JP7242284B2 - Medical image processing device, X-ray diagnostic device and medical image processing program - Google Patents

Description

TECHNICAL FIELD Embodiments of the present invention relate to a medical image processing apparatus, an X-ray diagnostic apparatus, and a medical image processing program.

2. Description of the Related Art Conventionally, in breast examination, tomosynthesis imaging is performed using an X-ray diagnostic apparatus. Specifically, the X-ray diagnostic apparatus first acquires a plurality of projection data while changing the X-ray irradiation angle with respect to the breast. Next, the X-ray diagnostic apparatus generates three-dimensional medical data through reconstruction processing based on the acquired plurality of projection data. Here, a user such as a doctor can sequentially interpret a plurality of tomographic images (slices) included in the three-dimensional medical data to observe lesions.

Furthermore, a technique is known for generating two-dimensional image data (composite 2D image) by synthesizing a plurality of slices included in three-dimensional medical data. Such a synthetic 2D image is two-dimensional medical data with information of the whole breast. That is, the user can observe the entire breast by interpreting the synthesized 2D image. However, in the synthesized 2D image, lesions appearing in individual slices are blurred, making observation difficult in some cases.

Japanese Patent Publication No. 2017-510323

The problem to be solved by the present invention is to improve the visibility of a lesion in a synthesized 2D image obtained by tomosynthesis imaging.

A medical image processing apparatus according to an embodiment includes a detection unit and an image processing unit. The detection unit detects the positions of lesions in the breast in a plurality of tomographic images included in three-dimensional medical data obtained by tomosynthesis imaging the breast of the subject. The image processing unit generates two-dimensional image data by synthesizing the plurality of tomographic images based on the detected positions of the lesions.

FIG. 1 is a block diagram showing an example of the configuration of a medical image processing system according to the first embodiment. FIG. 2 is a block diagram showing an example configuration of the X-ray diagnostic apparatus according to the first embodiment. FIG. 3A is a diagram for explaining tomosynthesis imaging according to the first embodiment; FIG. 3B is a diagram for explaining tomosynthesis imaging according to the first embodiment; FIG. 3C is a diagram for explaining tomosynthesis imaging according to the first embodiment; FIG. 4 is a diagram for explaining reconstruction of three-dimensional medical data according to the first embodiment. FIG. 5 is a diagram showing an example of three-dimensional medical data according to the first embodiment. FIG. 6A is a diagram showing an example of lesion detection according to the first embodiment. FIG. 6B is a diagram illustrating an example of lesion detection according to the first embodiment; FIG. 7 is a diagram illustrating an example of weighting processing according to the first embodiment. FIG. 8 is a diagram illustrating an example of a synthesized 2D image according to the first embodiment; FIG. 9A is a diagram showing an example of lesion detection according to the first embodiment. 9B is a diagram illustrating an example of lesion detection according to the first embodiment; FIG. FIG. 10 is a diagram illustrating an example of weighting processing according to the first embodiment. FIG. 11 is a flowchart for explaining a series of processing flows of the medical image processing apparatus according to the first embodiment. FIG. 12 is a diagram illustrating an example of weight distribution according to the second embodiment. FIG. 13 is a diagram illustrating an example of weighting processing according to the second embodiment. FIG. 14A is a diagram showing an example of lesion detection according to the second embodiment. FIG. 14B is a diagram illustrating an example of lesion detection according to the second embodiment. FIG. 15 is a diagram illustrating an example of calculation of a combined distribution according to the second embodiment. FIG. 16 is a diagram illustrating an example of weighting processing according to the second embodiment. FIG. 17 is a flowchart for explaining a series of processing flows of the medical image processing apparatus according to the second embodiment. FIG. 18 is a diagram for explaining generation of a synthesized 2D image according to the third embodiment. FIG. 19 is a flowchart for explaining a series of processing flows of the medical image processing apparatus according to the third embodiment. FIG. 20 is a block diagram showing an example of a processing circuit according to the fourth embodiment;

A medical image processing apparatus, an X-ray diagnostic apparatus, and a medical image processing program according to embodiments will be described below with reference to the drawings.

(First embodiment)
First, the first embodiment will be explained. In the first embodiment, a medical image processing system 1 including an X-ray diagnostic apparatus 10, an image storage apparatus 20 and a medical image processing apparatus 30 will be described.

FIG. 1 is a block diagram showing an example of the configuration of a medical image processing system 1 according to the first embodiment. As shown in FIG. 1, the medical image processing system 1 according to the first embodiment includes an X-ray diagnostic apparatus 10, an image storage apparatus 20, and a medical image processing apparatus 30. FIG. As shown in FIG. 1, an X-ray diagnostic apparatus 10, an image storage apparatus 20 and a medical image processing apparatus 30 are interconnected via a network NW.

The X-ray diagnostic apparatus 10 is an apparatus that collects medical data from a subject. Specifically, the X-ray diagnostic apparatus 10 acquires three-dimensional medical data by performing tomosynthesis imaging of the breast P of the subject. The configuration of the X-ray diagnostic apparatus 10 will be described later.

The image storage device 20 is a device that stores medical data collected by the X-ray diagnostic apparatus 10 . For example, the image storage device 20 acquires three-dimensional medical data from the X-ray diagnostic device 10 via the network NW, and stores the acquired three-dimensional medical data in a memory provided inside or outside the device. For example, the image storage device 20 is realized by computer equipment such as a server device.

The medical image processing apparatus 30 acquires three-dimensional medical data from the image storage apparatus 20 via the network NW, and executes various processes based on the acquired three-dimensional medical data. For example, the medical image processing device 30 generates a synthetic 2D image based on 3D medical data. Processing performed by the medical image processing apparatus 30 will be described later. For example, the medical image processing apparatus 30 is implemented by computer equipment such as a workstation.

Note that the location where the X-ray diagnostic apparatus 10, the image storage apparatus 20, and the medical image processing apparatus 30 are installed is arbitrary as long as they can be connected via the network NW. For example, the medical image processing apparatus 30 may be installed in a hospital different from the X-ray diagnostic apparatus 10 . That is, the network NW may be configured by a closed local network within the hospital, or may be a network via the Internet. Also, although one X-ray diagnostic apparatus 10 is shown in FIG. 1 , the medical image processing system 1 may include a plurality of X-ray diagnostic apparatuses 10 .

As shown in FIG. 1, the medical image processing apparatus 30 has an input interface 31, a display 32, a memory 33, and a processing circuit .

The input interface 31 receives various input operations from the operator, converts the received input operations into electrical signals, and outputs the electrical signals to the processing circuit 34 . For example, the input interface 31 includes a mouse, a keyboard, a trackball, a switch, a button, a joystick, a touch pad that performs input operations by touching an operation surface, a touch screen that integrates a display screen and a touch pad, and an optical sensor. It is realized by the used non-contact input circuit, voice input circuit, or the like. The input interface 31 may be composed of a tablet terminal or the like capable of wireless communication with the main body of the medical image processing apparatus 30 . Also, the input interface 31 is not limited to those having physical operation parts such as a mouse and a keyboard. For example, an electrical signal processing circuit for receiving an electrical signal corresponding to an input operation from an external input device provided separately from the medical image processing apparatus 30 and outputting the electrical signal to the processing circuit 34 is also included in the input interface 31. included in the example.

The display 32 displays various information. For example, the display 32 displays a GUI (Graphical User Interface) for receiving various instructions and various settings from the operator via the input interface 31 . Also, the display 32 displays various image data collected about the breast P of the subject. For example, the display 32 sequentially displays multiple slices included in the three-dimensional medical data. Also, the display 32 displays a synthesized 2D image generated based on the three-dimensional medical data. For example, the display 32 is a liquid crystal display or a CRT (Cathode Ray Tube) display. The display 32 may be of a desktop type, or may be configured by a tablet terminal or the like capable of wireless communication with the main body of the medical image processing apparatus 30 .

The memory 33 is implemented by, for example, a RAM (Random Access Memory), a semiconductor memory device such as a flash memory, a hard disk, an optical disk, or the like. For example, the memory 33 stores the three-dimensional medical data of the breast P acquired from the image storage device 20 and the synthetic 2D image generated by the processing circuit 34 . In addition, for example, the memory 33 stores programs for the circuits included in the medical image processing apparatus 30 to implement their functions. Note that the memory 33 may be realized by a server group (cloud) connected to the medical image processing apparatus 30 via the network NW. Also, the memory 33 is an example of a storage unit.

