JP7790195B2 - Medical image processing device, medical image processing system, and medical image processing method - Google Patents

Description

This invention relates to a medical image processing device, a medical image processing system, and a medical image processing method, and in particular to a medical image processing device, a medical image processing system, and a medical image processing method for processing X-ray images showing the vertebral bodies of a subject.

Medical image processing devices that process X-ray images of a subject's vertebral bodies (the cylindrical parts of the vertebrae that make up the spine) are known (see, for example, Patent Document 1).

Patent Document 1 discloses a configuration for detecting vertebral bodies from X-ray images showing the vertebral bodies using a trained model trained by machine learning. The trained model used to detect vertebral bodies in Patent Document 1 detects vertebral bodies by distinguishing between bone (vertebral body) regions and non-bone (non-vertebral body) regions. Patent Document 1 is also configured to calculate the bone density (bone mineral content) of the detected vertebral bodies.

Japanese Patent Application Laid-Open No. 2019-209028

Images of vertebral bodies are used, for example, to diagnose osteoporosis. In osteoporosis diagnosis, bone density is calculated for each vertebra. Furthermore, the average bone density is used to diagnose osteoporosis. If a metallic member or other object is captured near a subject's vertebral body, or if a vertebral body has a fracture, the pixel values attributable to the metallic member and the location of the fracture are higher than the pixel values of the vertebral body, resulting in an inaccurate bone density calculation. Therefore, it is preferable to exclude vertebral bodies with metallic members or fractures from the bone density calculation target. However, with the configuration disclosed in Patent Document 1, bone density is calculated for all vertebral bodies detected by the trained model. Therefore, if a vertebra with a metallic member or a fracture is included, the accuracy of the bone density calculation result decreases because the pixel values of the metallic member or the fracture location are higher than the pixel values of the vertebral body.

This invention was made to solve the above-mentioned problems, and one object of this invention is to provide a medical image processing device, a medical image processing system, and a medical image processing method that can prevent a decrease in the accuracy of bone density calculation results.

In order to achieve the above object, a medical image processing device according to a first aspect of the present invention includes an image acquisition unit that acquires an X-ray image showing a plurality of vertebral bodies, a vertebral body region identification unit that identifies a vertebral body region, which is a region of the plurality of vertebral bodies, from the X-ray image, an input acceptance unit that accepts a selection operation by an operator , an exclusion region determination unit that determines, for each vertebral body region, an exclusion region that is a region to be excluded from bone density calculation targets, based on an exclusion selection operation input by the input acceptance unit, which is an exclusion selection operation for selecting one region to be excluded from the vertebral body region, and a bone density calculation unit that calculates bone density using a calculation target region that is a vertebral body region other than the exclusion region determined by the exclusion region determination unit.

In addition, in order to achieve the above object, a medical image processing system according to a second aspect of the present invention includes an X-ray source, a detector that detects X-rays irradiated from the X-ray source, and an image processing device that generates an X-ray image based on the detection result of the detector, and the image processing device includes an image acquisition unit that acquires the X-ray image, a vertebral body region identification unit that identifies a vertebral body region, which is a region of a plurality of vertebrae, from the X-ray image, an input receiving unit that receives a selection operation by an operator, an exclusion region determination unit that determines, for each vertebral body region, an exclusion region that is a region to be excluded from bone density calculation targets, based on an exclusion selection operation input by the input receiving unit, which is an exclusion selection operation that selects one region to be excluded from the vertebral body region, and a bone density calculation unit that calculates bone density using a calculation target region that is a vertebral body region other than the exclusion region determined by the exclusion region determination unit.

In order to achieve the above object, a medical image processing method according to a third aspect of the present invention includes the steps of acquiring an X-ray image showing a plurality of vertebral bodies, identifying a vertebral body region, which is a region of the plurality of vertebral bodies, from the X-ray image, determining an exclusion region, which is a region to be excluded from bone density calculation targets, from the vertebral body region for each region based on an exclusion selection operation, which is a selection operation by an operator, for selecting one region to be excluded from the vertebral body region, and calculating bone density using the calculation target region, which is a vertebral body region other than the determined exclusion region.

The medical image processing device according to the first aspect includes an exclusion region determination unit that determines, for each region, an exclusion region to be excluded from bone density calculation targets based on an exclusion selection operation input by the input receiving unit, which selects one region to be excluded from the vertebral body region , and a bone density calculation unit that calculates bone density using a calculation target region that is a vertebral body region other than the exclusion region determined by the exclusion region determination unit. Thus, the exclusion region determination unit determines the exclusion region to be excluded from the bone density calculation target region based on the operator's exclusion selection operation. For example, the operator can select a vertebral body region near which a metal member is captured and a vertebral body region where a fracture has occurred, thereby excluding the vertebral body near which the metal member is captured and the vertebral body where the fracture has occurred from the bone density calculation target region. As a result, bone density is calculated using the calculation target region that excludes the vertebral body near which the metal member is captured and the vertebral body where the fracture has occurred, thereby preventing a decrease in the accuracy of the bone density calculation result. Furthermore, as described above, the present invention includes an exclusion region determination unit that determines, for each vertebral region, an exclusion region that is a region to be excluded from the bone mineral density calculation target, based on the exclusion selection operation input by the input receiving unit. As a result, since the exclusion region determination unit determines the exclusion region for each region based on the exclusion selection operation by the operator, it is possible to easily exclude the exclusion region from the calculation target region with a single exclusion selection operation, unlike a configuration in which the exclusion region is determined by gradually deleting regions from the vertebral region using, for example, a drag operation by the operator.

In addition, the medical image processing system according to the second aspect includes an image processing device including: an exclusion region determiner that determines, for each vertebral body region, an exclusion region that is to be excluded from bone density calculation targets, based on an exclusion selection operation input by the input receiver, which selects one region to be excluded from the vertebral body region , and a bone density calculator that calculates bone density using a calculation target region that is a vertebral body region other than the exclusion region determined by the exclusion region determiner. This makes it possible to provide a medical image processing system that can prevent a decrease in the accuracy of bone density calculation results, similar to the medical image processing device according to the first aspect.

The medical image processing method according to the third aspect includes the steps of: determining, for each vertebral region, an exclusion region to be excluded from bone mineral density calculation targets, based on an exclusion selection operation performed by an operator to select one region to be excluded from the vertebral region; and calculating bone mineral density using the calculation target region, which is the vertebral region other than the determined exclusion region. This makes it possible to provide a medical image processing method capable of suppressing a decrease in the accuracy of bone mineral density calculation results, similar to the medical image processing device according to the first aspect.

1 is a schematic diagram showing the overall configuration of a medical image processing system according to an embodiment. 1 is a schematic diagram for explaining the configuration of an X-ray imaging apparatus according to an embodiment; FIG. 1 is a schematic diagram for explaining an X-ray image showing a vertebral body. FIG. 10 is a schematic diagram for explaining a superimposed image in which a marker and an interbody boundary line are superimposed on a vertebral body region. 1 is a schematic diagram for explaining a configuration in which a medical image processing apparatus according to an embodiment displays a superimposed image on a display unit. 10A and 10B are schematic diagrams illustrating an example of a screen displaying a superimposed image and a past superimposed image. 10A and 10B are schematic diagrams illustrating a state in which an operator has selected an area to be determined as an exclusion area by an exclusion area determination unit according to an embodiment. 10A and 10B are schematic diagrams for explaining a process in which an exclusion region determining unit according to an embodiment determines a vertebral body region selected by an operator as an exclusion region. 5A and 5B are schematic diagrams for explaining calculation results displayed by a calculation result display unit in one embodiment. 1A and 1B are schematic diagrams illustrating a configuration for adjusting the position of one end of an interbody boundary line in a medical image processing apparatus according to an embodiment. 10A and 10B are schematic diagrams illustrating a configuration for adjusting the position of the other end of the interbody boundary line in a medical image processing apparatus according to one embodiment. 10 is a flowchart illustrating a process in which a medical image processing apparatus according to an embodiment determines a calculation target region and calculates bone density. 10 is a flowchart for explaining processing for adjusting the position of an interbody boundary line performed by a medical image processing apparatus according to an embodiment.

