the breast image included in the MLO image Im1. the d_lowWhen
d = d₁/d_low
calculated as

Next, the relative positions d _p1 and d _p2 of the intersection points p ₁ and p ₂ in the direction of the breast base line Cw2 in the virtual schema image Im3′ are calculated.

Here, the relative positions d _p1 and d _p2 are defined as the distances d ₂₁ and d ₂₂ from the intersection points p ₁ and p ₂ to the nipple axis Nl2, and the distance from the nipple axis Nl2 to the breast end in the breast image included in the CC image Im2. When the distances between are d _in and d _out ,
_d21 = _dp1 / _din
d22 = _dp2 / _dout
calculated as

A reference candidate range RL in the CC image Im2 is defined as a range sandwiched between relative positions _d21 and _d22 defined in the breast base line Cw2 direction. In this example, the reference candidate range RL is displayed as two straight lines of length d ₂₁ +d ₂₂ .

Further, the positions of the two straight lines in the depth direction (Z direction in Im2) perpendicular to the breast base line Cw2 are based on the depth information representing the position in the depth direction (Z direction in Im1) of the region of interest RI in the MLO image Im1. A specified configuration may also be used. Specifically, for example, the distance z ₁ from the nipple position Ni1 from the breast baseline Cw1, the distance z _RI from the central breast baseline Cw1 in the depth direction (the Z direction in Im1) of the region of interest RI, and the nipple position Ni2 z ₂ is the distance from the breast baseline Cw2, and z _RL is the distance from the central breast baseline Cw2 in the depth direction (Z direction in Im2) of the reference candidate range RL,
_zRL = _zRI * _z2 / _z1
may be calculated as

In addition, instead of the calculation method described above, z _RL is calculated by calculating the Z-direction distance (Z ₁ −Z _RI ) in Im1 from the nipple position Ni1 to ZR1 in the MLO image Im1, and calculating from the nipple position Ni2 in the CC image Im2. A position distant (Z ₁ −Z _RI ) in the Z direction (chest wall direction) in Im2 may be obtained, and the height of the position in the Z direction may be set as Z _RL .

As a result, as shown in FIG. 27B, the attention area RI in the breast image included in the MLO image and the reference candidate range RL in the breast image included in the CC image are displayed.

Next, a mode of the reference candidate range RL displayed in the breast image included in the CC image when the position of the region of interest RI in the breast image included in the MLO image is different will be described.

FIGS. 28(a) to (c), FIGS. 29(a) and (b) show regions of interest existing at different positions in the breast images included in the MLO image, in the breast images included in the identified CC images. FIG. 4 is a schematic diagram showing a mode of a reference candidate range; As shown in the figure, the width of the reference candidate range RL displayed in the CC image in the direction of the breast base line Cw2 is displayed differently depending on the position of the region of interest RI in the MLO image in the direction of the breast base line Cw1. For example, as shown in FIGS. 28(a), (b), (c), or FIGS. 28(a), 29(a), and (b), the position of the region of interest RI in the breast image in the MLO image However, the width of the reference candidate range RL displayed in the CC image in the direction of the breast base line Cw2 is reduced as it approaches the edge of the breast in the direction of the base line Cw1 of the breast. In this case, the operator can narrow down the image range to be referred to, thereby improving the inspection efficiency.

(Operation of displaying reference candidate range on MLO image based on attention area of CC image)
FIG. 30(a) is a schematic diagram for explaining a method of specifying a reference candidate range in an MLO image based on information on an attention area in an attention area in a CC image.

Here, as in FIG. 27A, the reference candidate range RL in the CC image Im1 based on the attention area RI of the CC image Im2 is calculated as follows.

First, a virtual schema image Im3′ corresponding to the breast image included in the CC image Im2 and the breast image included in the MLO image Im1 is defined. The method of arranging the MLO image Im1, the CC image Im2, and the virtual schema image Im3′ in FIG. 30A is the same as the method of arranging the MLO image Im1, the CC image Im2, and the schema image Im3 in FIG.

Next, the relative position d ₂₁ of the region of interest RI in the breast image included in the CC image Im2 in the direction of the first breast base line Cw2 is calculated.

Next, corresponding to the relative position d ₂₁ in the CC image Im2, the intersections p ₁ and p ₂ where the straight line Nl2′ indicating the relative position d in the virtual schema image Im3′ and the peripheral line of the breast in the virtual schema image Im3′ are set. calculate.

Here, the relative position d in the virtual schema image Im3′ is defined as the distance to a straight line Nl2′ parallel to the nipple axis Nl2._inWhen
d = d_{twenty one}/d_in
calculated as

Next, the relative positions d _p1 and d _p2 of the intersection points p ₁ and p ₂ in the direction of the breast base line Cw1 in the virtual schema image Im3′ are calculated.

Here, the relative positions d _p1 and d _p2 are defined as the distances d ₁₁ and d ₁₂ from the intersection points p ₁ and p ₂ to the nipple axis Nl1, and the distance from the nipple axis Nl1 to the breast end in the breast image included in the CC image Im1. When the distance of is d _low ,
_d11 = _dp1 / _dlow
d12 = _dp2 / _dlow
calculated as

A reference candidate range RL in the CC image Im1 is defined as a range sandwiched between relative positions _d11 and _d12 defined in the breast base line Cw1 direction. In this example, the reference candidate range RL is displayed as two straight lines of length d ₁₁ +d ₁₂ .