The processing circuit 34 controls the overall operation of the medical image processing apparatus 30 by executing a control function 341 , a detection function 342 , an image processing function 343 and a display control function 344 . Here, the detection function 342 is an example of a detection unit. Also, the image processing function 343 is an example of an image processing unit. Also, the display control function 344 is an example of a display control unit.

For example, the processing circuit 34 reads a program corresponding to the control function 341 from the memory 33 and executes it, thereby controlling various functions of the processing circuit 34 based on an input operation received from the operator via the input interface 31. do. Also, the control function 341 controls transmission and reception of data via the network NW. For example, the control function 341 receives three-dimensional medical data of the breast P from the image storage device 20 and stores it in the memory 33 . Also, for example, the control function 341 transmits the synthesized 2D image generated by the image processing function 343 to the image storage device 20 .

Further, for example, the processing circuit 34 reads a program corresponding to the detection function 342 from the memory 33 and executes it, thereby detecting the position of the lesion in a plurality of tomographic images (slices) included in the three-dimensional medical data of the breast P. do. Further, the processing circuit 34 reads a program corresponding to the image processing function 343 from the memory 33 and executes it to generate two-dimensional image data (composite 2D image) from a plurality of slices included in the three-dimensional medical data. Processing by the detection function 342 and the image processing function 343 will be described later. In addition, the processing circuit 34 displays a synthesized 2D image generated by the image processing function 343 on the display 32 by reading out and executing a program corresponding to the display control function 344 from the memory 33 .

In the medical image processing apparatus 30 shown in FIG. 1, each processing function is stored in the memory 33 in the form of a computer-executable program. The processing circuit 34 is a processor that implements a function corresponding to each program by reading the program from the memory 33 and executing the program. In other words, the processing circuit 34 in a state where each program is read has a function corresponding to the read program. In FIG. 1, the control function 341, the detection function 342, the image processing function 343, and the display control function 344 are realized by the single processing circuit 34, but processing is performed by combining a plurality of independent processors. The circuit 34 may be configured and each processor may implement the function by executing a program. Further, each processing function of the processing circuit 34 may be appropriately distributed or integrated in a single or a plurality of processing circuits and implemented.

Next, the X-ray diagnostic apparatus 10 will be described. The X-ray diagnostic apparatus 10 performs tomosynthesis imaging on the breast P of the subject and collects three-dimensional medical data. In the following description, the X-ray diagnostic apparatus 10 is assumed to be a mammography apparatus.

For example, the X-ray diagnostic apparatus 10 has a base 101 and a stand 102 as shown in FIG. A stand 102 is erected on a base 101 and supports an imaging table 103, a compression plate 104, an X-ray tube 105, an X-ray diaphragm 106, an X-ray detector 107, and a signal processing circuit 108. do. Here, the stand 102 supports the imaging table 103, the compression plate 104, the X-ray detector 107, and the signal processing circuit 108 so as to be vertically movable. Note that FIG. 2 is a block diagram showing an example of the configuration of the X-ray diagnostic apparatus 10 according to the first embodiment.

The imaging table 103 is a table for supporting the breast P of the subject, and has a support surface on which the breast P is placed. The compression plate 104 is arranged above the imaging stand 103 and provided so as to face the imaging stand 103 in parallel. Here, the compression plate 104 is provided so as to be movable in the directions of contacting and separating from the imaging table 103 . For example, the compression plate 104 compresses the breast P supported on the imaging table 103 by moving in a direction approaching the imaging table 103 . The breast compressed by the compression plate 104 is spread thinly, and overlapping of the mammary glands within the breast is reduced.

The X-ray tube 105 is a vacuum tube having a cathode (filament) that generates thermoelectrons and an anode (target) that generates X-rays upon collision with thermoelectrons. The X-ray tube 105 uses a high voltage supplied from the X-ray high voltage device 111 to irradiate thermoelectrons from the cathode to the anode, thereby generating X-rays. Here, the X-ray tube 105 is configured to be movable so as to change the irradiation angle of the X-rays to the breast P. Note that the movement of the X-ray tube 105 will be described later.

The X-ray restrictor 106 is arranged between the X-ray tube 105 and the compression plate 104 to control the X-rays generated by the X-ray tube 105 . For example, the X-ray restrictor 106 has a collimator that narrows down the irradiation range of X-rays and a filter that adjusts the X-rays.

The collimator in the X-ray diaphragm 106 has, for example, four slidable diaphragm blades. By sliding these diaphragm blades, the X-rays generated by the X-ray tube 105 are focused and irradiated to the breast P. . Here, the aperture blade is a plate-like member made of lead or the like, and is provided near the X-ray irradiation port of the X-ray tube 105 in order to adjust the X-ray irradiation range.

The filter in the X-ray restrictor 106 changes the quality of transmitted X-rays depending on its material and thickness for the purpose of reducing the exposure dose to the subject and improving the image quality of the X-ray image data, and is easily absorbed by the subject. It reduces soft-line components and high-energy components that reduce image contrast. Also, the filter changes the X-ray dose and irradiation range depending on its material, thickness, position, etc., and attenuates the X-rays so that the X-rays irradiated to the breast P have a predetermined distribution.

For example, the X-ray diaphragm 106 has a drive mechanism such as a motor and an actuator, and controls X-ray irradiation by operating the drive mechanism under the control of a processing circuit 114, which will be described later. For example, the X-ray diaphragm 106 applies a driving voltage to the driving mechanism according to the control signal received from the processing circuit 114, thereby adjusting the opening degree of the diaphragm blades of the collimator and irradiating the breast P. Controls the irradiation range of X-rays. Further, for example, the X-ray diaphragm 106 adjusts the position of the filter by applying a driving voltage to the driving mechanism in accordance with the control signal received from the processing circuit 114, thereby irradiating the breast P. Control the X-ray dose distribution.

The X-ray detector 107 is, for example, an X-ray flat panel detector (FPD) having detection elements arranged in a matrix. The X-ray detector 107 detects X-rays emitted from the X-ray tube 105 and transmitted through the breast P, and outputs a detection signal corresponding to the detected X-ray dose to the signal processing circuit 108 . The X-ray detector 107 may be an indirect conversion type detector having a grid, a scintillator array, and a photosensor array, or a direct conversion type detector having a semiconductor element that converts incident X-rays into electrical signals. It may be a detector.

For example, the X-ray detector 107 detects an X-ray pulse emitted from the X-ray tube 105 and generates a detection signal corresponding to the detected X-ray dose. Here, the X-ray detector 107 holds the generated detection signal. Also, the X-ray detector 107 outputs a detection signal to the signal processing circuit 108 after irradiation with the X-ray pulse. Then, the signal processing circuit 108 generates projection data based on the detection signal output from the X-ray detector 107 and stores it in the memory 112 .

Further, as shown in FIG. 2, the X-ray diagnostic apparatus 10 has an input interface 109, an elevation drive device 110, an X-ray high voltage device 111, a memory 112, a display 113, and a processing circuit 114.

The input interface 109 receives various input operations from the operator, converts the received input operations into electrical signals, and outputs the electrical signals to the processing circuit 114 . For example, the input interface 109 includes a mouse, a keyboard, a trackball, a switch, a button, a joystick, a touch pad that performs input operations by touching an operation surface, a touch screen that integrates a display screen and a touch pad, and an optical sensor. It is realized by the used non-contact input circuit, voice input circuit, or the like. Note that the input interface 109 may be configured by a tablet terminal or the like capable of wireless communication with the processing circuit 114 . Also, the input interface 109 is not limited to having physical operation parts such as a mouse and a keyboard. For example, the input interface 109 also includes an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the main body of the X-ray diagnostic apparatus 10 and outputs the electrical signal to the processing circuit 114 . included in the examples of

The lift drive device 110 is connected to the imaging table 103 and the compression plate 104 . For example, the elevation driving device 110 raises and lowers the imaging table 103 in the vertical direction. Further, for example, the elevation driving device 110 raises and lowers the compression plate 104 in the vertical direction (the direction of contacting and separating from the imaging table 103). For example, the elevation driving device 110 has a driving mechanism such as a motor and an actuator, and operates the driving mechanism under the control of the processing circuit 114 to control elevation of the imaging table 103 and the compression plate 104 .