The following describes an embodiment of the present invention with reference to the drawings.

The configuration of a medical image processing system 300 according to one embodiment will be described with reference to Figures 1 and 2. As shown in Figure 1, the medical image processing system 300 includes a medical image processing device 100 and an X-ray image capturing device 200. The medical image processing device 100 is located in the same room (examination room or operating room) as the X-ray image capturing device 200, and is connected to the X-ray image capturing device 200 via a wired or wireless connection. The medical image processing device 100 is an imaging device that calculates the bone density of a subject 80 (see Figure 2) and can be used to diagnose osteoporosis. The medical image processing device 100 is an example of an "image processing device" as defined in the claims.

(Configuration of medical image processing device)
As shown in FIG. 1, the medical image processing apparatus 100 includes an image acquisition unit 1, an image processing unit 2, an input reception unit 3, a display unit 4, and a storage unit 5.

The image acquisition unit 1 is configured to acquire an X-ray image 10 that shows multiple vertebral bodies 40 (see Figure 3). In this embodiment, the image acquisition unit 1 is configured to acquire the X-ray image 10, for example, from an X-ray imaging device 200. The image acquisition unit 1 includes, for example, an input/output interface.

The image processing unit 2 is configured to calculate the bone density of multiple vertebral bodies 40 (see Figure 3) in the X-ray image 10. The image processing unit 2 is also configured to generate a superimposed image 11, which will be described later. The image processing unit 2 is a computer, processor, or circuitry that includes a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and GPU (Graphics Processing Unit). The image processing unit 2, which is comprised of hardware such as a CPU, also includes the following software (program) functional blocks: a vertebral body region identification unit 20, an exclusion region determination unit 21, a bone density calculation unit 22, a boundary line display unit 23, a superimposed image generation unit 24, an image display control unit 25, and a calculation result output unit 26. By executing the programs stored in the memory unit 5, the image processing unit 2 functions as a vertebral body region identification unit 20, an exclusion region determination unit 21, a bone density calculation unit 22, a boundary line display unit 23, a superimposed image generation unit 24, an image display control unit 25, and a calculation result output unit 26. Details of each functional block of the image processing unit 2 will be described later.

The input acceptance unit 3 is configured to accept operations by the operator. In this embodiment, the input acceptance unit 3 is configured to accept an operation 30 by the operator to select a vertebral body region 41 (see FIG. 3) and an operation (adjustment operation 31) to adjust the position of the interbody boundary line 13 (see FIG. 4), which will be described later. The input acceptance unit 3 includes input devices such as a mouse, keyboard, and touch-panel LCD display.

The display unit 4 is configured to display an X-ray image 10. In this embodiment, the display unit 4 is configured to display a superimposed image 11. In this embodiment, the display unit 4 also displays a past superimposed image 11a stored in the memory unit 5 together with the superimposed image 11. In this embodiment, the display unit 4 is also configured to display a calculation result 32 calculated by the bone density calculation unit 22. Details of the superimposed image 11, the past superimposed image 11a, and the calculation result 32 will be described later. The display unit 4 includes a display device such as a liquid crystal monitor or an organic EL (Electro-Luminescence) monitor.

The memory unit 5 is configured to store the superimposed image 11. The memory unit 5 is also configured to store a past superimposed image 11a. The memory unit 5 also stores a trained model 33 that the vertebral body region identification unit 20 uses to identify the vertebral body 40 (see Figure 3) that appears in the X-ray image 10. The memory unit 5 also stores a program executed by the image processing unit 2. The memory unit 5 includes, for example, non-volatile memory such as a hard disk drive (HDD) or a solid state drive (SSD).

The trained model 33 is created by individually estimating vertebral bodies 40 (see FIG. 3) shown in the X-ray image 10 and learning to estimate interbody boundaries 13 (see FIG. 4), which are the boundaries between each individually estimated vertebral body 40. The trained model 33 is trained using the X-ray image 10 showing the vertebral bodies 40 and an image in which the vertebral body region 41 and interbody boundary line 13 are superimposed on the X-ray image 10 as training data. When the X-ray image 10 is input, the trained model 33 outputs information (position information) about the vertebral body region 41 (see FIG. 4) and information (position information) about the interbody boundary line 13. The trained model 33 is created, for example, by training a convolutional neural network (CNN) or a learning model that partially includes a convolutional neural network.

(Configuration of X-ray imaging device)
As shown in FIG. 2 , the X-ray imaging device 200 includes an X-ray source 50, an X-ray detection unit 51, an imaging device control unit 52, and an imaging device image processing unit 53. The imaging device control unit 52 is electrically connected to the X-ray source 50 and the imaging device image processing unit 53. The X-ray detection unit 51 is electrically connected to the imaging device image processing unit 53. The X-ray imaging device 200 captures an image of a subject 80 to generate an X-ray image 10 showing multiple vertebral bodies 40 (see FIG. 3 ). The X-ray imaging device 200 also sends the generated X-ray image 10 to the medical image processing device 100. In the example shown in FIG. 2 , electrical connections are indicated by dashed lines, and input and output of information are indicated by solid arrows. The X-ray detection unit 51 is an example of a "detector" as defined in the claims.

The X-ray source 50 generates X-rays when a high voltage is applied. The X-rays generated by the X-ray source 50 are configured to be emitted in the direction in which the X-ray detection unit 51 is located. The X-ray source 50 is, for example, an X-ray generating device including an X-ray tube (not shown) and a voltage supply unit (not shown).

The X-ray detection unit 51 detects X-rays emitted from the X-ray source 50 and converts the detected X-rays into an electrical signal. The X-ray detection unit 51 is, for example, a flat panel detector (FPD). The detection signal (image signal) from the X-ray detection unit 51 is sent to the imaging device image processing unit 53.

The imaging device control unit 52 is configured to control the X-ray image imaging device 200. The imaging device control unit 52 is, for example, a computer, processor, or circuitry configured to include a CPU, ROM, RAM, and GPU.

The imaging device image processing unit 53 is configured to generate an X-ray image 10 based on the detection signal sent from the X-ray detection unit 51. The imaging device image processing unit 53 includes, for example, a processor or circuitry such as a GPU or an FPGA (Field-Programmable Gate Array) configured for image processing.

The X-ray image 10 generated by the imaging device image processing unit 53 is sent to the medical image processing device 100.

(X-ray images and superimposed images)
Next, the X-ray image 10 and the superimposed image 11 will be described with reference to FIGS.

As shown in FIG. 3, the X-ray image 10 is an image showing the vertebral bodies 40 of the subject 80 (see FIG. 2). In the example shown in FIG. 3, the first vertebral body 40a, the second vertebral body 40b, the third vertebral body 40c, and the fourth vertebral body 40d are shown. The trained model 33 (see FIG. 1) distinguishes between vertebral body regions 41 and regions other than the vertebral bodies 40 by individually estimating the vertebral bodies 40 in the X-ray image 10. Specifically, the trained model 33 is configured to distinguish one vertebral body region 41 from multiple vertebral bodies 40. For convenience, in the example shown in FIG. 3, the vertebral body region 41 is illustrated with a dashed line. In this embodiment, the X-ray image 10 is an image showing the lumbar vertebral bodies 40 (lumbar vertebrae) of the subject 80.