As in FIG. 27A, the positions of the two straight lines in the depth direction (Z direction in Im1) perpendicular to the breast base line Cw1 correspond to the depth direction (Z direction in Im1) of the region of interest RI in the MLO image Im2. ) may be specified based on the depth information representing the position. For example, the distance z RI from the central breast baseline Cw2 in the depth direction (Z direction in Im2) of the region of interest _RI , the distance from the central breast baseline Cw1 in the depth direction (Z direction in Im1) of the reference candidate range RL When _zRL is
_zRL = _zRI * _z1 / _z2
may be calculated as

Also here, z _RL is calculated by calculating the Z-direction distance (Z ₂ −Z _RI ) in Im2 from the nipple position Ni2 to ZR1 in the CC image Im2 instead of the calculation method described above, and calculating the nipple position Ni1 in the MLO image Im1. A position distant (Z ₂ -Z _RI ) in the Z direction (thoracic wall direction) in I12 from is obtained, and the height of the position in the Z direction may be defined as Z _RL .

Furthermore, the positions of the two straight lines in the direction perpendicular to the breast base line Cw2 may be specified based on the existence range of the attention area RI in the CC image Im2 in the depth direction (the Z direction in Im2).

Thereby, as shown in FIG. 30(b), the attention area RI in the breast image included in the CC image and the reference candidate range RL in the breast image included in the MLO image are displayed.

Again, as shown in FIGS. 31(a) to (c) and FIGS. 32(a) and (b), the region of interest RI in the CC image is displayed in the MLO image according to the position in the breast baseline Cw2 direction. The width of the reference candidate range RL is different. For example, the closer the position of the attention area RI in the breast image in the CC image is to the end of the breast in the direction of the breast baseline Cw2, the smaller the width of the reference candidate range RL displayed in the MLO image in the direction of the breast baseline Cw1 ( 31(a), (b), (c) or FIG. 31(a), FIG. 32(a), (b)).

<Effect>
FIG. 33(a) is an example of a CC image in which a region of interest is displayed in a breast image, and (b) is an example of an MLO image in which a reference candidate range RL is displayed in a breast image.

In this way, in the CC image Im2 of the mammography image, based on the designated attention area RI as shown in FIG. 33(a), as shown in FIG. It can be seen that the reference candidate range RL is displayed.

As described above, in the medical image generating apparatus according to the third embodiment, the region of interest RI specified in the MLO image Im1 or the CC image Im2 is calculated via the virtual schema image Im3′ on the other image. By narrowing down the reference candidate range RL, it is possible to narrow down the image range that the operator should refer to and improve the inspection efficiency.

Further, the range in the depth direction (Z direction) of the reference candidate range RL is further narrowed down in the depth direction (Z direction) by specifying the range in the depth direction (Z direction) of the attention area RI in the mammography image. It is possible to present the reference candidate range RL, and it is possible to further narrow down the image range to be referred to by the operator and further improve the inspection efficiency.

In the medical image generation apparatus 4 according to Embodiments 1 and 2, the schema image generation unit generates a schema image by reflecting the front information and the depth information, and the display control unit outputs the schema image to the display device. It is configured to be displayed. However, in the medical image generating apparatus according to Embodiment 3, based on the region of interest RI designated in one of the MLO image Im1 and CC image Im2 in the mammography image, the reference candidate range RL is displayed in the other image. In addition to the configuration, a schema image may be generated by reflecting at least the front information, and the schema image based on the front information may be displayed.

Further, the imaging angle of the MLO image Im1 is not limited to the above, and it goes without saying that, for example, it may be defined as the angle formed by the radiation irradiation direction when acquiring the MLO image Im1 with respect to the horizontal direction.

<<Modification>>
Although the present disclosure has been described above based on the embodiments, the present disclosure is not limited to the above embodiments, and various modifications are possible.

(1) In the above-described embodiment, the embodiment of the medical image generating apparatus has been described as an example. However, for example, the medical image generating apparatus may be a medical image generating program that causes the computer PC to execute.

(2) In the above-described embodiment, the shama images generated by the server device 4 are displayed on the display device 6 constituting the medical image archiving and communication system 10 (PACS). However, the schema image may be displayed on the display screen of the ultrasonic diagnostic apparatus or the display device 33 of the diagnostic support device (CAD) 3, for example.

Further, in the above-described embodiment, an example in which the program related to the schema image generation processing is executed by the control unit 41 of the server device 4 has been described. However, the schema image generation processing may be configured to be executed in the ultrasonic diagnostic apparatus.

Thereby, the operator can grasp the scanning position in the ultrasonic examination on the display screen of the ultrasonic diagnostic apparatus, and the operation becomes easier.

(3) In the above-described embodiment, the schema image generator 414 classifies the depth of the attention region R1 into "shallow, medium, and deep" and classifies the thickness in the depth direction into "small, medium, and large." In addition, the mode of the indicator representing the position of the attention area in the front direction is made different in accordance with the classification depth and the classification thickness. However, the depth of the attention area R1 is classified into "shallow and deep", and the thickness in the depth direction is classified into "small and large", and the mode of the indicator representing the position of the attention area is changed according to the classification. good too. This makes it possible to roughly grasp the scanning position in the ultrasonic examination. Alternatively, it may be classified into four or more stages. The scanning position can be grasped more accurately.