X-ray high voltage device 111 supplies high voltage to X-ray tube 105 under the control of processing circuitry 114 . For example, the X-ray high-voltage device 111 has electric circuits such as a transformer and a rectifier, and includes a high-voltage generator that generates a high voltage to be applied to the X-ray tube 105 and a voltage that the X-ray tube 105 emits. and an X-ray control device for controlling an output voltage according to X-rays. The high voltage generator may be of a transformer type or an inverter type.

The memory 112 is implemented by, for example, a RAM, a semiconductor memory device such as a flash memory, a hard disk, an optical disk, or the like. For example, the memory 112 stores projection data generated by the signal processing circuit 108 and three-dimensional medical data generated by the processing circuit 114 . In addition, for example, the memory 112 stores programs for the circuits included in the X-ray diagnostic apparatus 10 to implement their functions. Note that the memory 112 may be realized by a server group (cloud) connected to the X-ray diagnostic apparatus 10 via the network NW.

The display 113 displays various information. For example, the display 113 displays a GUI for receiving various instructions and various settings from the operator via the input interface 109 . Moreover, the display 113 displays various image data collected about the breast P. FIG. For example, the display 113 sequentially displays multiple slices included in the three-dimensional medical data. Also, for example, the display 113 displays a composite 2D image generated by the medical image processing apparatus 30 . For example, display 113 is a liquid crystal display or a CRT display. The display 113 may be of a desktop type, or may be configured by a tablet terminal or the like capable of wireless communication with the main body of the X-ray diagnostic apparatus 10 .

The processing circuit 114 controls the overall operation of the X-ray diagnostic apparatus 10 by executing a control function 114a and a display control function 114b.

For example, the processing circuit 114 reads a program corresponding to the control function 114a from the memory 112 and executes it, thereby controlling various functions of the processing circuit 114 based on an input operation received from the operator via the input interface 109. do.

The control function 114a also controls acquisition of three-dimensional medical data. Specifically, first, the control function 114a performs tomosynthesis imaging on the breast P to collect a plurality of projection data. Next, the control function 114a performs correction processing such as logarithmic conversion processing, offset correction, sensitivity correction, and beam hardening correction on the collected projection data to generate corrected projection data. Processing circuitry 114 then reconstructs three-dimensional medical data based on the corrected projection data.

The control function 114a can also acquire mammography images such as MLO (Mediolateral-Oblique) images and CC (Cranio-Caudal) images. Specifically, the control function 114a fixes the positions of the imaging table 103 and the compression plate 104 in the MLO direction and the CC direction, and irradiates the breast P with X-rays while keeping the X-ray irradiation angle constant. , mammography images such as MLO and CC images.

The control function 114a also controls transmission and reception of data via the network NW. For example, the control function 114 a transmits three-dimensional medical data obtained by tomosynthesis imaging of the breast P of the subject to the image storage device 20 . The processing circuit 34 also displays various image data on the display 32 by reading out and executing a program corresponding to the display control function 344 from the memory 33 .

In the X-ray diagnostic apparatus 10 shown in FIG. 2, each processing function is stored in the memory 112 in the form of a computer-executable program. The signal processing circuit 108 and the processing circuit 114 are processors that implement functions corresponding to each program by reading and executing the programs from the memory 112 . In other words, the signal processing circuit 108 and the processing circuit 114 from which the program has been read have functions corresponding to the read program. 2, the control function 114a and the display control function 114b are realized by the single processing circuit 114, but the processing circuit 114 is configured by combining a plurality of independent processors, and each processor The function may be implemented by executing a program. Further, each processing function of the processing circuit 114 may be appropriately distributed or integrated in a single or a plurality of processing circuits and implemented.

The term "processor" used in the above description includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an application specific integrated circuit (ASIC), a programmable logic device (e.g., Circuits such as Simple Programmable Logic Device (SPLD), Complex Programmable Logic Device (CPLD), and Field Programmable Gate Array (FPGA)). The processor implements its functions by reading and executing programs stored in the memory 33 or memory 112 .

1 and 2, it is assumed that the single memory 33 or memory 112 stores programs corresponding to each processing function. However, embodiments are not so limited. For example, a plurality of memories 33 may be arranged in a distributed manner, and the processing circuit 34 may be configured to read corresponding programs from individual memories 33 . Similarly, a plurality of memories 112 may be distributed and the processing circuit 114 may read corresponding programs from the individual memories 112 . Also, instead of storing the program in the memory 33 and memory 112, the program may be configured to be directly embedded in the circuit of the processor. In this case, the processor implements its functions by reading and executing the program embedded in the circuit.

Also, the processing circuit 34 and the processing circuit 114 may realize their functions using a processor of an external device connected via the network NW. For example, the processing circuit 34 reads and executes a program corresponding to each function from the memory 33, and uses a server group (cloud) connected to the medical image processing apparatus 30 via the network NW as computational resources. , implement the functions shown in FIG.

The medical image processing system 1 including the medical image processing apparatus 30 has been described above. With such a configuration, the medical image processing apparatus 30 in the medical image processing system 1 improves the visibility of lesions in a synthesized 2D image obtained by tomosynthesis imaging through processing by the processing circuit 34 .

First, tomosynthesis imaging will be described. In tomosynthesis imaging, the breast P of the subject is placed on the imaging table 103 and then compressed between the imaging table 103 and the compression plate 104 . Next, the control function 114a acquires a plurality of projection data by irradiating the breast P with X-rays while changing the X-ray irradiation angle. Acquisition of a plurality of projection data will be described below with reference to FIGS. 3A, 3B, and 3C. 3A, 3B, and 3C are diagrams for explaining tomosynthesis imaging according to the first embodiment.

3A shows the case where the X-ray tube 105 is positioned at position X1, FIG. 3B shows the case where the X-ray tube 105 is positioned at position X2, and FIG. 3C shows the case where the X-ray tube 105 is positioned at position X3. indicate the case. As shown in FIGS. 3A, 3B, and 3C, the x-ray tube 105 moves in a trajectory that includes positions X1, X2, and X3. Broken lines in FIGS. 3A, 3B, and 3C indicate part of the X-rays emitted from the X-ray tube 105. FIG. As shown in FIGS. 3A, 3B, and 3C, the X-ray tube 105 moves so as to change the irradiation angle of X-rays.

Specifically, the control function 114a first irradiates the breast P with X-rays from the X-ray tube 105 at the position X1, as shown in the upper diagram of FIG. 3A. Here, the X-rays that have passed through the breast P are detected by the X-ray detector 107 . Also, the signal processing circuit 108 generates projection data A101 shown in the lower diagram of FIG. 3A based on the detection signal output from the X-ray detector 107 . In the projection data A101, the regions a, b, and c of the breast P are projected to the positions shown in the lower diagram of FIG. 3A.

Next, as shown in the upper diagram of FIG. 3B, the control function 114a changes the X-ray irradiation angle from the state shown in FIG. 3A to irradiate the breast P from the X-ray tube 105 at the position X2. Here, the X-rays that have passed through the breast P are detected by the X-ray detector 107 . The signal processing circuit 108 also generates projection data A102 shown in the lower diagram of FIG. 3B based on the detection signal output from the X-ray detector 107 . In the projection data A102, the regions a, b, and c of the breast P are projected so as to be superimposed on the positions shown in the lower diagram of FIG. 3B.

Further, the control function 114a changes the X-ray irradiation angle from the state shown in FIG. 3B to irradiate the breast P from the X-ray tube 105 at the position X3, as shown in the upper diagram of FIG. 3C. Here, the X-rays that have passed through the breast P are detected by the X-ray detector 107 . Also, the signal processing circuit 108 generates projection data A103 shown in the lower diagram of FIG. 3C based on the detection signal output from the X-ray detector 107 . In the projection data A103, the regions a, b, and c of the breast P are projected to the positions shown in the lower diagram of FIG. 3C.

For example, the control function 114a continuously changes the position of the X-ray tube 105 and generates X-rays when the X-ray tube 105 reaches a position X1, a position X2, a position X3, or the like. Further, for example, the control function 114a alternately repeats the movement of the X-ray tube 105 and the irradiation of X-rays to generate X-rays from the X-ray tube 105 stopped at the position X1, the position X2, the position X3, or the like. . The trajectory along which the X-ray tube 105 moves may be arc-shaped as shown in FIGS. 3A, 3B, and 3C, may be linear, or may be other trajectories.