Next, the superimposed image 11 will be described with reference to FIG. 4. The superimposed image 11 is generated by the superimposed image generating unit 24 (see FIG. 1). The superimposed image 11 is an image in which markers 12 that identify vertebral body regions 41 are superimposed on the X-ray image 10 (see FIG. 3). In this embodiment, in the superimposed image 11, one marker 12 is superimposed on multiple vertebral bodies 40 (see FIG. 3). As shown in FIG. 4, the superimposed image 11 also displays intervertebral body boundary lines 13. As shown in FIG. 4, the intervertebral body boundary lines 13 are displayed at positions between adjacent vertebral bodies 40 (see FIG. 3). In other words, multiple intervertebral body boundary lines 13 are displayed in the superimposed image 11. Rectangular markers 14a and 14b are displayed at one end 13a and the other end 13b of the intervertebral body boundary line 13, respectively. The markers 14a and 14b are operated when adjusting the position of the intervertebral body boundary line 13. Adjusting the position of the interbody boundary line 13 will be described later.

The superimposed image 11 also displays the soft tissue region 15 so that it can be distinguished from the vertebral body region 41. The soft tissue region 15 is used when calculating bone density. In the example shown in Figure 4, the soft tissue region 15 is hatched differently from the vertebral body region 41, thereby showing that the soft tissue region 15 and the vertebral body region 41 can be distinguished.

(Display of superimposed images)
Next, a configuration in which the image processing unit 2 generates the superimposed image 11 and displays it on the display unit 4 will be described with reference to FIG.

In this embodiment, the vertebral body region identification unit 20 is configured to identify a vertebral body region 41 (see FIG. 4), which is a region of multiple vertebral bodies 40 (see FIG. 3), from an X-ray image 10. The vertebral body region identification unit 20 individually identifies the regions of multiple vertebral bodies 40 and identifies the vertebral body region 41 by connecting the individually identified vertebral body 40 regions. Specifically, the vertebral body region identification unit 20 acquires an X-ray image 10 from the image acquisition unit 1. The vertebral body region identification unit 20 also reads a trained model 33 from the memory unit 5. The vertebral body region identification unit 20 inputs the acquired X-ray image 10 into the trained model 33 to acquire vertebral body region information 20a. The vertebral body region information 20a is coordinate information of the vertebral body region 41 (see FIG. 3) in the X-ray image 10. The vertebral body region identification unit 20 also outputs the acquired vertebral body region information 20a to the superimposed image generation unit 24.

The vertebral body region identification unit 20 is also configured to acquire intervertebral body boundary lines 13 (see Figure 4), which are the boundaries between adjacent vertebral bodies 40 in the vertebral body region 41. Specifically, the vertebral body region identification unit 20 acquires intervertebral body boundary line information 20b by inputting the X-ray image 10 to the trained model 33. The intervertebral body boundary line information 20b is coordinate information for each intervertebral body boundary line 13. The vertebral body region identification unit 20 also outputs the acquired intervertebral body boundary line information 20b to the boundary line display unit 23.

The superimposed image generation unit 24 is configured to generate a superimposed image 11 in which a marker 12 (see FIG. 3) that can identify a vertebral body region 41 (see FIG. 3) is superimposed on the X-ray image 10. Specifically, the superimposed image generation unit 24 acquires vertebral body region information 20a input from the vertebral body region identification unit 20. The superimposed image generation unit 24 superimposes the marker 12 on the X-ray image 10 based on the acquired vertebral body region information 20a. More specifically, the superimposed image generation unit 24 superimposes the marker 12 on a pixel region corresponding to the vertebral body region 41 based on the vertebral body region coordinate information, which is the vertebral body region information 20a. Note that the vertebral body region information 20a is information on the coordinate range representing the vertebral body region 41. The superimposed image generation unit 24 outputs the generated superimposed image 11 to the boundary line display unit 23.

The boundary display unit 23 is configured to display the intervertebral body boundary line 13 in the vertebral body region 41. The boundary display unit 23 acquires intervertebral body boundary line information 20b from the vertebral body region identification unit 20. The boundary display unit 23 also acquires the superimposed image 11 from the superimposed image generation unit 24. The boundary display unit 23 displays the intervertebral body boundary line 13 in the superimposed image 11 based on the intervertebral body boundary line information 20b. Specifically, the boundary display unit 23 displays (superimposes) the intervertebral body boundary line 13 on the superimposed image 11 based on the coordinate information of the intervertebral body boundary line 13, which is the intervertebral body boundary line information 20b. The boundary display unit 23 outputs the superimposed image 11 displaying the intervertebral body boundary line 13 to the image display control unit 25.

The image display control unit 25 is configured to control the display of the superimposed image 11 on the display unit 4. Specifically, the image display control unit 25 acquires the superimposed image 11 from the boundary line display unit 23. The image display control unit 25 also outputs the acquired superimposed image 11 to the display unit 4.

In addition, in this embodiment, the image display control unit 25 is configured to control the display unit 4 to display a past superimposed image 11a stored in the memory unit 5 together with the superimposed image 11. Specifically, the image display control unit 25 acquires the past superimposed image 11a by reading it from the memory unit 5. The image display control unit 25 also outputs the acquired past superimposed image 11a to the display unit 4 together with the superimposed image 11.

The display unit 4 displays the superimposed image 11 input from the image display control unit 25 and the past superimposed image 11a.

Figure 6 is a schematic diagram of the first display screen 16 on which the display unit 4 (see Figure 5) displays the superimposed image 11.

The first display screen 16 displays a first button 16a for changing to an exclusion area determination mode in which an exclusion area 41a (see Figure 8) to be excluded from bone density calculation is determined, a second button 16b for changing to an interbody boundary line 13 adjustment operation mode, an OK button 16c, and a cancel button 16d.

The processing that occurs when the first button 16a is pressed and the processing that occurs when the second button 16b is pressed will be described later.

When the OK button 16c is pressed, the calculation result output unit 26 (see Figure 1) performs a bone density calculation process, and then the first display screen 16 is closed. When the Cancel button 16d is pressed, the bone density calculation process is not performed and the first display screen 16 is closed.

As shown in Figure 6, the display unit 4 (see Figure 5) displays the superimposed image 11 and the previous superimposed image 11a side by side. The first display screen 16 also displays a subtraction image 11b. The subtraction image 11b is an image obtained by subtracting two X-ray images 10 (see Figure 2) taken by changing the tube voltage applied to the X-ray source 50 (see Figure 2). The subtraction image 11b is an image in which soft tissue has been removed by subtracting the X-ray images 10 taken by changing the tube voltage.

(Determination of Exclusion Areas)
Here, if the vertebral body region 41 includes a vertebra (not shown) in which a metal member of the vertebral body 40 (see FIG. 3) is located nearby, or a vertebra (not shown) in the vertebral body 40 where a fracture has occurred, the pixel value of the metal member or the pixel value of the fractured portion may be higher than the pixel value of the vertebral body 40, resulting in an inaccurate bone density calculation result 32 (see FIG. 1). Therefore, in this embodiment, the exclusion region determiner 21 (see FIG. 1) is configured to determine a region to be excluded from the bone density calculation from the vertebral body region 41 based on a selection operation 30 (see FIG. 1) performed by the operator.

As shown in FIG. 7, the exclusion region determination unit 21 (see FIG. 1) is configured to select a region that is a candidate for the exclusion region 41a (see FIG. 8) based on the selection operation 30 (see FIG. 1). Note that the operator presses the first button 16a (see FIG. 6) before performing the selection operation 30. That is, when the exclusion region determination unit 21 detects the operator's operation input of the first button 16a, it transitions to an operation mode for determining the exclusion region 41a. Furthermore, the example shown in FIG. 7 illustrates a case where the operator selects the third vertebral body 40c as a candidate for the region to be excluded from the calculation target region 41b.