(4) In the above-described embodiment, the description has been given by exemplifying the configuration in which designation of the region of interest by the interpreting doctor and generation and display of the shamanic image are performed as continuous processing. However, the present disclosure is not limited to this, and by once storing the information of the attention area designated by the interpretation doctor in a medical image database, etc. and display may be performed separately.

(5) In the above-described embodiment, an examination of a breast cancer tumor has been exemplified and explained as an example of medical image generation. However, the present disclosure is not limited to this, and may be applied to generation of schema images for various diseases using mammography.

(6) In the above-described embodiment, an example of mammography using X-rays has been described as a modality for capturing a mammography tomographic image of a subject. As tomographic image information, a schematic image is formed using mammography tomographic image information obtained by various modalities for taking tomographic images of the subject, such as mammography using positron emission tomography (PET). may

(7) In the medical image generating apparatus according to the above embodiment, the data storage unit, which is a storage device, is included in the medical image archiving communication system (PACS). , a hard disk drive, an optical disk drive, a magnetic storage device, etc. may be externally connected to a medical image archival communication system (PACS) via the Internet.

(8) Although the present disclosure has been described based on the above embodiments, the present disclosure is not limited to the above embodiments, and the following cases are also included in the present invention.

For example, the present invention may be a computer system comprising a microprocessor and memory, the memory storing the computer program, and the microprocessor operating according to the computer program. For example, it may be a computer system that has a computer program of the medical image generation method of the present disclosure and operates according to this program (or instructs each connected part to operate).

The present invention also includes a case in which the entire medical image generating apparatus is configured by a computer system including a microprocessor, a recording medium such as a ROM and a RAM, a hard disk unit, and the like. The RAM or hard disk unit stores a computer program that achieves the same operations as the above devices. Each device achieves its function by the microprocessor operating according to the computer program.

Also, part or all of the components constituting each of the devices described above may be configured from one system LSI (Large Scale Integration). A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple components on a single chip. Specifically, it is a computer system that includes a microprocessor, ROM, RAM, etc. . These may be made into one chip individually, or may be made into one chip so as to include part or all of them. LSIs are also called ICs, system LSIs, super LSIs, and ultra LSIs depending on the degree of integration. The RAM stores a computer program that achieves the same operations as those of the above devices. The system LSI achieves its functions by the microprocessor operating according to the computer program. For example, the present invention includes a case where the medical image generation method of the present invention is stored as an LSI program, and this LSI is inserted into a computer to execute a predetermined program (medical image generation method).

It should be noted that the method of circuit integration is not limited to LSI, and may be implemented using a dedicated circuit or a general-purpose processor. An FPGA (Field Programmable Gate Array) that can be programmed after the LSI is manufactured, or a reconfigurable processor (Reconfigurable Processor) that can reconfigure the connections and settings of the circuit cells inside the LSI may be used.

Furthermore, if an integration technology that replaces the LSI appears due to advances in semiconductor technology or another derived technology, the technology may naturally be used to integrate the functional blocks.

Also, part or all of the functions of the medical image generating apparatus according to each embodiment may be realized by a processor such as a CPU executing a program. It may be a non-temporary computer-readable recording medium in which a program for executing the operation of the medical image generating apparatus is recorded. By recording a program or a signal on a recording medium and transferring it, the program may be executed by another independent computer system, and the above program can be distributed via a transmission medium such as the Internet. Needless to say.

Further, each component of the medical image generation apparatus according to the above embodiments may be implemented by a programmable device such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a processor, and software. The latter configuration is the so-called GPGPU (General-Purpose computing on Graphics Processing Unit). These components can be single circuit components or aggregates of multiple circuit components. Also, a plurality of components can be combined to form a single circuit component, or a plurality of circuit components can be aggregated.

Also, the division of functional blocks in the block diagram is an example, and a plurality of functional blocks can be realized as one functional block, one functional block can be divided into a plurality of functional blocks, and some functions can be moved to other functional blocks. may Moreover, single hardware or software may process the functions of a plurality of functional blocks having similar functions in parallel or in a time-sharing manner.

Also, the order in which the above steps are performed is for illustrative purposes in order to specifically describe the present invention, and orders other than the above may be used. Also, some of the above steps may be executed concurrently (in parallel) with other steps.

Also, at least part of the functions of the medical image generating apparatus according to each embodiment and its modifications may be combined. Furthermore, the numbers used above are all examples for specifically describing the present invention, and the present invention is not limited to the numbers shown.

Furthermore, the present invention also includes various modifications in which those skilled in the art make modifications to the present embodiment.

≪Summary≫
As described above, the medical image generating apparatus according to the embodiment is a medical image generating apparatus that generates a schema image based on a mammography image and information on a region of interest, wherein the mammography image and the information on the region of interest are obtained from a storage medium. Represents the range in the front direction of the attention area for the attention area in the breast image included in the mammography image based on the image and information acquisition unit that acquires the acquired mammography image and the information on the attention area A position of the attention area in the depth direction of the attention area in the breast image included in the mammography image is calculated based on a front information calculation unit that calculates front information and information on the obtained mammography image and the attention area. a depth information calculation unit that calculates depth information to represent; a schema image generation unit that generates a schema image by reflecting the front information and the depth information; and a display control unit that displays the schema image on a display device. characterized by

In another aspect, in any one of the above-described aspects, the schema image generation unit changes the aspect of the identification unit reflecting the front information in the schema image according to the content of the depth information. An image may be generated.