Next, the control function 114a reconstructs three-dimensional medical data from a plurality of projection data acquired by changing irradiation angles. The reconstruction of three-dimensional medical data will be described below with reference to FIG. FIG. 4 is a diagram for explaining reconstruction of three-dimensional medical data according to the first embodiment. In FIG. 4, as an example, the case of reconstructing three-dimensional medical image data by the shift addition method will be described.

For example, the control function 114a adds projection data A101, projection data A102, and projection data A103 shown in the upper diagram of FIG. Create a slice Ba shown. In the slice Ba, since the projected portions of the projection data A101, the projection data A102, and the projection data A103 are aligned and added, the portion a is clearly expressed.

Here, although the individual projection data also include signals indicating the region b and the region c, the signals of the region b and the region c are superimposed on the slice Ba with their positions shifted. Diffused and blurred. Therefore, as shown in the lower diagram of FIG. 4, only the portion a is clearly expressed in the slice Ba. That is, the control function 114a generates a slice Ba using a plane including the region a in the breast P as a reconstruction plane.

Note that the direction parallel to the reconstruction plane is hereinafter referred to as the X direction. Also, hereinafter, the direction parallel to the reconstruction plane and perpendicular to the X direction is defined as the Y direction. That is, as shown in FIG. 4, the X and Y directions are parallel to the slice. Also, hereinafter, the direction orthogonal to the X direction and the Y direction is defined as the Z direction. That is, the Z direction is the direction perpendicular to the slice. Note that the Z direction is also referred to as the slice direction.

Similarly, the control function 114a adds the projection data A101, the projection data A102, and the projection data A103 so as to align the projected position of the part b on each projection data to generate a slice Bb (not shown). That is, the control function 114a generates the slice Bb using the XY plane including the region b in the breast P as a reconstruction plane. Further, the control function 114a adds the projection data A101, the projection data A102, and the projection data A103 so as to align the projected position of the part c on each projection data, and generates a slice Bc (not shown). That is, the control function 114a generates the slice Bc using the XY plane including the site c in the breast P as a reconstruction plane.

As described above, the control function 114a uses a plurality of projection data to generate a plurality of slices while shifting the reconstruction plane. That is, the control function 114a reconstructs three-dimensional medical data including multiple slices using multiple projection data.

As an example, a case where the control function 114a reconstructs the three-dimensional medical data shown in FIG. 5 will be described below. That is, hereinafter, the case where the three-dimensional medical data includes "50" slices from slice B101 to slice B150 will be described. FIG. 5 is a diagram showing an example of three-dimensional medical data according to the first embodiment. The control function 114 a transmits the generated three-dimensional medical data to the image storage device 20 .

The medical image processing apparatus 30 receives the 3D medical data from the image storage apparatus 20 and stores it in the memory 33 . Note that the medical image processing apparatus 30 may receive three-dimensional medical data from the X-ray diagnostic apparatus 10 without going through the image storage apparatus 20 . The medical image processing apparatus 30 also generates two-dimensional image data (composite 2D image) by synthesizing a plurality of slices included in the three-dimensional medical data.

Here, as a method of generating a synthetic 2D image from three-dimensional medical data, for example, maximum intensity projection (MIP) and minimum intensity projection (MinIP) are known. However, when a synthetic 2D image is generated by these methods, lesions appearing in individual slices may become blurred and difficult to observe. That is, when a slice in which a lesion appears is synthesized with a slice in which no lesion appears, a lesion that was clear before synthesis may be blurred.

Therefore, the processing circuit 34 detects the positions of the lesions of the breast P in a plurality of slices included in the three-dimensional medical data, and generates a composite 2D image from the plurality of slices based on the detected positions of the lesions. Improve visibility of lesions in 2D images. The generation of the composite 2D image by the processing circuitry 34 will be described in detail below.

First, the detection function 342 detects the positions of lesions in the breast P in multiple slices included in the three-dimensional medical data. Specifically, detection function 342 performs computer-aided detection (CAD) on each of slices B101 to B150 shown in FIG. Locate the P lesion.

A case where a lesion is detected in the slice B120 shown in FIG. 6A and the slice B130 shown in FIG. 6B among the slices B101 to B150 will be described below. Specifically, the detection function 342 detects the lesion L1 at the position shown in FIG. 6A in the slice B120, and detects the lesion L2 at the position shown in FIG. 6B in the slice B130. Note that the pixel P1 shown in FIG. 6A is an example of the pixel at the position of the lesion L1. Hereinafter, the XY coordinates of the pixel P1 are assumed to be (x1, y1). A pixel P2 shown in FIG. 6B is an example of a pixel at the position of the lesion L2. Hereinafter, let the XY coordinates of the pixel P2 be (x2, y2). 6A and 6B are diagrams showing an example of lesion detection according to the first embodiment.

The image processing function 343 then determines weights for each pixel in the multiple slices based on the locations of the lesions detected by the detection function 342 . That is, the image processing function 343 determines weights for each slice and each pixel based on the location of the lesion.

For example, the image processing function 343 determines the weight of the pixel P1 (the pixel at the position of the lesion L1) to be "1", and determines the weight of the pixel having the same XY coordinates as the pixel P1 in another slice to be "0". do. That is, as shown in FIG. 7, the image processing function 343 determines the weight of the pixel P1 whose XY coordinates are (x1, y1) in the slice B120 to be "1", and the other slices (slices B101 to B119, Then, the weight of the pixel whose XY coordinates are (x1, y1) in slices B121 to B150) is set to "0". FIG. 7 is a diagram showing an example of weighting processing according to the first embodiment.

As described above, the image processing function 343 determines the pixel weights in each slice for the XY coordinates (x1, y1). The image processing function 343 also determines the weight of the pixel in each slice for each of the XY coordinates included in the position of the lesion L1, in the same manner as for the XY coordinates (x1, y1).

Similarly, the image processing function 343 sets the weight of the pixel P2 (the pixel at the position of the lesion L2) to "1", and sets the weight of the pixel having the same XY coordinates as the pixel P2 in another slice to "0". ” is decided. That is, as shown in FIG. 7, the image processing function 343 determines the weight of the pixel P2 whose XY coordinates are (x2, y2) in the slice B130 to be "1", and the other slices (slice B101 to slice B129, Also, in slices B131 to B150), the weight of pixels whose XY coordinates are (x2, y2) is determined to be "0". The image processing function 343 also determines the weight of the pixel in each slice for each of the XY coordinates included in the position of the lesion L2, similarly to the XY coordinates (x2, y2).

Note that one lesion may be detected over a plurality of slices depending on the size of the lesion. That is, a lesion may be detected over a plurality of positions overlapping in the slice direction (Z direction). In this case, the image processing function 343 performs multiple slices so that the weight of the pixel at the position with the smallest pixel value among the multiple positions where the lesion is detected is greater than the weight of the pixel at the other positions. Determine the weight for each pixel.

For example, when the lesion L1 is detected over slices B119, B120, and B121, the image processing function 343 compares pixel values among the slices for each XY coordinate of the position where the lesion L1 is detected. Note that the image processing function 343 compares the pixel values without inverting the gradation. That is, the image processing function 343 compares the pixel values, assuming that the pixel values are smaller in regions where X-rays are less likely to pass (bone, lesion, etc.).

For example, the image processing function 343 processes a pixel with XY coordinates of (x1, y1) in slice B119, a pixel P1 with XY coordinates of (x1, y1) in slice B120, and a pixel with XY coordinates of (x1, y1) in slice B121. A pixel value is compared with a pixel that becomes (x1, y1). Here, for example, when the pixel value of the pixel P1 is the smallest, the image processing function 343 sets the weight of the pixel P1 to be higher than the weight of each pixel whose XY coordinates are (x1, y1) in the slices B119 and B121. Determine the weights for each pixel in multiple slices in increasing order. For example, the image processing function 343 sets the weight of the pixel P1 to "1" and sets the weight of each pixel whose XY coordinates are (x1, y1) to "0" in the slices B119 and B121. In other words, the image processing function 343 determines a weight for each pixel in a plurality of slices such that, when a lesion is detected over a plurality of overlapping positions in the slice direction, the clearer the lesion, the greater the weight. .