The input receiving unit 3 is configured to receive an operation input for selecting a marker 12 in the superimposed image 11 as a selection operation 30. The cursor 60 shown in FIG. 7 indicates that the third vertebral body 40c is selected by the selection operation 30 received by the input receiving unit 3. The exclusion region determination unit 21 is configured to determine the calculation target region 41b based on the operation input for selecting the marker 12. Specifically, the exclusion region determination unit 21 is configured to exclude from bone density calculation the region sandwiched between the intervertebral body boundary lines 13 on both the upper and lower sides of the exclusion region 41a selected based on the selection operation 30.

The exclusion region determination unit 21 (see FIG. 1) acquires position information of the selected position 61 on the superimposed image 11 based on the operation input for selecting the marker 12. The exclusion region determination unit 21 also acquires the nearest boundary line 13c, which is the intervertebral body boundary line 13 closest to the selected position 61. Specifically, the exclusion region determination unit 21 acquires coordinate information of the selected position 61 (the position coordinates of the selected position 61). Then, as indicated by arrows 62 and 63, the exclusion region determination unit 21 acquires a distance W1 from the selected position 61 to the nearest intervertebral body boundary line 13 below the selected position 61, and a distance W2 from the selected position 61 to the nearest intervertebral body boundary line 13 above the selected position 61. The exclusion region determination unit 21 then compares the distances W1 and W2 and determines the intervertebral body boundary line 13 with the smaller distance as the nearest boundary line 13c. Additionally, for the selected position 61, the opposite side boundary line 13d, which is the interbody boundary line 13 on the opposite side of the nearest boundary line 13c, is obtained.

8, the exclusion region determination unit 21 (see FIG. 1) is configured to determine an exclusion region 41a based on a selection operation 30 (see FIG. 1). In this embodiment, the exclusion region determination unit 21 is configured to determine, for each vertebral region 41, an exclusion region 41a to be excluded from bone mineral density calculation targets, based on the selection operation 30 input by the input receiving unit 3 (see FIG. 1). Specifically, the exclusion region determination unit 21 (see FIG. 1) is configured to exclude the vertebral region 41 between the nearest boundary line 13c and the opposite boundary line 13d from the calculation target region 41b. The superimposed image generation unit 24 (see FIG. 1) is configured to differentiate the display mode of the calculation target region 41b and the region other than the calculation target region 41b (exclusion region 41a) within the vertebral region 41. Specifically, in the superimposed image 11c after the exclusion region 41a has been determined, the superimposed image generation unit 24 superimposes markers 12 on the calculation target region 41b, but does not superimpose markers 12 on the region other than the calculation target region 41b (exclusion region 41a). In the example shown in FIG. 8, the third vertebral body 40c has been determined as the exclusion region 41a, and therefore the region corresponding to the third vertebral body 40c is displayed in a different manner from the vertebral body region 41 above the third vertebral body 40c (calculation target region 41b) and the vertebral body region 41 below the third vertebral body 40c (calculation target region 41b). Note that in the example shown in FIG. 8, the presence or absence of hatching represents the difference in display manner between the calculation target region 41b and the region other than the calculation target region 41b (exclusion region 41a).

The bone density calculation unit 22 (see FIG. 1) is configured to calculate bone density using a calculation target region 41b, which is the vertebral body region 41 other than the exclusion region 41a determined by the exclusion region determination unit 21. In this embodiment, the bone density calculation unit 22 is configured to calculate BMD (Bone Mineral Density) as the bone density of the vertebral body region 41. Note that the vertebral body 40 (see FIG. 3) shown in the X-ray image 10 (see FIG. 3) is imaged based on X-rays that have passed through the vertebral body 40 and soft tissues that overlap with the vertebral body 40 (see FIG. 3) in the X-ray irradiation axis direction. Therefore, the pixel value of the vertebral body region 41 is a value based on the intensity of the X-rays that have passed through the vertebral body 40 and soft tissues. Therefore, in this embodiment, in order to remove the effect of X-ray attenuation by soft tissue from the pixel values of the vertebral body region 41, the bone density calculation unit 22 calculates bone density based on the pixel values of the calculation target region 41b and the pixel values of the soft tissue region 15. Note that in this embodiment, the bone density calculation unit 22 acquires the bone mineral density (BMD) of each vertebral body 40 using the DXA (Dual Energy X-ray Absorptometry) method, which calculates bone density using the difference image 11b (see FIG. 6). Note that the method by which the bone density calculation unit 22 calculates bone density is not limited.

(Display of calculation results)
Next, a configuration for displaying the calculation results 32 (see FIG. 1) will be described with reference to FIG. 9. In this embodiment, the calculation result output unit 26 (see FIG. 1) is configured to output the calculation results 32 obtained by the bone mineral density calculation unit 22 (see FIG. 1). The second display screen 17 shown in FIG. 9 is an example of a screen for displaying the calculation results 32.

As shown in Figure 9, the bone mineral density calculation unit 22 outputs a first calculation result 34 and a second calculation result 35 as calculation results 32 to the display unit 4 (see Figure 1).

The first calculation result 34 is the calculation result of bone density for each vertebral body 40. The second calculation result 35 is the average value of the calculation results 32 for each vertebral body 40.

The first calculation result 34 includes a measurement date 34a, an age 34b, a location 34c, and a bone density 34d. In the example shown in Figure 9, the location 34c displays the bone density 34d of the vertebral bodies 40 of L1 (first vertebral body 40a), L2 (second vertebral body 40b), and L4 (fourth vertebral body 40d). Note that the example shown in Figure 9 is a calculation result 32 (see Figure 1) after excluding L3 (third vertebral body 40c) as the excluded region 41a (see Figure 8), and therefore the bone density value is not displayed in the L3 bone density 34d column.

The second calculation result 35 also includes a measurement date 35a, an age 35b, a location 35c, a bone density 35d, a same-age comparison 35e, and a younger-age comparison 35f. The bone density 35d is the average value of the bone density values at the location 35c. In this embodiment, since L3 (the third vertebral body 40c) is the excluded region 41a (see Figure 8), the bone density 35d is the average value of the bone density values of the vertebral bodies 40 at L1 (the first vertebral body 40a), L2 (the second vertebral body 40b), and L4 (the fourth vertebral body 40d).

Same-age comparison 35e is the ratio of the bone density 35d value of subject 80 to the average bone density value of people of the same age as subject 80 (see Figure 2). Younger-age comparison 35f is the ratio of bone density 35d of subject 80 to the average bone density value of people of a younger age (for the lumbar spine, 20 to 44 years old).

In addition, the second display screen 17 displays the superimposed image 11 together with the first calculation result 34 and the second calculation result 35.

(Adjustment of interbody boundaries)
Next, a configuration in which the image processing unit 2 (see FIG. 1 ) according to this embodiment adjusts the position of the intervertebral body boundary line 13 will be described. The vertebral body region identification unit 20 (see FIG. 5 ) acquires the position of the intervertebral body boundary line 13 using a trained model 33 (see FIG. 5 ). At this time, the position of the intervertebral body boundary line 13 displayed in the superimposed image 11 may not be located between the actual vertebral bodies 40. Furthermore, even if the position of the intervertebral body boundary line 13 displayed in the superimposed image 11 is located between the actual vertebral bodies 40, the operator may want to adjust the position of the intervertebral body boundary line 13.

For this reason, in this embodiment, the image processing unit 2 (see FIG. 1) is configured to be able to adjust the position of the intervertebral boundary line 13. Specifically, the input receiving unit 3 (see FIG. 1) is configured to receive an operation to adjust the position of the intervertebral boundary line 13 (adjustment operation 31 (see FIG. 1)) as an input operation by the operator.

Figures 10(A) and 10(B) are schematic diagrams illustrating adjusting the position of one end 13a of the interbody boundary line 13. The example shown in Figures 10(A) and 10(B) is a configuration for adjusting the position of one end 13a of the interbody boundary line 13 between the first vertebral body 40a and the second vertebral body 40b.