With such a configuration, the ultrasound operator can grasp the position of the region of interest in the depth direction of the breast from the schema image during ultrasound examination. At this time, by changing the mode of the schema image itself, the position of the region of interest in the depth direction of the breast is displayed. From the change in the mode, the position of the attention area in the depth direction of the breast can be recognized through vision in a state closer to the operator's sense. Therefore, it is possible to eliminate the work of the ultrasound operator to read out the mammography image and confirm the position of the region of interest in the depth direction during the ultrasound examination, thereby reducing the work burden on the ultrasound operator and improving the examination efficiency. can.

In another aspect, in any one of the above aspects, the depth information calculation unit calculates the total breast thickness of the breast image and the attention area depth of the attention area in the breast image, and calculates the attention area. A ratio of the depth to the total breast thickness may be calculated, and the ratio may be used as the depth information.

With such a configuration, the ultrasound operator can grasp the position of the region of interest in the depth direction of the breast from the schema image during ultrasound examination.

In another aspect, in any one of the above aspects, the depth information calculation unit further calculates a total breast thickness of the breast image and a region-of-interest thickness of the region of interest in the breast image, A configuration may be employed in which the ratio of the thickness of the region of interest to the total thickness of the breast is calculated, and the ratio is used as the depth information.

With such a configuration, the ultrasound operator can grasp the depth-direction thickness of the breast of the region of interest from the schema image in addition to the depth-direction position of the region of interest during ultrasound examination.

In another aspect, in any one of the above aspects, the schema image generation unit sets the aspect of the identification unit of the front information in the schema image to the type of the parameter that constitutes the depth information, and the type of the parameter that constitutes the depth information. It is good also as a structure which differs according to the magnitude of a value.

With such a configuration, the ultrasound operator can grasp a plurality of types of parameters in the depth direction of the breast in the region of interest from the schema image during ultrasound examination.

In another aspect, in any one of the above-described aspects, the schema image generation unit includes the color, density, brightness, transparency, texture, width, length, or area of the identifying portion of the front information in the schema image. The element selected from may correspond to the type of the parameter, and the corresponding element may differ according to the value of the parameter.

With such a configuration, the ultrasound operator can determine the magnitude of the values of a plurality of types of parameters in the depth direction of the breast in the region of interest during ultrasound examination, in the schema image identification unit corresponding to each type of parameter. It can be grasped from the change.

Further, in another aspect, in any one of the above aspects, when the information on the region of interest is information indicating a region having a size less than a predetermined value in the mammography image, the schema image generation unit generates the schema The identification portion of the front information in the image may be linear.

Further, in another aspect, in any one of the above aspects, when the information on the region of interest is information indicating a region having a size equal to or greater than a predetermined value in the mammography image, the schema image generation unit generates the schema One or more elements selected from the width, length, or area of the identification portion of the front information in the image may be changed according to the content of the depth information.

With such a configuration, it is possible to perform display in which the size of the region of interest in the mammography image and the size of the identification portion in the schema image are quantitatively approximated, making it easy to confirm the range to be scanned in ultrasonic examination. .

In another aspect, in any one of the above aspects, when there are a plurality of pieces of information on the attention area corresponding to a plurality of attention areas, the schema image generation unit may generate the front information in the schema image. One or more elements selected from the color, density, brightness, transparency, texture, width, length, or area of the identification portion may be different for each region of interest.

With such a configuration, it becomes easy to distinguish identification portions in schema images corresponding to different attention areas, and to easily recognize an intersection of a pair of identification portions corresponding to the same attention area.

In another aspect, in any one of the above aspects, the schema image generation unit changes the aspect of the identification unit reflecting the front information in the schema image to the area of interest corresponding to the depth of the area of interest. It is also possible to generate different schema images according to the size of the possible range in the front direction.

With such a configuration, it is possible to realize an error display function that accurately displays a range in which the attention area RI may exist in accordance with the magnitude of the error in the actual position estimation result. Reflecting the estimation error, it is possible to display a schema image indicating a possible range of the region of interest RI.

In another aspect, in any one of the above aspects, the deeper the depth of the attention area, the larger the size of the range in which the attention area may exist in the front direction. In another aspect, in any one of the above aspects, the schema image generation unit selects from the width, length, or area of the identification part of the front information in the schema image as the depth of the region of interest increases. It is good also as a structure which makes it different so that any element carried out may increase.

With such a configuration, it is possible to display a schema image using an indicator that allows the operator to more intuitively recognize the possible existence range of the attention area RI.

In another aspect, in any one of the above aspects, the total breast thickness is the distance from the nipple position in the breast image to the intersection of the depth direction axis and the breast baseline in the mammography image through the nipple position. The depth of the attention area is the distance from the intersection of a straight line passing through a reference point in the attention area and parallel to the depth direction axis and a straight line passing through the nipple position and perpendicular to the depth direction axis to the reference point. A certain configuration may be used.

In another aspect, in any one of the above aspects, the depth of the region of interest is calculated from the intersection of a breast skin line and a straight line passing through a reference point in the region of interest and parallel to the depth direction axis in the mammography image. It is the distance to the reference point, and the total breast thickness may be the distance from the intersection of the straight line and the breast skin line to the intersection of the straight line and the breast baseline.

With such a configuration, it is possible to calculate the position of the attention area in the depth direction.