Next, the image processing function 343 generates a synthesized 2D image by synthesizing multiple slices included in the three-dimensional medical data based on the determined weights. For example, the image processing function 343 generates two-dimensional image data C101 (composite 2D image) shown in FIG. 8 by executing weighted average processing based on the determined weight for each XY coordinate. Note that FIG. 8 is a diagram showing an example of a synthesized 2D image according to the first embodiment.

For example, the image processing function 343 performs weighted average processing using the pixel value of each pixel and the weight determined for each pixel for the XY coordinates (x1, y1). Specifically, the image processing function 343 calculates the product of the pixel value of the pixel P1 whose XY coordinates are (x1, y1) in the slice B120 and the weight "1", and the XY coordinates are ( x1, y1) and the weight "0" is calculated. Then, the image processing function 343 sets the sum of the calculated products as the pixel value of the pixel whose XY coordinates are (x1, y1) in the two-dimensional image data C101.

Also, the image processing function 343 determines the pixel values of the two-dimensional image data C101 for each of the XY coordinates included in the position of the lesion L1 in the same manner as the XY coordinates (x1, y1). In this case, in the two-dimensional image data C101, the area having the same XY coordinates as the lesion L1 in the slice B120 (area R1 in FIG. 8) reflects the lesion L1 in the slice B120 as it is.

The image processing function 343 also performs weighted average processing using the pixel value of each pixel and the weight determined for each pixel for the XY coordinates (x2, y2). Specifically, the image processing function 343 calculates the product of the pixel value of the pixel P2 whose XY coordinates are (x2, y2) in the slice B130 and the weight "1", and the XY coordinates are ( x2, y2) and the weight "0" are multiplied. Then, the image processing function 343 sets the sum of the calculated products as the pixel value of the pixel whose XY coordinates are (x2, y2) in the two-dimensional image data C101.

The image processing function 343 also determines the pixel values of the two-dimensional image data C101 for each of the XY coordinates included in the position of the lesion L2, similarly to the XY coordinates (x2, y2). In this case, in the two-dimensional image data C101, the lesion L2 in the slice B130 is directly reflected in the region having the same XY coordinates as the lesion L2 in the slice B130 (region R2 in FIG. 8).

Note that the image processing function 343 uses an arbitrary method to convert the XY coordinates (XY coordinates excluding the XY coordinates of the lesion L1 and the XY coordinates of the lesion L2) from which no lesion was detected in any slice to the two-dimensional image data C101. Pixel values can be determined. For example, the image processing function 343 determines pixel values of the two-dimensional image data C101 by minimum intensity projection without determining weights for XY coordinates in which no lesion is detected in any slice. That is, the image processing function 343 executes weighted averaging processing based on the determined weight when pixels with determined weights exist in the slice direction, and when pixels with determined weights do not exist in the slice direction. generates two-dimensional image data C101 by executing minimum intensity projection processing.

As described above, the image processing function 343 determines the weight of each pixel in a plurality of slices based on the position of the lesion, and synthesizes the plurality of slices based on the determined weights to obtain the two-dimensional image data C101. Generate. As a result, the two-dimensional image data C101 represents lesions in individual slices. For example, in the two-dimensional image data C101, the region R1 represents the lesion L1 in the slice B120, and the region R2 represents the lesion L2 in the slice B130. Therefore, the medical image processing apparatus 30 can improve the visibility of lesions in the two-dimensional image data C101 obtained by tomosynthesis imaging.

Here, types of lesions include tumors, calcifications, architectural disturbances, and the like. Among them, tumors and calcifications are expressed with lower pixel values than other regions in the three-dimensional medical data. Therefore, tumors and calcifications usually also appear on synthetic 2D images generated by minimum intensity projection or the like. On the other hand, structural disturbance is a change in the shape of the mammary gland, so the pixel values in the three-dimensional medical data are not significantly different from those in other regions. Therefore, even if construction distortion appears in some slices, the construction distortion is usually blurred in a synthesized 2D image generated by minimum intensity projection or the like. On the other hand, the medical image processing apparatus 30 detects the positions of lesions in a plurality of slices included in the three-dimensional medical data, generates a synthetic 2D image based on the positions of the lesions, and visually recognizes structural disturbances. It can also improve sexuality.

Note that the medical image processing apparatus 30 may generate a composite 2D image using not only the lesion position but also the lesion type. In this case, first, the detection function 342 detects the positions and types of lesions in the breast P in multiple slices included in the three-dimensional medical data. For example, detection function 342 detects the location and type of lesions in breast P in slices B101-B150 by performing computer-aided detection (CAD) on each of slices B101-B150 shown in FIG. do.

As an example, a case where a lesion is detected in the slice B120 shown in FIG. 9A and the slice B130 shown in FIG. 9B among the slices B101 to B150 will be described below. 9A and 9B are diagrams showing an example of lesion detection according to the first embodiment.

Specifically, the detection function 342 detects the lesion L3, which is the “construction disturbance”, at the position shown in FIG. 9A in the slice B120. Note that the pixel P3 shown in FIG. 9A is an example of the pixel at the position of the lesion L3. Hereinafter, the XY coordinates of pixel P3 are assumed to be (x3, y3). Further, the detection function 342 detects a “tumor” lesion L4 and a “calcification” lesion L5 at the positions shown in FIG. 9B in the slice B130. Note that the pixel P4 shown in FIG. 9B is an example of the pixel at the position of the lesion L4. The XY coordinates of pixel P4 are the same as those of pixel P3 (x3, y3). A pixel P5 shown in FIG. 9A is an example of the pixel at the position of the lesion L5. Hereinafter, the XY coordinates of pixel P5 are assumed to be (x5, y5).

The image processing function 343 then determines weights for each pixel in multiple slices based on the location and type of lesions detected by the detection function 342 . Weighting processing based on the position and type of lesion will be described below with reference to FIG. FIG. 10 is a diagram illustrating an example of weighting processing according to the first embodiment.

For example, for the XY coordinates where one lesion is detected, the image processing function 343 determines the weight of the pixels of the slice in which the lesion is detected to be "1" and the weight of the pixels of the other slices to be "0". do. For example, as shown in FIGS. 9A and 9B, only one lesion L5 is detected at XY coordinates (x5, y5). In this case, the image processing function 343 sets the weight of the pixel P5 (the pixel at the position of the lesion L5) to "1" and sets the weight of the pixel having the same XY coordinates as the pixel P5 in another slice to "0". decide. That is, as shown in FIG. 10, the image processing function 343 determines the weight of the pixel P5 whose XY coordinates are (x5, y5) in the slice B130 to be "1", and the other slices (slices B101 to B129, Then, the weight of the pixel whose XY coordinates are (x5, y5) in the slices B131 to B150) is set to "0". Also, the image processing function 343 determines the weight of the pixel in each slice for each of the XY coordinates included in the position of the lesion L5, similarly to the XY coordinates (x5, y5).

In addition, the image processing function 343 determines the pixel weight of each slice using the type of lesion for the XY coordinates where multiple lesions are detected. For example, as shown in FIGS. 9A and 9B, two lesions (lesion L3 and lesion L4) are detected at XY coordinates (x3, y3). In this case, the image processing function 343 weights the pixel P3 (the pixel at the position of the lesion L3) and the pixel P4 (the pixel at the position of the lesion L4) according to the type of lesion, thereby obtaining the image of each slice. Determine the pixel weight.

For example, the image processing function 343 acquires a ratio set for each combination of lesion types according to the types of lesion L3 and lesion L4. Such a ratio may be a preset value set in advance or may be set by the user. For example, the image processing function 343 acquires from the memory 33 the ratio of "architectural disturbance" to "tumor" of "4:1" according to the types of the lesions L3 and L4.

Then, the image processing function 343 adjusts the pixels of each slice so that the ratio of the weight of the pixel P3 at the position of "construction disorder" and the weight of the pixel P4 at the position of "tumor" is "4:1". give weight to For example, as shown in FIG. 10, the image processing function 343 assigns a weight of "0.8" to the pixel P3 whose XY coordinates are (x3, y3) in slice B120, and the XY coordinates are ( A weight of "0.2" is assigned to the pixel P4 that is x3, y3). That is, when the lesion L3, which is a "disturbance of architecture", is detected, the image processing function 343 assigns a weight to the pixel P3 at the position of "disturbance of architecture" and to the pixel at the position of the lesion L4, which is a "tumor". give a greater weight than

In addition, the image processing function 343 applies the weight " 0” is attached. That is, the image processing function 343 assigns weights according to the types of lesions to the pixels at the positions of the lesions L3 and L4, thereby determining weights for each pixel in a plurality of slices. Similarly, the image processing function 343 determines weights of pixels in each slice for each of the XY coordinates included in the positions of the lesions L3 and L4, similarly to the XY coordinates (x3, y3).