As shown in FIG. 10(A), the input receiving unit 3 (see FIG. 1) receives the selection operation of the marker 14a using the cursor 60 as the adjustment operation 31 (see FIG. 1). Then, as shown in the superimposed image 11d shown in FIG. 10(B), the input receiving unit 3 receives the drag-and-drop operation on the marker 14a as the adjustment operation 31. Note that the cursor 60a shown by the dashed line in FIG. 10(B) is the cursor 60 before movement shown in FIG. 10(A). In the example shown in FIG. 10(B), the cursor 60 is dragged and dropped from the position of the cursor 60a to the position of the cursor 60. This adjusts the position of the end 13a on one side of the interbody boundary line 13.

Figures 11(A) and 11(B) are schematic diagrams illustrating adjusting the position of the other end 13b of the interbody boundary line 13. The example shown in Figures 10(A) and 10(B) is a configuration for adjusting the position of the other end 13b of the interbody boundary line 13 between the first vertebral body 40a and the second vertebral body 40b.

As shown in FIG. 11(A), the input receiving unit 3 (see FIG. 1) receives the selection operation of the marker 14b using the cursor 60 as the adjustment operation 31 (see FIG. 1). Then, as shown in the superimposed image 11e shown in FIG. 11(B), the input receiving unit 3 receives the drag-and-drop operation on the marker 14b as the adjustment operation 31. Note that the cursor 60b shown by the dashed line in FIG. 11(B) is the cursor 60 before movement shown in FIG. 11(B). In the example shown in FIG. 11(B), the cursor 60 is dragged and dropped from the position of the cursor 60b to the position of the cursor 60. This adjusts the position of the other end 13b of the interbody boundary line 13.

In this embodiment, the exclusion area determination unit 21 (see Figure 1) is configured to exclude from the bone density calculation the area (vertebral body area 41) sandwiched between the intervertebral body boundary lines 13 after their positions have been adjusted based on the operation to adjust the intervertebral body boundary lines 13, as shown in Figures 10 and 11.

(Determination of calculation target area and calculation of bone density)
Next, with reference to FIG. 12, a process in which the image processing unit 2 (see FIG. 5) according to this embodiment determines the calculation target region 41b (see FIG. 8) and calculates bone density will be described.

In step 101, the image acquisition unit 1 (see Figure 5) acquires an X-ray image 10 (see Figure 5) showing multiple vertebral bodies 40 (see Figure 3).

In step 102, the vertebral body region identification unit 20 (see Figure 5) identifies a vertebral body region 41 (see Figure 3), which is a region of multiple vertebral bodies 40, from the X-ray image 10.

In step 103, the superimposed image generation unit 24 (see Figure 5) generates the superimposed image 11 (see Figure 5). In the processing of step 103, the boundary line display unit 23 (see Figure 5) displays the intervertebral boundary line 13 (see Figure 4) in the superimposed image 11. In addition, the image display control unit 25 (see Figure 5) displays the superimposed image 11 generated by the superimposed image generation unit 24 on the display unit 4 (see Figure 5).

In step 104, the image processing unit 2 determines whether or not an operation input has been made to determine the exclusion area 41a (see Figure 8). If an operation input has been made to determine the exclusion area 41a, the process proceeds to step 105. If an operation input has not been made to determine the exclusion area 41a, the process proceeds to step 107. The operation input to determine the exclusion area 41a is the process of pressing the first button 16a (see Figure 6).

In step 105, the exclusion area determination unit 21 (see Figure 1) determines whether or not the operator has input a selection operation 30 (see Figure 1). If a selection operation 30 has been input, the process proceeds to step 106. If a selection operation 30 has not been input, the process of step 105 is repeated.

In step 106, the exclusion region determination unit 21 determines, for each vertebral body region 41 (see FIG. 7), an exclusion region 41a (see FIG. 8) that is to be excluded from bone density calculation targets, based on the operator's selection operation 30. Specifically, the exclusion region determination unit 21 acquires position information for the selected position 61 (see FIG. 7). The exclusion region determination unit 21 also acquires the nearest boundary line 13c (see FIG. 7) and the opposite-side boundary line 13d (see FIG. 7) of the selected position 61. The exclusion region determination unit 21 then determines the vertebral body region 41 between the nearest boundary line 13c and the opposite-side boundary line 13d as the exclusion region 41a.

In step 107, the bone density calculation unit 22 (see FIG. 1) determines whether or not there has been an operational input to calculate bone density. If there has been an operational input to calculate bone density, the process proceeds to step 108. If there has not been an operational input to calculate bone density, the process proceeds to step 109. Note that the operational input to calculate bone density is, for example, an operational input to press the OK button 16c (see FIG. 7).

In step 108, the bone density calculation unit 22 calculates bone density using the calculation target region 41b (see Figure 8), which is the vertebral body region 41 other than the determined exclusion region 41a. The bone density calculation unit 22 outputs the calculated calculation result 32 (see Figure 1) to the display unit 4. Thereafter, the processing ends.

If the process proceeds from step 107 to step 109, the image processing unit 2 determines in step 109 whether or not there has been an operation input to close the first display screen 16 (see FIG. 6). If there has been an operation input to close the first display screen 16, the process ends without calculating bone density. If there has not been an operation input to close the first display screen 16, the process proceeds to step 105. Note that an operation input to close the first display screen 16 is, for example, an operation input to press the second button 16b (see FIG. 6).

(Adjusting the position of the interbody boundary line)
Next, a process of adjusting the position of the interbody boundary line 13 (see FIG. 10A) performed by the image processing unit 2 (see FIG. 1) according to this embodiment will be described with reference to Fig. 13. The process of adjusting the position of the interbody boundary line 13, which will be described with reference to Fig. 13, is started when the second button 16b (see Fig. 6) is operated (pressed) on the first display screen 16 (see Fig. 6).

In step 201, the image processing unit 2 (see Figure 1) determines whether an operation to adjust the position of the interbody boundary line 13 (see Figure 19(A)) (adjustment operation 31 (see Figure 1)) has been input. If adjustment operation 31 has been input, the process proceeds to step 202. If adjustment operation 31 has not been input, the process proceeds to step 204.

In step 202, the image processing unit 2 (see FIG. 1) adjusts the position of the interbody boundary line 13. Specifically, as shown in FIGS. 10 and 11, the image processing unit 2 (see FIG. 1) adjusts the position of the interbody boundary line 13 (see FIG. 10(A)) based on operational inputs that move the marker 14a (see FIG. 10(A)) and the marker 14b (see FIG. 11(A)).

In step 203, the image processing unit 2 displays the superimposed image 11e (see Figure 11(B)) after the position of the interbody boundary line 13 has been adjusted on the display unit 4 (see Figure 1).

In step 204, the image processing unit 2 determines whether or not an operation input has been made to end the adjustment of the position of the interbody boundary line 13. An operation input to end the adjustment of the position of the interbody boundary line 13 is, for example, a selection operation 30 (see Figure 1) of the marker 12 shown in the superimposed image 11 (see Figure 7) or an operation to press the OK button 16c (see Figure 7). If an operation input to end the adjustment of the position of the interbody boundary line 13 has been made, the processing ends. If an operation input to end the adjustment of the position of the interbody boundary line 13 has not been made, the processing proceeds to step 201.

(Effects of this embodiment)
In this embodiment, the following effects can be obtained.