In another aspect, in any one of the above aspects, the depth of the region of interest is a normal line that passes through a reference point in the region of interest and passes through the nipple position to the breast baseline, or a line between the upper and lower ends of the nipple. It is the distance from the intersection of a straight line parallel to the straight line passing through the nipple position and the breast skin line to the reference point, and the total breast thickness is the distance between the straight line and the breast skin line. It may be the distance from the intersection to the intersection of the straight line and the breast base line.

With this configuration, it is possible to calculate the position of the region of interest in the depth direction even when the depth direction in the shama image does not match the direction of the breast baseline.

In another aspect, in any one of the above-described aspects, the thickness of the region of interest is defined by a straight line passing through a reference point in the region of interest and parallel to the depth direction axis in the mammography image and the outer edge of the region of interest. The total breast thickness is the distance between intersection points, and the total breast thickness may be the distance from the intersection of the straight line and the breast skin line to the intersection of the straight line and the breast baseline.

With such a configuration, it is possible to calculate the thickness of the region of interest in the depth direction.

In another aspect, in any one of the above aspects, the thickness of the region of interest is a normal line that passes through a reference point in the region of interest and passes through the nipple position to the breast base line, or the upper end and the lower end of the nipple. is the distance between two intersections of a straight line parallel to a straight line passing through the nipple position and the outer edge of the region of interest, and the total breast thickness is the distance from the intersection of the straight line and the breast skin line, It may be the distance to the intersection of the straight line and the breast base line.

With this configuration, it is possible to calculate the thickness of the region of interest in the depth direction even when the depth direction in the shama image does not match the direction of the breast base line.

In another aspect, in any one of the above aspects, the display control unit causes the display device to display the mammography image, and further displays the mammography image while the mammography image is displayed on the display device. An operation input unit that receives an operation input from the operator to change any one selected from the position of the nipple, the end of the breast, and the midpoint position of the end of the breast detected from the mammography image in the breast image included in the The front information calculation unit calculates corrected front information representing the range of the attention area in the front direction based on the mammography image, the attention area information, and the operation input, and the depth information calculation unit calculates the mammography image. , based on the information on the attention area and the operation input, corrected depth information representing the position of the attention area in the depth direction is calculated, and the schema image generation unit reflects the corrected front information and the corrected depth information. may generate a modified schema image, and the display control unit may display the modified schema image on a display device.

With such a configuration, the operator can correct the detected positional deviation of the nipple reference line, the nipple position, and the position of the breast end from the MLO image and the CC image, and the position of the attention area RI displayed as a schema image. And thickness accuracy can be improved. As a result, the position and thickness of the attention area RI estimated with higher accuracy based on the detection result corrected by the operator can be reflected in the schema image and displayed.

In another aspect, in any one of the aspects described above, the operation input unit may be configured to receive an operation input for changing the position of the nipple.

With such a configuration, the visibility of the mammography image is improved, and only the nipple position can be corrected, which enables easy and effective correction with good operability.

In another aspect, in any one of the above aspects, the operation input unit receives an operation input for changing the inclination of the nipple reference line, or the operation input unit changes the position of the breast end. A configuration may be adopted in which an operation input for changing is received.

Here, the nipple reference line is a straight line passing through the position of the nipple and perpendicular to the breast base line, a straight line passing through the position of the nipple and passing through the upper and lower ends of the nipple, and a straight line perpendicular to the straight line passing through the nipple position and the midpoint of the breast end. or a horizontal line in the mammography image that passes through the location of the nipple in the mammography image.

In another aspect, in any one of the above aspects, the mammography image may include an MLO image and a CC image.

With such a configuration, it is possible to calculate the range of the attention area in the front direction for the attention area in the mammography image based on the MLO image, the CC image, and the attention area information.

In another aspect, in any one of the above aspects, the information on the region of interest is either a region specified by a radiologist in the mammography image or a region selected by CAD based on the mammography image. It is good also as a structure which is.

With such a configuration, the position in the depth direction of the attention area in the mammography image can be calculated based on the information about the attention area obtained by the mammography image and the CAD.

In another aspect, a medical image generating apparatus for calculating a reference candidate range based on information on a region of interest and a mammography image including an MLO image and a CC image,
an image and information acquisition unit that acquires information on the mammography image and the region of interest;
a front information calculation unit that calculates front information representing a range of the attention area in the front direction for the attention area in the breast image included in the mammography image, based on the obtained mammography image and information about the attention area; ,
A reference candidate for calculating a reference candidate range for calculating a reference candidate range for the other image based on the attention area specified in one of the MLO image and the CC image, reflecting the front information. a range calculator;
The configuration may include a display control unit that causes a display device to display the mammography image and the reference candidate range.

With such a configuration, it is possible to display the reference candidate range RL on the other image based on the region of interest RI designated on one image, thereby narrowing down the image range to be referred to by the operator and improving the inspection efficiency. .

In another aspect, in any one of the aspects described above, the reference candidate range calculation unit calculates the first breast image included in the one image and the second breast image included in the other image. Set the corresponding virtual schema image,
Relative positions of the region of interest in the first breast image in a first breast baseline direction of the first breast image are relatively projected onto the second breast image via the virtual schema image. The range in the direction of the second breast base line of the second breast image may be used as the reference candidate range.