Then, the image processing function 343 generates a synthesized 2D image by synthesizing a plurality of slices included in the three-dimensional medical data based on the determined weights. For example, the image processing function 343 generates two-dimensional image data C102 (not shown) by executing weighted average processing based on the determined weight for each XY coordinate.

For example, the image processing function 343 performs weighted average processing using the pixel value of each pixel and the weight determined for each pixel for the XY coordinates (x3, y3). Specifically, the image processing function 343 calculates the product of the pixel value of the pixel P3 whose XY coordinates are (x3, y3) in the slice B120 and the weight "0.8", and the XY coordinates are ( x3, y3) and the weight "0.2", and the pixel value of the pixel with the XY coordinates of (x3, y3) and the weight "0" in another slice is calculated. Calculate the product. Then, the image processing function 343 sets the sum of the calculated products as the pixel value of the pixel whose XY coordinates are (x3, y3) in the two-dimensional image data C102. Also, the image processing function 343 determines the pixel values of the two-dimensional image data C102 for each of the XY coordinates included in the positions of the lesions L3 and L4 in the same manner as the XY coordinates (x3, y3).

The image processing function 343 also performs weighted average processing using the pixel value of each pixel and the weight determined for each pixel for the XY coordinates (x5, y5). The image processing function 343 also determines the pixel values of the two-dimensional image data C102 for each of the XY coordinates included in the position of the lesion L5, similarly to the XY coordinates (x5, y5). In addition, the image processing function 343 determines pixel values of the two-dimensional image data C102 by minimum intensity projection without determining weights for XY coordinates in which no lesion is detected in any slice. That is, the image processing function 343 executes weighted averaging processing based on the determined weight when pixels with determined weights exist in the slice direction, and when pixels with determined weights do not exist in the slice direction. generates two-dimensional image data C102 by executing minimum intensity projection processing.

As described above, the image processing function 343 determines the weight of each pixel in a plurality of slices based on the position and type of lesion, and synthesizes the plurality of slices based on the determined weights to obtain two-dimensional image data. Generate C102. As a result, the two-dimensional image data C102 appropriately represents each of the lesions even when multiple lesions overlap in the slice direction.

For example, in the case where “architectural disturbance” and “tumor” overlap in the slice direction, the image processing function 343 assigns weights to pixels at positions of “turbulence” and pixels at positions of “tumor”. 2D image data C102 is generated by assigning a greater weight than . Here, the "structural disturbance" is not so different in pixel values from other regions in the three-dimensional medical data, so that it is easily blurred when synthesizing a plurality of slices. On the other hand, the image processing function 343 can reduce the blur caused by the "disturbance of construction" in the two-dimensional image data C102 by assigning a large weight to the pixels in the position of "disturbance of construction". In addition, other types of lesions such as "tumor" and "calcification" are represented by pixel values lower than those of other regions in the three-dimensional medical data, and even if a small weight is assigned, the two-dimensional It can be sufficiently visually recognized in the image data C102. That is, the image processing function 343 generates the two-dimensional image data C102 based on the position and type of the lesion, thereby making it possible to visually recognize both the “structural disorder” and other types of lesions. can.

After the synthetic 2D image is generated by the image processing function 343 , the display control function 344 causes the display 32 to display the synthetic 2D image. Thereby, the display control function 344 can present the synthesized 2D image to the user of the medical image processing apparatus 30 . Alternatively, the control function 341 may output the generated synthetic 2D image to an external device. For example, the control function 341 transmits the composite 2D image to the X-ray diagnostic apparatus 10 via the network NW. In this case, the display control function 114b in the X-ray diagnostic apparatus 10 causes the display 113 to display the synthesized 2D image. Thereby, the display control function 114b can present a synthesized 2D image to the user of the X-ray diagnostic apparatus 10. FIG. Note that the gradation may be reversed when displaying the synthesized 2D image.

Next, an example of the procedure of processing by the medical image processing apparatus 30 will be described with reference to FIG. 11 . FIG. 11 is a flowchart for explaining a series of processing flows of the medical image processing apparatus 30 according to the first embodiment. Step S<b>101 is a step corresponding to the control function 341 . Step S<b>102 is a step corresponding to the detection function 342 . Steps S 103 , S 104 , S 105 , S 106 , S 107 , S 108 , S 109 , S 110 and S 111 are steps corresponding to the image processing function 343 . Note that FIG. 11 describes a case where a synthetic 2D image is generated based on the position and type of lesion.

First, the processing circuit 34 acquires three-dimensional medical data from the X-ray diagnostic apparatus 10 or the image storage apparatus 20 (step S101). Next, the processing circuitry 34 detects the positions and types of lesions in the breast P in a plurality of slices included in the three-dimensional medical data (step S102).

Next, the processing circuit 34 sets one of the XY coordinates included in the three-dimensional medical data (step S103), and determines whether a lesion exists in the slice direction (Z direction) at the set XY coordinates. Determine (step S104). That is, the processing circuit 34 determines whether or not there is a slice in which a lesion is detected at the set XY coordinates.

Here, if a lesion exists in the slice direction (Yes at step S104), the processing circuit 34 determines whether or not there are a plurality of lesions (step S105). If there are a plurality of lesions as in the XY coordinates (x3, y3) shown in FIGS. 9A and 9B (Yes in step S105), the processing circuit 34 calculates the ratio set for each combination of lesion types. Acquire (step S106).

If there are not a plurality of lesions as in the XY coordinates (x5, y5) shown in FIGS. 9A and 9B (No in step S105), or after step S106, the processing circuit 34 detects a plurality of lesions included in the three-dimensional medical data. A weight for each pixel in the slice is determined (step S106). Here, if there are multiple lesions, processing circuitry 34 uses the ratios obtained in step S106 to determine weights for each pixel in multiple slices.

Next, the processing circuit 34 performs weighted average processing based on the determined weights (step S108). Alternatively, if no lesion exists in the slice direction (No at step S104), the processing circuitry 34 executes minimum intensity projection processing (step S109). Step S108 or step S109 allows processing circuitry 34 to determine the pixel values of the pixels of the composite 2D image at the set XY coordinates.

Here, the processing circuit 34 determines whether or not all pixels in the three-dimensional medical data have been processed (step S110). If unprocessed pixels remain (No at step S110), the processing circuit 34 changes the setting of the XY coordinates (step S111), and proceeds to step S104 again. On the other hand, if all pixels have been processed (Yes at step S110), the processing circuitry 34 determines that the generation of the composite 2D image has been completed, and terminates the process.

As described above, according to the first embodiment, the detection function 342 detects the position of the lesion of the breast P in a plurality of slices included in the three-dimensional medical data obtained by tomosynthesis imaging the breast P of the subject. to detect The image processing function 343 also generates two-dimensional image data (composite 2D image) by combining a plurality of slices based on the detected lesion positions. Therefore, the medical image processing apparatus 30 according to the first embodiment can improve the visibility of lesions in a synthesized 2D image obtained by tomosynthesis imaging. As a result, the medical image processing apparatus 30 can improve the throughput of image interpretation and reduce the possibility of overlooking lesions.

Also, as described above, according to the first embodiment, the detection function 342 further detects lesion types in multiple slices. The image processing function 343 also uses the lesion type to generate a composite 2D image. Therefore, the medical image processing apparatus 30 according to the first embodiment can generate a composite 2D image that appropriately represents each lesion even when a plurality of lesions overlap in the slice direction.

Further, as described above, according to the first embodiment, when architectural disturbance is detected as a lesion, the image processing function 343 assigns the position of another type of lesion to the pixel at the position of the architectural disturbance. Give more weight than you give to pixels. Therefore, the medical image processing apparatus 30 according to the first embodiment can prevent architectural disturbances from being blurred during synthesis, and can improve the visibility of architectural disturbances in a synthesized 2D image.