In this embodiment, as described above, the medical image processing device 100 includes an image acquisition unit 1 that acquires an X-ray image 10 showing multiple vertebral bodies 40; a vertebral body region identification unit 20 that identifies a vertebral body region 41, which is a region of the multiple vertebral bodies 40, from the X-ray image 10; an input reception unit 3 that receives a selection operation 30 from an operator; an exclusion region determination unit 21 that determines, for each vertebral body region 41, an exclusion region 41a that is a region to be excluded from bone density calculation targets based on the selection operation 30 input by the input reception unit 3; and a bone density calculation unit 22 that calculates bone density using a calculation target region 41b, which is a vertebral body region 41 other than the exclusion region 41a determined by the exclusion region determination unit 21.

As a result, the exclusion region determiner 21 determines the exclusion region 41a, which is a region to be excluded from the bone density calculation target region 41b, in response to the operator's selection operation 30. For example, by selecting a region of a vertebral body 40 in which a metal member is nearby and a region of a vertebral body 40 in which a fracture has occurred, the operator can exclude the vertebral body 40 in which a metal member is nearby and the vertebral body 40 in which a fracture has occurred from the bone density calculation target region. As a result, bone density is calculated using the calculation target region 41b that excludes the vertebral body 40 in which a metal member is nearby and the vertebral body 40 in which a fracture has occurred, thereby preventing a decrease in the accuracy of the bone density calculation result 32. Furthermore, because the exclusion region determiner 21 determines the exclusion region 41a region by region in response to the operator's selection operation 30, the exclusion region 41a can be easily excluded from the calculation target region 41b with a single selection operation 30, unlike a configuration in which the exclusion region 41a is determined by gradually removing regions from the vertebral body region 41 using, for example, a drag operation by the operator.

In addition, in this embodiment, as described above, the medical image processing system 300 comprises an X-ray source 50, a detector (X-ray detection unit 51) that detects X-rays irradiated from the X-ray source 50, and an image processing device (medical image processing device 100) that generates an X-ray image 10 based on the detection results of the detector (X-ray detection unit 51).The image processing device (medical image processing device 100) comprises an image acquisition unit 1 that acquires the X-ray image 10, a vertebral body region identification unit 20 that identifies a vertebral body region 41, which is a region of multiple vertebral bodies 40, from the X-ray image 10, an input receiving unit 3 that receives a selection operation 30 by the operator, an exclusion region determination unit 21 that determines, for each region, an exclusion region 41a, which is a region to be excluded from the bone density calculation target, from the vertebral body region 41 based on the selection operation 30 input by the input receiving unit 3, and a bone density calculation unit 22 that calculates bone density using a calculation target region 41b, which is a vertebral body region 41 other than the exclusion region 41a determined by the exclusion region determination unit 21. This makes it possible to provide a medical image processing system 300 that, like the medical image processing device 100, can prevent a decrease in the accuracy of the bone density calculation results 32.

Furthermore, in this embodiment, as described above, the medical image processing method includes the steps of acquiring an X-ray image 10 showing multiple vertebral bodies 40, identifying vertebral body regions 41, which are regions of the multiple vertebral bodies 40, from the X-ray image 10, determining, for each vertebral body region 41, an exclusion region 41a, which is a region to be excluded from bone density calculation targets, based on a selection operation 30 by the operator, and calculating bone density using a calculation target region 41b, which is a vertebral body region 41 other than the determined exclusion region 41a. This makes it possible to provide a medical image processing method that, similar to the medical image processing device 100, can suppress a decrease in the accuracy of the bone density calculation result 32.

In addition, the above embodiment has the following configuration, which provides the following additional advantages:

In other words, in this embodiment, as described above, the vertebral body region identification unit 20 is configured to acquire intervertebral body boundary lines 13, which are the boundaries between adjacent vertebral bodies 40 in the vertebral body region 41, and further includes a boundary line display unit 23 that displays the intervertebral body boundary lines 13 in the vertebral body region 41. The exclusion region determination unit 21 is configured to exclude from bone density calculation the region sandwiched between the intervertebral body boundary lines 13 on both the upper and lower sides of the exclusion region 41a selected based on the selection operation 30. As a result, the exclusion region determination unit 21 determines the exclusion region 41a in the vertebral body region 41 based on the region selected by the selection operation 30 and the intervertebral body boundary lines 13 above and below the selected region, making it possible to easily determine the exclusion region 41a in the vertebral body region 41 that includes multiple vertebral bodies 40, on a vertebral body 40-by-vertebral body 40 basis.

As described above, this embodiment further includes a display unit 4 that displays the X-ray image 10, a superimposed image generation unit 24 that generates a superimposed image 11 in which a marker 12 that can identify the vertebral body region 41 is superimposed on the X-ray image 10, and an image display control unit 25 that controls the display of the superimposed image 11 on the display unit 4. The input reception unit 3 is configured to receive an operation input for selecting the marker 12 in the superimposed image 11 as a selection operation 30, and the exclusion region determination unit 21 is configured to determine the calculation target region 41b based on the operation input for selecting the marker 12. This allows the operator to select the marker 12 while visually viewing the superimposed image 11, thereby making it easy to select the exclusion region 41a.

In addition, in this embodiment, as described above, the exclusion region determiner 21 is configured to acquire position information of the selected position 61 on the superimposed image 11 based on the operation input for selecting the marker 12, and to acquire the nearest boundary line 13c, which is the intervertebral body boundary line 13 closest to the selected position 61, and the opposite-side boundary line 13d, which is the intervertebral body boundary line 13 on the opposite side of the nearest boundary line 13c with respect to the selected position 61, and to exclude the vertebral body region 41 between the nearest boundary line 13c and the opposite-side boundary line 13d from the calculation target region 41b. As a result, the exclusion region determiner 21 acquires the nearest boundary line 13c and the opposite-side boundary line 13d based on the position information of the selected position 61, making it possible to easily acquire the intervertebral body boundary lines 13 on both sides of the exclusion region 41a in the vertical direction.

Furthermore, in this embodiment, as described above, the superimposed image generation unit 24 is configured to display the calculation target region 41b of the vertebral body region 41 in a different manner from the region other than the calculation target region 41b. This allows the operator to visually easily distinguish between the calculation target region 41b and the region other than the calculation target region 41b in the superimposed image 11. As a result, the operator can visually easily determine whether the desired region of the vertebral body region 41 has been determined as the exclusion region 41a.

As described above, this embodiment further includes a memory unit 5 that stores the superimposed image 11, and the image display control unit 25 is configured to control the display unit 4 to display the previous superimposed image 11a stored in the memory unit 5 along with the superimposed image 11. As a result, the superimposed image 11 and the previous superimposed image 11a are displayed on the display unit 4, allowing the operator to easily compare the superimposed image 11 with the previous superimposed image 11a at a glance. As a result, by checking the previous superimposed image 11a, the operator can determine in advance whether there is an area in the superimposed image 11 that should be selected as an exclusion area 41a.

In addition, in this embodiment, as described above, the input receiving unit 3 is configured to receive an operation (adjustment operation 31) for adjusting the position of the intervertebral body boundary line 13, and the exclusion region determining unit 21 is configured to exclude from the bone density calculation target the region sandwiched between the intervertebral body boundary lines 13 after their positions have been adjusted based on the operation for adjusting the intervertebral body boundary line 13. This allows the operator to easily correct the position of the intervertebral body boundary line 13, for example, if the position of the intervertebral body boundary line 13 acquired by the vertebral body region identifying unit 20 is incorrect. Furthermore, even if the position of the intervertebral body boundary line 13 acquired by the vertebral body region identifying unit 20 is correct, the operator may wish to adjust the position of the intervertebral body boundary line 13 to a desired position. In this case, the above configuration allows the operator to adjust the position of the intervertebral body boundary line 13 to the desired position and then determine the exclusion region 41a. As a result, convenience (usability) for the operator can be improved.