In another aspect, in any one of the above aspects, the reference candidate range calculation unit calculates a straight line passing through the position corresponding to the relative position in the virtual schema image and an outer circumference of the breast in the virtual schema image. Calculate the two points of intersection where
calculating two relative positions of the two intersection points in the second breast baseline direction in the virtual schema image;
A configuration may be employed in which a range sandwiched between the two relative positions defined in the direction of the second breast base line in the second breast image is calculated as the reference candidate range.

With such a configuration, it is possible to obtain the reference candidate range RL on the other image, which is calculated via the virtual schema image Im3′ from the region of interest RI specified in the MLO image Im1 or the CC image Im2. can be displayed on the image.

In another aspect, in any one of the above aspects, the region of interest in the breast image included in the mammography image is further determined based on the acquired mammography image and the information of the region of interest. A depth information calculation unit that calculates depth information representing the position in the depth direction of
The reference candidate range calculation unit specifies a depth direction range of the reference candidate range in the second breast image based on depth information representing a position in the depth direction of the attention area in the first breast image. may be

With such a configuration, it is possible to present the reference candidate range RL further narrowed down in the height direction, and it is possible to further narrow down the image range to be referred to by the operator and further improve the inspection efficiency.

A medical image generation program according to an embodiment causes a computer to execute a medical image generation process executed by a medical image generation apparatus that generates a schema image based on a mammography image and information on a region of interest in the mammography image. A generation program, in which medical image generation processing acquires information on the mammography image and the region of interest from a storage medium, and based on the acquired mammography image and the information on the region of interest, is included in the mammography image. For the attention area in the breast image, front information representing the range of the attention area in the front direction is calculated, and based on the obtained mammography image and the information of the attention area, the breast image included in the mammography image. For an attention area, depth information representing a position of the attention area in the depth direction is calculated, a schema image is generated by reflecting the front information and the depth information, and the schema image is displayed on a display device. and

In another aspect, in any one of the above aspects, in generating the schema image, a schema image is generated in which the aspect of the identification part of the front information in the schema image is changed according to the content of the depth information. It is good also as a structure which generates.

With such a configuration, it is possible to realize a medical image generating apparatus that allows an ultrasound operator to grasp the position of the region of interest in the depth direction of the breast from the schema image during ultrasound examination. Therefore, it is possible to implement a program that can reduce the work load on the ultrasound operator and improve the examination efficiency.

In another aspect, in the medical image generation program, in the generation of the schema image, the aspect of the identification unit reflecting the front information in the schema image is set to the front direction of the attention area corresponding to the attention area depth. It is also possible to generate different schema images according to the size of the possible range of existence in .

With such a configuration, a schema image indicating the possible existence range of the attention area RI is generated by reflecting the position estimation error according to the depth information (distance) in accordance with the magnitude of the error in the actual position estimation result. A program to display can be realized.

In another aspect, the medical image generation program according to any one of the above aspects,
Further, causing a display device to display the mammography image showing the position of the nipple, the edge of the breast, and the nipple reference line in the breast image included in the mammography image detected from the mammography image,
receiving an operation input from an operator to change any one selected from the position of the nipple, the edge of the breast, and the inclination of the nipple reference line while the mammography image is displayed on the display device;
Based on the mammography image, the information of the attention area, and the operation input, corrected front information representing the range of the attention area in the front direction is calculated;
calculating corrected depth information representing the position of the attention area in the depth direction based on the mammography image, the attention area information, and the operation input;
A modified schema image may be generated by reflecting the modified front face information and the modified depth information, and the modified schema image may be displayed on a display device.

With such a configuration, it is possible to realize a program that reflects and displays the position and thickness of the attention area RI estimated with higher accuracy in the schema image based on the detection result corrected by the operator.

In another aspect, the medical image generation program causes a computer to execute medical image generation processing for calculating a reference candidate range based on a mammography image including an MLO image and a CC image and information on a region of interest in the mammography image. A medical image generation program,
Medical image generation processing
Acquiring information on the mammography image and the region of interest;
Based on the obtained mammography image and the information of the attention area, for the attention area in the breast image included in the mammography image, front information representing the range in the front direction of the attention area is calculated;
reflecting the front information and calculating a reference candidate range in the other image based on the region of interest designated in either one of the MLO image and the CC image;
The mammography image and the reference candidate range may be displayed on a display device.

With such a configuration, a program that displays a reference candidate range RL on the other image based on a region of interest RI specified in one image, narrows down the image range that the operator should refer to, and contributes to improving inspection efficiency. realizable.

Further, in another aspect, the medical image generation program further comprises, based on the obtained mammography image and information on the region of interest, the region of interest in the breast image included in the mammography image. calculating depth information representing the position in the depth direction of the target area, and specifying the range in the depth direction of the reference candidate range in the other image based on the depth information representing the position in the depth direction of the attention area in the one image may be configured. With such a configuration, it is possible to present a reference candidate range RL further narrowed down in the height direction, and it is possible to realize a program capable of further narrowing down the image range to be referred to by the operator and further improving inspection efficiency.

The medical image generating apparatus and the medical image generating program according to the embodiment of the present disclosure can grasp the position of the region of interest in the mammography image in the depth direction of the breast from the schema image. It is widely applicable to systems (PACS).