Also, as described above, according to the first embodiment, the image processing function 343 generates a synthetic 2D image from 3D medical data. That is, the medical image processing apparatus 30 according to the first embodiment can acquire both three-dimensional image data and two-dimensional image data by one-time imaging. Therefore, the medical image processing apparatus 30 can reduce the number of times of imaging and reduce the exposure dose of the subject.

Note that the display control function 344 may display three-dimensional medical data in addition to the synthesized 2D image. For example, the display control function 344 causes the display 32 to display one of a plurality of slices in the three-dimensional medical data, or sequentially display a plurality of slices. Here, since the user has already grasped the position and type of the lesion from the synthesized 2D image, the three-dimensional medical data can be interpreted quickly. The user can then focus on a slice in which a lesion appears particularly clearly and perform more detailed observation.

(Second embodiment)
In the first embodiment described above, a case has been described in which the weight for each pixel in multiple slices is determined based on the position and type of lesion. On the other hand, in the second embodiment, a case will be described in which the weight for each pixel in a plurality of slices is determined based on the weight distribution in the slice direction of the three-dimensional medical data.

The medical image processing system 1 according to the second embodiment has the same configuration as the medical image processing system 1 according to the first embodiment shown in FIGS. part is different. The same reference numerals as in FIGS. 1 and 2 are assigned to the same configurations as those described in the first embodiment, and descriptions thereof are omitted.

First, the memory 33 stores weight distribution in the slice direction (Z direction) of three-dimensional medical data for each type of lesion. Here, the weight distribution stored in the memory 33 will be described with reference to FIG. FIG. 12 is a diagram illustrating an example of weight distribution according to the second embodiment.

The distribution of weights shown in FIG. 12 is preset and stored in memory 33 prior to generation of the synthetic 2D image. The setting of the weight distribution may be automatically performed by the medical image processing apparatus 30 or may be performed by the user via the input interface 109 . Alternatively, the medical image processing apparatus 30 may receive the weight distribution set by an external device and store it in the memory 33 .

The horizontal axis in FIG. 12 corresponds to the slice direction (Z direction). More specifically, the horizontal axis in FIG. 12 corresponds to the number of slices. Note that the horizontal axis in FIG. 12 may be the length in the slice direction instead of the number of slices. Also, the horizontal axis in FIG. 12 corresponds to the magnitude of the weight given to each pixel of a plurality of slices.

FIG. 12 shows three weight distributions set for three types of lesions: "calcification", "arrangement disorder" and "tumor". Here, the three weight distributions are set so that the spread in the slice direction becomes smaller in the order of "calcification", "arrangement disorder", and "tumor". That is, the memory 33 stores a distribution of weights set such that the smaller the size of the lesion, the greater the kurtosis. Also, the three weight distributions are set so that the total areas are approximately the same. For example, the weight distribution of "tumor" is set so that the peak value decreases as the spread in the slice direction increases. Note that although FIG. 12 shows a continuously set distribution, the memory 33 may store a discretely set distribution (array).

Processing circuitry 34 generates a composite 2D image by determining weights for each pixel in multiple slices using the distribution of weights shown in FIG. Specifically, first, the detection function 342 detects the position and type of lesion in multiple slices. Next, the image processing function 343 applies weight distribution according to the type of lesion to the positions of the lesions detected by the detection function 342 to determine weights for each pixel in a plurality of slices. Further, the image processing function 343 generates a composite 2D image by combining multiple slices based on the determined weights.

In the following, as an example, among the slices B101 to B150, a “tumor” lesion L1 is detected in the slice B120 shown in FIG. 6A, and a “calcification” lesion L2 is detected in the slice B130 shown in FIG. 6B. I will explain the case where In this case, the image processing function 343 applies the “tumor” weight distribution to the location of the lesion L1 in slice B120, and the “calcification” weight distribution to the location of the lesion L2 in slice B130. to determine the weight for each pixel in multiple slices.

Specifically, the image processing function 343 first reads from the memory 33 the weight distribution of “tumor” among the three weight distributions shown in FIG. Next, the image processing function 343 applies the “tumor” weight distribution to the pixel P1 at the location of the lesion L1 in the slice B120 so that (0) in the “tumor” weight distribution is aligned with the slice B120. do. That is, the image processing function 343 aligns (-2) with slice B118, (-1) with slice B119, (0) with slice B120, (+1) with slice B121, and (+2). Apply the "tumor" weight distribution to pixel P1 to fit slice B122.

By applying the "tumor" weight distribution to pixel P1, image processing function 343 determines the weights of pixel P1 in slice B120 and pixels with the same XY coordinates as pixel P1 in other slices. . For example, image processing function 343 determines the weight of each pixel having the same XY coordinates (x1, y1) as pixel P1, as shown in FIG. FIG. 13 is a diagram showing an example of weighting processing according to the second embodiment.

That is, as shown in FIG. 13, the image processing function 343 determines the weight of the pixel whose XY coordinates are (x1, y1) in the slice B118 to be "0.1", and in the slice B119, the XY coordinates are (x1, y1). y1) is determined to be "0.2", the weight of the pixel P1 whose XY coordinates are (x1, y1) in slice B120 is determined to be "0.4", and the weight of pixel P1 whose XY coordinates are (x1, y1) is determined to be "0.4" in slice B121. The weight of a pixel having (x1, y1) is determined to be "0.2", and the weight of a pixel having XY coordinates of (x1, y1) in slice B122 is determined to be "0.1". In addition, the image processing function 343 determines the weight of pixels whose XY coordinates are (x1, y1) in other slices (slices B101 to B117 and slices B123 to B150) to be "0". The image processing function 343 also determines the weight of the pixel in each slice for each of the XY coordinates included in the position of the lesion L1, in the same manner as for the XY coordinates (x1, y1).

Further, the image processing function 343 reads from the memory 33 the weight distribution of “calcification” among the three weight distributions shown in FIG. Next, the image processing function 343 calculates the weight distribution of "tumor" for the pixel P2 at the position of the lesion L2 in the slice B130 so that (0) in the weight distribution of "calcification" is aligned with the slice B130. Apply. That is, the image processing function 343 aligns (-2) with slice B128, (-1) with slice B129, (0) with slice B130, (+1) with slice B131, and (+2). Apply the distribution of "calcification" weights to pixel P2 to match slice B132. Accordingly, the image processing function 343 determines weights for pixel P2 in slice B130 and pixels having the same XY coordinates as pixel P2 in other slices.

That is, as shown in FIG. 13, the image processing function 343 determines the weight of the pixel whose XY coordinates are (x2, y2) in the slice B129 to be "0.1", and in the slice B130, the XY coordinates are (x2, y2). y2) is determined to be "0.8", and the weight of the pixel whose XY coordinates are (x2, y2) in slice B131 is determined to be "0.1". In addition, the image processing function 343 determines the weight of pixels whose XY coordinates are (x2, y2) in other slices (slices B101 to B128 and slices B132 to B150) to be "0". The image processing function 343 also determines the weight of the pixel in each slice for each of the XY coordinates included in the position of the lesion L2, similarly to the XY coordinates (x2, y2).

Then, the image processing function 343 generates a synthesized 2D image by synthesizing a plurality of slices included in the three-dimensional medical data based on the determined weights. For example, the image processing function 343 generates two-dimensional image data C201 (not shown) by executing weighted average processing based on the determined weight for each XY coordinate. For example, the image processing function 343 executes weighted averaging processing based on the determined weight when pixels with determined weights exist in the slice direction, and when pixels with determined weights do not exist in the slice direction. generates two-dimensional image data C201 by executing minimum intensity projection processing.

Next, a case where a plurality of lesions overlap in the slice direction will be described. As an example below, among the slices B101 to B150, a “tumor” lesion L6 is detected in the slice B120 shown in FIG. 14A, and a “calcification” lesion L7 is detected in the slice B130 shown in FIG. 14B. A case will be described. 14A and 14B are diagrams showing an example of lesion detection according to the second embodiment.

Pixels P6 and P7 shown in FIG. 14A are examples of pixels at the position of lesion L6. Hereinafter, let the XY coordinates of the pixel P6 be (x6, y6), and let the XY coordinates of the pixel P7 be (x7, y7). A pixel P8 shown in FIG. 14B is an example of the pixel at the position of the lesion L7. The XY coordinates of pixel P8 are the same as those of pixel P6 (x6, y6). A pixel P9 shown in FIG. 14B is a pixel having the same XY coordinates (x7, y7) as the pixel P7. As shown in FIG. 14B, pixel P9 is not included in the location of lesion L7.