In addition, as described above, this embodiment further includes a calculation result output unit 26 that outputs the calculation results 32 obtained by the bone density calculation unit 22. As a result, the calculation result output unit 26 can present the calculation results 32 to the operator, for example, by outputting the calculation results 32 to the display unit 4. As a result, the operator can easily understand the calculation results 32.

[Modification]
The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims rather than the description of the above embodiments, and further includes all modifications (variations) within the meaning and scope of the claims.

For example, in the above embodiment, an example configuration was shown in which the medical image processing device 100 includes a boundary line display unit 23 and displays the intervertebral body boundary line 13 in the superimposed image 11, but the present invention is not limited to this. For example, the medical image processing device 100 does not need to include a boundary line display unit 23. In this case, the vertebral body region identification unit 20 can identify individual vertebral body regions for each of the multiple vertebral bodies 40 and display them as individual regions in the superimposed image 11.

In addition, while the above embodiment has shown an example of a configuration in which the medical image processing device 100 includes a superimposed image generation unit 24, the present invention is not limited to this. For example, the medical image processing device 100 does not have to include a superimposed image generation unit 24. However, if the medical image processing device 100 does not include a superimposed image generation unit 24, the operator will select the exclusion region 41a based on the X-ray image 10. In this case, the exclusion region 41a must be selected based on the X-ray image 10 in which the vertebral body region 41 is not easily distinguishable from other regions, and depending on the operator's level of skill, selecting the exclusion region 41a may be difficult. Therefore, it is preferable that the medical image processing device 100 includes a superimposed image generation unit 24.

In addition, in the above embodiment, an example was shown in which the exclusion region determiner 21 determines the exclusion region 41a based on a selection operation 30 that selects a marker 12 displayed on the superimposed image 11, but the present invention is not limited to this. For example, the exclusion region determiner 21 may be configured to determine the exclusion region 41a based on an input such as a number (numeric value) that is preset for each vertebral body 40.

In addition, in the above embodiment, an example was shown in which the exclusion region determiner 21 acquires the nearest boundary line 13c and the opposite-side boundary line 13d, and excludes the vertebral body region 41 between the nearest boundary line 13c and the opposite-side boundary line 13d from the bone density calculation target. However, the present invention is not limited to this. For example, the exclusion region determiner 21 may be configured to acquire the intervertebral body boundary line 13 above the selected position 61 and the intervertebral body boundary line 13 below the selected position 61, regardless of the distance from the selected position 61, and exclude the vertebral body region 41 between the acquired intervertebral body boundary line 13 from the bone density calculation target.

Furthermore, in the above embodiment, an example of a configuration in which the image display control unit 25 displays a past superimposed image 11a was shown, but the present invention is not limited to this. The image display control unit 25 does not have to display a past superimposed image 11a.

Furthermore, in the above embodiment, an example of a configuration in which the medical image processing system 300 includes the medical image processing device 100 and the X-ray image capturing device 200 has been shown, but the present invention is not limited to this. For example, if the image processing unit (capturing device image processing unit 53) included in the X-ray image capturing device 200 is configured to function as the medical image processing device 100, the medical image processing system 300 does not need to include the medical image processing device 100.

Furthermore, in the above embodiment, an example of a configuration in which the medical image processing device 100 excludes a vertebral body region 41 corresponding to one vertebral body 40 has been shown, but the present invention is not limited to this. For example, the medical image processing device 100 may exclude multiple vertebral body regions 41 (exclusion regions 41a). For example, the medical image processing device 100 may determine a vertebral body region 41 corresponding to two vertebral bodies 40 as the exclusion region 41a, or may determine a vertebral body region 41 corresponding to three vertebral bodies 40 as the exclusion region 41a.

In addition, in the above embodiment, an example of a configuration was shown in which the medical image processing device 100 adjusts the position of the intervertebral body boundary line 13 based on an operation on the marker 14a at one end 13a of the intervertebral body boundary line 13 and the marker 14b at the other end 13b, but the present invention is not limited to this. For example, the medical image processing device 100 may be configured to adjust the position of the intervertebral body boundary line 13 based on an operation on the intervertebral body boundary line 13 itself (the line of the intervertebral body boundary line 13).

In addition, in the above embodiment, an example of a configuration in which the medical image processing device 100 acquires X-ray images 10 from the X-ray imaging device 200 has been shown, but the present invention is not limited to this. For example, the medical image processing device 100 may be configured to acquire X-ray images 10 that have been captured in advance by the X-ray imaging device 200 and stored on a server or the like.

Furthermore, in the above embodiment, an example of a configuration in which the medical image processing device 100 and the X-ray image capturing device 200 are placed in the same room is shown, but the present invention is not limited to this. For example, the medical image processing device 100 may be placed in a workstation or examination room, and the X-ray image capturing device 200 may be placed in an examination room or operating room. In this case, the medical image processing device 100 and the X-ray image capturing device 200 may be connected, for example, via a network.

In addition, while the above embodiment illustrates an example of a configuration in which the exclusion region determiner 21 determines the exclusion region 41a based on the operator's selection operation 30, the present invention is not limited to this. For example, the exclusion region determiner 21 may be configured to pre-exclude regions of the vertebral body region 41 that are to be excluded from bone density calculation based on the pixel values of the vertebral body region 41. With this configuration, vertebrae with nearby metal members and vertebrae with fractures can be pre-excluded before the operator selects them. As a result, the burden on the operator can be reduced.

In addition, while the above embodiment illustrates an example of a configuration in which the trained model 33 estimates the intervertebral body boundary line 13, the present invention is not limited to this. For example, the trained model 33 may be configured to estimate only the vertebral body region 41. In this case, the vertebral body region identification unit 20 may be configured to acquire the intervertebral body boundary line 13. In this case, the vertebral body region identification unit 20 may acquire, as the intervertebral body boundary line 13, the position between the lower end of the upper vertebral body 40 and the upper end of the lower vertebral body 40 of the vertebral bodies 40 estimated by the trained model 33.

[Aspects]
It will be appreciated by those skilled in the art that the exemplary embodiments described above are examples of the following aspects.

(Item 1)
an image acquisition unit for acquiring X-ray images showing a plurality of vertebrae;
a vertebral body region identifying unit for identifying vertebral body regions, which are regions of the plurality of vertebrae, from the X-ray image;
an input receiving unit that receives a selection operation by an operator;
an exclusion region determining unit that determines, for each of the vertebral body regions, an exclusion region that is a region to be excluded from a bone mineral density calculation target, based on the selection operation input by the input receiving unit;
a bone density calculation unit that calculates bone density using a calculation target region that is the vertebral body region other than the exclusion region determined by the exclusion region determination unit.

(Item 2)
the vertebral body region identifying unit is configured to acquire an intervertebral body boundary line, which is a boundary between adjacent vertebral bodies in the vertebral body region;
a boundary line display unit for displaying the intervertebral boundary line in the vertebral body region;
2. The medical image processing device according to item 1, wherein the exclusion region determination unit is configured to exclude, from bone density calculation, a region sandwiched between the intervertebral body boundary lines on both upper and lower sides of the exclusion region selected based on the selection operation.

(Item 3)
a display unit that displays the X-ray image;
a superimposed image generating unit that generates a superimposed image in which a mark that can identify the vertebral body region is superimposed on the X-ray image;
an image display control unit that controls the display of the superimposed image on the display unit,
the input receiving unit is configured to receive, as the selection operation, an operation input for selecting the sign in the superimposed image;
3. The medical image processing apparatus according to item 2, wherein the exclusion region determination unit is configured to determine the calculation target region based on an operation input for selecting the marker.

(Item 4)
Item 4. The medical image processing device according to item 3, wherein the exclusion region determination unit is configured to acquire position information of a selected position on the superimposed image based on an operation input for selecting the marker, acquire a nearest boundary line that is the intervertebral body boundary line closest to the selected position, and an opposite-side boundary line that is the intervertebral body boundary line on the opposite side of the nearest boundary line with respect to the selected position, and exclude the vertebral body region between the nearest boundary line and the opposite-side boundary line from the calculation target region.