100 Medical Imaging System 10 Medical Image Archival Communication System (PACS)
1 Mammography device 3 Diagnosis support device (CAD)
31 CPUs
32 memory 33 display device (display)
34 input/output unit (interface)
4 server device, (medical image generation device)
41, 41A control unit 411 image and information acquisition unit 412 front information calculation unit 413 depth information calculation unit 414 schema image generation unit 414A reference candidate range calculation unit 42 memory 43 display control unit 44 operation input unit 45 input/output unit (interface)
5 medical image database,
6 Display device (display)

Claims (21)

A medical image generating apparatus that generates a schema image based on a mammography image and information on a region of interest,
an image and information acquisition unit that acquires information on the mammography image and the region of interest;
a front information calculation unit that calculates front information representing a range of the attention area in the front direction for the attention area in the breast image included in the mammography image, based on the obtained mammography image and information about the attention area; ,
a depth information calculation unit that calculates depth information representing a position of the region of interest in the depth direction for the region of interest in a breast image included in the mammography image, based on the acquired mammography image and information of the region of interest; ,
a schema image generation unit that generates a schema image by reflecting the front information and the depth information;
a display control unit for displaying the schema image on a display device,
The schema image generation unit generates a schema image in which the aspect of the identification unit reflecting the front information in the schema image is changed according to the content of the depth information.
Medical imaging equipment. A medical image generating apparatus that generates a schema image based on a mammography image and information on a region of interest,
an image and information acquisition unit that acquires information on the mammography image and the region of interest;
a front information calculation unit that calculates front information representing a range of the attention area in the front direction for the attention area in the breast image included in the mammography image, based on the obtained mammography image and information about the attention area; ,
a depth information calculation unit that calculates depth information representing a position of the region of interest in the depth direction for the region of interest in a breast image included in the mammography image, based on the acquired mammography image and information of the region of interest; ,
a schema image generation unit that generates a schema image by reflecting the front information and the depth information;
a display control unit for displaying the schema image on a display device,
The depth information calculation unit calculates the total breast thickness of the breast image and the attention area depth of the attention area in the breast image, calculates the ratio of the attention area depth to the total breast thickness, and calculates the Use the ratio as depth information
Medical imaging equipment. The depth information calculation unit further calculates a total breast thickness of the breast image and an attention area thickness of the attention area in the breast image, calculates a ratio of the attention area thickness to the total breast thickness, The medical image generating apparatus according to claim 2 , wherein the ratio is depth information. 4. The schema image generator according to any one of claims 1 to 3, wherein the identification unit of the front information in the schema image varies according to the types of parameters constituting the depth information and the magnitude of the parameter values. 1. The medical image generating apparatus according to 1. When the information on the region of interest is information indicating a region having a size less than a predetermined value in the mammography image,
4. The medical image generating apparatus according to any one of claims 1 to 3 , wherein the schema image generation unit makes the form of the identification unit of the front information in the schema image linear. When the information on the region of interest is information indicating a region having a size equal to or larger than a predetermined value in the mammography image,
4. The schema image generation unit varies one or more elements selected from the width, length, or area of the identifying portion of the front information in the schema image according to the content of the depth information. The medical image generating apparatus according to any one of . When there are a plurality of pieces of information on the attention area corresponding to a plurality of attention areas,
The schema image generation unit generates one or more elements selected from the color, density, brightness, transparency, texture, width, length, or area of the identifying portion of the front information in the schema image for each region of interest. The medical image generation apparatus according to any one of claims 1 to 3, which is made different. The schema image generation unit adjusts the form of the identification unit that reflects the front information in the schema image according to the size of the range in which the attention area corresponding to the attention area depth is likely to exist in the front direction. 3. The medical image generating apparatus according to claim 2 , which generates different schema images. 9. The medical image generating apparatus according to claim 8 , wherein the deeper the depth of the region of interest, the larger the size of the range in which the region of interest may exist in the front direction. 3. The schema image generation unit increases any element selected from the width, length, or area of the identifying portion of the front information in the schema image as the depth of the region of interest increases. 10. The medical image generating device according to 8 or 9 . The total breast thickness is the distance from the position of the nipple in the breast image to the intersection of the depth direction axis and the breast baseline in the mammography image through the nipple position,
2. The depth of the attention area is the distance from the intersection of a straight line passing through a reference point in the attention area and parallel to the depth direction axis and a straight line passing through the nipple position and perpendicular to the depth direction axis to the reference point. The medical image generating apparatus according to . The depth of the region of interest is the distance from the intersection of the breast skin line and a straight line passing through a reference point in the region of interest and parallel to the depth direction axis in the mammography image to the reference point,
The medical image generating apparatus according to claim 2 , wherein the total breast thickness is a distance from an intersection point of the straight line and the breast skin line to an intersection point of the straight line and the breast baseline line. The region-of-interest depth is parallel to a straight line passing through a reference point in the region-of-interest and normal to the breast base line through the nipple position, or a straight line passing through the upper and lower ends of the nipple and a straight line passing through the nipple position. A straight line and the distance from the intersection of the breast skin line to the reference point,
The medical image generating apparatus according to claim 2 , wherein the total breast thickness is a distance from an intersection point of the straight line and the breast skin line to an intersection point of the straight line and the breast baseline line. The thickness of the region of interest is the distance between two intersections of a straight line passing through a reference point in the region of interest and parallel to the depth direction axis in the mammography image and the outer edge of the region of interest,
The medical image generating apparatus according to claim 3 , wherein the total breast thickness is the distance from the intersection of the straight line and the breast skin line to the intersection of the straight line and the breast baseline. The thickness of the region of interest is parallel to a straight line passing through a reference point in the region of interest, a normal line to the breast base line through the nipple position, or a straight line passing through the upper and lower ends of the nipple and a straight line passing through the nipple position. is the distance between two intersections of a straight line and the outer edge of the region of interest;
The medical image generating apparatus according to claim 3 , wherein the total breast thickness is the distance from the intersection of the straight line and the breast skin line to the intersection of the straight line and the breast baseline. A medical image generating apparatus that generates a schema image based on a mammography image and information on a region of interest,
an image and information acquisition unit that acquires information on the mammography image and the region of interest;
a front information calculation unit that calculates front information representing a range of the attention area in the front direction for the attention area in the breast image included in the mammography image, based on the obtained mammography image and information about the attention area; ,
a depth information calculation unit that calculates depth information representing a position of the region of interest in the depth direction for the region of interest in a breast image included in the mammography image, based on the acquired mammography image and information of the region of interest; ,
a schema image generation unit that generates a schema image by reflecting the front information and the depth information;
a display control unit for displaying the schema image on a display device,
The display control unit causes a display device to display the mammography image showing the position of the nipple, the breast edge, and the nipple reference line in the breast image included in the mammography image detected from the mammography image,
Further, an operation input unit that receives an operation input from an operator to change any one selected from the position of the nipple, the edge of the breast, and the inclination of the nipple reference line while the mammography image is displayed on the display device. with
The front information calculation unit calculates corrected front information representing the range of the attention area in the front direction based on the mammography image, the attention area information, and the operation input,
The depth information calculation unit calculates corrected depth information representing the position of the attention area in the depth direction based on the mammography image, the attention area information, and the operation input,
The schema image generating unit generates a modified schema image by reflecting the modified front information and the modified depth information,
The display control unit causes a display device to display the modified schema image.
Medical imaging equipment. A medical image generation apparatus for calculating a reference candidate range based on information on a mammography image including an MLO image and a CC image and information on a region of interest,
an image and information acquisition unit that acquires information on the mammography image and the region of interest;
a front information calculation unit that calculates front information representing a range of the attention area in the front direction for the attention area in the breast image included in the mammography image, based on the obtained mammography image and information about the attention area; ,
A reference candidate for calculating a reference candidate range for calculating a reference candidate range for the other image based on the attention area specified in one of the MLO image and the CC image, reflecting the front information. a range calculator;
a display control unit that causes a display device to display the mammography image and the reference candidate range,
The reference candidate range calculation unit sets virtual schema images corresponding to a first breast image included in the one image and a second breast image included in the other image,
Relative positions of the region of interest in the first breast image in a first breast baseline direction of the first breast image are relatively projected onto the second breast image via the virtual schema image. The range in the direction of the second breast base line of the second breast image is set as the reference candidate range.
Medical imaging equipment. The reference candidate range calculation unit calculates two points of intersection where a straight line passing through a position corresponding to the relative position in the virtual schema image intersects an outer peripheral line of the breast in the virtual schema image,
calculating two relative positions of the two intersection points in the second breast baseline direction in the virtual schema image;
18. The medical image generating apparatus according to claim 17 , wherein a range sandwiched between the two relative positions defined in the direction of the second breast base line in the second breast image is calculated as the reference candidate range. Further, depth information calculation for calculating depth information representing a position of the region of interest in the depth direction for the region of interest in a breast image included in the mammography image based on the obtained mammography image and information of the region of interest. having a department,
The reference candidate range calculation unit specifies a depth direction range of the reference candidate range in the second breast image based on depth information representing a position in the depth direction of the attention area in the first breast image. 19. The medical image generating apparatus according to Item 17 or 18 . A medical image generation program for causing a computer to execute a medical image generation process executed by a medical image generation apparatus that generates a schema image based on a mammography image and information on a region of interest in the mammography image, comprising:
Medical image generation processing
Acquiring information on the mammography image and the region of interest;
Based on the obtained mammography image and the information of the attention area, for the attention area in the breast image included in the mammography image, front information representing the range in the front direction of the attention area is calculated;
Based on the acquired mammography image and the information of the attention area, depth information representing the position of the attention area in the depth direction for the attention area in the breast image included in the mammography image is calculated,
generating a schema image reflecting the front information and the depth information;
displaying the schema image on a display device ;
When generating the schema image, a schema image is generated in which the aspect of the identification portion reflecting the front information in the schema image is changed according to the content of the depth information.
A medical image generation program. A medical image generation program for causing a computer to execute a medical image generation process for calculating a reference candidate range based on a mammography image including an MLO image and a CC image and information on a region of interest in the mammography image, comprising:
Medical image generation processing
Acquiring information on the mammography image and the region of interest;
Based on the obtained mammography image and the information of the attention area, for the attention area in the breast image included in the mammography image, front information representing the range in the front direction of the attention area is calculated;
reflecting the front information and calculating a reference candidate range in the other image based on the region of interest designated in either one of the MLO image and the CC image;
displaying the mammography image and the reference candidate range on a display device ;
When calculating the reference candidate range, setting a virtual schema image corresponding to a first breast image included in the one image and a second breast image included in the other image,
Relative positions of the region of interest in the first breast image in a first breast baseline direction of the first breast image are relatively projected onto the second breast image via the virtual schema image. The range in the direction of the second breast base line of the second breast image is set as the reference candidate range.
A medical image generation program.

Images