(Item 5)
5. The medical image processing apparatus according to item 3 or 4, wherein the superimposed image generation unit is configured to differentiate a display mode between the calculation target region and a region other than the calculation target region in the vertebral body region.

(Item 6)
further comprising a storage unit that stores the superimposed image;
6. The medical image processing device according to any one of items 3 to 5, wherein the image display control unit is configured to control the display unit to display the past superimposed image stored in the storage unit together with the superimposed image.

(Item 7)
the input receiving unit is configured to receive an operation for adjusting a position of the interbody boundary line as an input operation by an operator,
7. The medical image processing device according to any one of items 2 to 6, wherein the exclusion area determination unit is configured to exclude an area sandwiched between the interbody boundary lines after the position has been adjusted based on the operation of adjusting the interbody boundary line from a target for bone density calculation.

(Item 8)
8. The medical image processing apparatus according to any one of items 1 to 7, further comprising a calculation result output unit that outputs the calculation result by the bone density calculation unit.

(Item 9)
an X-ray source;
a detector for detecting X-rays emitted from the X-ray source;
an image processing device that generates an X-ray image based on the detection result of the detector,
The image processing device includes:
an image acquisition unit for acquiring the X-ray image;
a vertebral body region identifying unit for identifying vertebral body regions, which are regions of the plurality of vertebrae, from the X-ray image;
an input receiving unit that receives a selection operation by an operator;
an exclusion region determining unit that determines, for each of the vertebral body regions, an exclusion region that is a region to be excluded from a bone mineral density calculation target, based on the selection operation input by the input receiving unit;
a bone density calculation unit that calculates bone density using a calculation target region that is the vertebral body region other than the exclusion region determined by the exclusion region determination unit.

(Item 10)
acquiring x-ray images of a plurality of vertebral bodies;
identifying vertebral body regions, which are regions of the plurality of vertebral bodies, from the X-ray image;
determining, for each vertebral region, an exclusion region that is an area to be excluded from the calculation of bone mineral density, based on a selection operation by an operator;
calculating bone density using a calculation target region that is the vertebral body region other than the determined exclusion region.

REFERENCE SIGNS LIST 1 Image acquisition unit 3 Input acceptance unit 4 Display unit 5 Storage unit 10 X-ray image 11, 11c, 11d, 11e Superimposed image 11a Previous superimposed image 12 Mark 13 Intervertebral body boundary line 13c Nearest boundary line 13d Opposite side boundary line 20 Vertebral body region identification unit 21 Exclusion region determination unit 22 Bone density calculation unit 23 Boundary line display unit 24 Superimposed image generation unit 25 Image display control unit 26 Calculation result output unit 30 Selection operation 31 Adjustment operation (operation to locate the position of the intervertebral body boundary line)
32 Calculation results 40 Vertebral body (multiple vertebral bodies)
41 Vertebral body region 41a Exclusion region 41b Calculation target region 50 X-ray source 51 X-ray detection unit (detector)
100 Medical image processing device (image processing device)
200 X-ray image capturing device 300 Medical image processing system

Claims (13)

an image acquisition unit for acquiring X-ray images showing a plurality of vertebrae;
a vertebral body region identifying unit for identifying vertebral body regions, which are regions of the plurality of vertebrae, from the X-ray image;
an input receiving unit that receives a selection operation by an operator;
an exclusion region determination unit that determines , for each vertebral body region, an exclusion region that is a region to be excluded from a bone mineral density calculation target, based on an exclusion selection operation that selects one region to be excluded from the vertebral body region, which is the selection operation input by the input receiving unit;
a bone density calculation unit that calculates bone density using a calculation target region that is the vertebral body region other than the exclusion region determined by the exclusion region determination unit. the vertebral body region identifying unit is configured to acquire an intervertebral body boundary line, which is a boundary between adjacent vertebral bodies in the vertebral body region;
a boundary line display unit for displaying the intervertebral boundary line in the vertebral body region;
2. The medical image processing device according to claim 1, wherein the exclusion area determination unit is configured to exclude an area sandwiched between the intervertebral body boundary lines on both the upper and lower sides of the exclusion area selected based on the exclusion selection operation from the calculation of bone density. a display unit that displays the X-ray image;
a superimposed image generating unit that generates a superimposed image in which a mark that can identify the vertebral body region is superimposed on the X-ray image;
an image display control unit that controls the display of the superimposed image on the display unit,
the input receiving unit is configured to receive, as the selection operation, an operation input for selecting the sign in the superimposed image;
The medical image processing apparatus according to claim 2 , wherein the exclusion region determining unit is configured to determine the calculation target region based on an operation input for selecting the marker. The medical image processing device of claim 3, wherein the exclusion region determination unit is configured to acquire position information of the selected position on the superimposed image based on an operation input for selecting the marker, acquire a nearest boundary line that is the intervertebral body boundary line closest to the selected position, and an opposite boundary line that is the intervertebral body boundary line on the opposite side of the nearest boundary line from the selected position, and exclude the vertebral body region between the nearest boundary line and the opposite boundary line from the calculation target region. The medical image processing device according to claim 3 or 4, wherein the superimposed image generation unit is configured to differentiate the display mode of the calculation target region and the region other than the calculation target region within the vertebral body region. further comprising a storage unit that stores the superimposed image;
The medical image processing device according to any one of claims 3 to 5, wherein the image display control unit is configured to control the display unit to display the past superimposed images stored in the memory unit together with the superimposed image. the input receiving unit is configured to receive an operation for adjusting a position of the interbody boundary line as an input operation by an operator,
The medical image processing device according to any one of claims 2 to 6, wherein the exclusion area determination unit is configured to exclude the area sandwiched between the intervertebral boundary lines after the position has been adjusted based on the operation of adjusting the intervertebral boundary line from the target for bone density calculation. The medical image processing device according to any one of claims 1 to 7, further comprising a calculation result output unit that outputs the calculation results obtained by the bone density calculation unit. The medical image processing apparatus according to claim 1 , wherein the bone density calculation unit is configured to calculate bone densities of a plurality of vertebrae other than the excluded region by using the calculation target region. an X-ray source;
a detector for detecting X-rays emitted from the X-ray source;
an image processing device that generates an X-ray image based on the detection result of the detector,
The image processing device includes:
an image acquisition unit for acquiring the X-ray image;
a vertebral body region identifying unit for identifying a vertebral body region, which is a region of a plurality of vertebrae, from the X-ray image;
an input receiving unit that receives a selection operation by an operator;
an exclusion region determination unit that determines , for each vertebral body region, an exclusion region that is a region to be excluded from a bone mineral density calculation target, based on an exclusion selection operation that selects one region to be excluded from the vertebral body region, which is the selection operation input by the input receiving unit;
a bone density calculation unit that calculates bone density using a calculation target region that is the vertebral body region other than the exclusion region determined by the exclusion region determination unit. The medical image processing system according to claim 10 , wherein the bone density calculation unit is configured to calculate bone densities of a plurality of vertebrae other than the excluded region by using the calculation target region. acquiring x-ray images of a plurality of vertebral bodies;
identifying vertebral body regions, which are regions of the plurality of vertebral bodies, from the X-ray image;
determining an exclusion region, which is a region to be excluded from a bone mineral density calculation target, from the vertebral body region for each region based on an exclusion selection operation, which is a selection operation by an operator, for selecting one region to be excluded from the vertebral body region;
calculating bone density using a calculation target region that is the vertebral body region other than the determined exclusion region. The medical image processing method according to claim 12 , wherein in the step of calculating bone density, the bone densities of a plurality of vertebrae other than the excluded region are calculated using the calculation target region.