As shown in FIGS. 14A and 14B, the detection function 342 detects lesions L6 and L7 at XY coordinates (x6, y6). That is, FIGS. 14A and 14B show a case where the detection function 342 detects multiple lesions at overlapping positions in the slice direction. In this case, the image processing function 343 calculates a composite distribution by, for example, synthesizing distributions of weights according to the types of the plurality of lesions according to the positions of the plurality of lesions, and calculates the composite distribution based on the calculated composite distribution. to determine the weight for each pixel in multiple slices.

Specifically, the image processing function 343 first reads from the memory 33 the weight distribution of “tumor” and the weight distribution of “calcification” among the three weight distributions shown in FIG. Next, the image processing function 343 adjusts the weight distribution of "tumor" to the slice B120 and the weight distribution of "calcification" to the slice B130. , and the distribution of the weights of , to calculate a composite distribution.

Note that when one lesion is detected over a plurality of slices, the image processing function 343 calculates a combined distribution in a state in which the distribution of weights is adjusted to the central slice among the plurality of slices, for example. For example, as shown in FIG. 15, when a lesion L6 that is a "tumor" is detected over slices B119, B120, and B121, the image processing function 343 assigns a weight of "tumor" to the central slice B120. Align the distribution. Then, the image processing function 343 matches the weight distribution of "tumor" to the slice B120 and the weight distribution of "calcification" to the slice B130, and calculates the distribution of the weight of "tumor" and the weight of "calcification". 15 is calculated by combining the distribution of the weights of . Note that FIG. 15 is a diagram showing an example of calculation of the composite distribution according to the second embodiment.

The image processing function 343 then determines a weight for each pixel in a plurality of slices based on the calculated composite distribution. For example, the image processing function 343 adjusts the pixel P6 at the position of the lesion L6 and the pixel P8 at the position of the lesion L7 so that the peak m1 in the composite distribution shown in the lower diagram of FIG. Apply the composite distribution to . Accordingly, the image processing function 343 determines weights for pixel P6 in slice B120, pixel P8 in slice B130, and pixels having the same XY coordinates as pixel P6 and pixel P8 in other slices.

For example, as shown in FIG. 16, the image processing function 343 determines the weight of a pixel whose XY coordinates are (x6, y6) in slice B118 to be "0.05", and in slice B119, its XY coordinates are (x6, y6). y6) is determined to be "0.1", the weight of the pixel whose XY coordinates are (x6, y6) in slice B120 is determined to be "0.2", and in slice B121 the XY coordinates are ( x6, y6) is determined to be "0.1", and the weight of the pixel whose XY coordinates are (x6, y6) in slice B122 is determined to be "0.05". Further, the image processing function 343 determines the weight of the pixel whose XY coordinates are (x6, y6) to be "0.05" in the slice B129, and the weight of the pixel whose XY coordinates are (x6, y6) in the slice B130. is set to "0.4", and the weight of the pixel whose XY coordinates are (x6, y6) in slice B131 is set to "0.05". In addition, the image processing function 343 sets the weight of pixels whose XY coordinates are (x6, y6) to "0" in other slices (slices B101 to B117, slices B123 to B128, and slices B132 to B150). to decide. Also, the image processing function 343 determines the weight of the pixel in each slice for each of the XY coordinates included in the positions of both the lesion L6 and the lesion L7, similarly to the XY coordinates (x6, y6). FIG. 16 is a diagram showing an example of weighting processing according to the second embodiment.

Also, as shown in FIGS. 14A and 14B, only one lesion (lesion L6, which is a “tumor”) is detected at the XY coordinates (x7, y7). In this case, the image processing function 343 applies the “tumor” weight distribution to the location of the lesion L6 detected by the detection function 342 to determine the weight for each pixel in multiple slices.

Specifically, the image processing function 343 reads out the “tumor” weight distribution from the memory 33 and applies the “tumor” weight distribution to the pixel P7 at the position of the lesion L6 in the slice B120. Accordingly, the image processing function 343 determines weights for pixel P7 in slice B120 and pixels having the same XY coordinates as pixel P7 in other slices. For example, image processing function 343 determines the weight of each pixel having the same XY coordinates (x7, y7) as pixel P7, as shown in FIG.

That is, the image processing function 343 determines the weight of the pixel whose XY coordinates are (x7, y7) to be "0.1" in the slice B118, and the weight of the pixel whose XY coordinates are (x7, y7) in the slice B119. is set to "0.2", the weight of the pixel P7 whose XY coordinates are (x7, y7) in slice B120 is set to "0.4", and the XY coordinates are (x7, y7) in slice B121. The weight of the pixel is determined to be "0.2", and the weight of the pixel whose XY coordinates are (x7, y7) in slice B122 is determined to be "0.1". In addition, the image processing function 343 determines the weight of pixels whose XY coordinates are (x7, y7) in other slices (slices B101 to B117 and slices B123 to B150) to be "0". In addition, the image processing function 343 calculates the weight of the pixel in each slice for each of the XY coordinates included in the position of the lesion L6 and not included in the position of the lesion L7, similarly to the XY coordinates (x7, y7). decide.

Then, the image processing function 343 generates a synthesized 2D image by synthesizing a plurality of slices included in the three-dimensional medical data based on the determined weights. For example, the image processing function 343 generates two-dimensional image data C202 (not shown) by executing weighted average processing based on the determined weight for each XY coordinate. For example, the image processing function 343 executes weighted averaging processing based on the determined weight when pixels with determined weights exist in the slice direction, and when pixels with determined weights do not exist in the slice direction. generates two-dimensional image data C202 by executing minimum intensity projection processing.

Next, an example of the procedure of processing by the medical image processing apparatus 30 will be described using FIG. FIG. 17 is a flowchart for explaining a series of processing flows of the medical image processing apparatus 30 according to the second embodiment. Step S<b>201 is a step corresponding to the control function 341 . Step S<b>202 is a step corresponding to the detection function 342 . Steps S 203 , S 204 , S 205 , S 206 , S 207 , S 208 , S 209 , S 210 , S 211 and S 212 are steps corresponding to the image processing function 343 .

First, the processing circuit 34 acquires three-dimensional medical data from the X-ray diagnostic apparatus 10 or the image storage apparatus 20 (step S201). Next, the processing circuitry 34 detects the positions and types of lesions in the breast P in a plurality of slices included in the three-dimensional medical data (step S202).

Next, the processing circuit 34 sets one of the XY coordinates included in the three-dimensional medical data (step S203), and determines whether a lesion exists in the slice direction (Z direction) at the set XY coordinates. Determine (step S204). That is, the processing circuit 34 determines whether or not there is a slice in which a lesion is detected at the set XY coordinates.

Here, if a lesion exists in the slice direction (Yes at step S204), the processing circuit 34 determines whether or not there are a plurality of lesions (step S205). When there are a plurality of lesions as in the XY coordinates (x6, y6) shown in FIGS. 14A and 14B (Yes in step S205), the processing circuit 34 calculates weight distributions corresponding to the types of each of the plurality of lesions. is read from the memory 33, and a composite distribution is calculated by combining the plurality of read distributions according to the positions of the plurality of lesions (step S206).

If there are not multiple lesions, such as the XY coordinates (x7, y7) shown in FIGS. 9A and 9B (No at step S205), or after step S206, processing circuitry 34 applies the distribution to multiple slices. (Step S207). Here, if the number of lesions is not plural, the processing circuit 34 applies the distribution according to the type of lesion to plural slices. On the other hand, if there are multiple lesions, the processing circuitry 34 applies the composite distribution calculated in step S206 to multiple slices. Processing circuitry 34 thereby determines a weight for each pixel in a plurality of slices (step S208).

Next, the processing circuit 34 executes weighted average processing based on the determined weights (step S209). Alternatively, if no lesion exists in the slice direction (No at step S204), the processing circuitry 34 executes minimum intensity projection processing (step S210). Step S209 or step S210 allows processing circuitry 34 to determine the pixel values of the pixels of the composite 2D image at the set XY coordinates.

Here, the processing circuit 34 determines whether or not all pixels in the three-dimensional medical data have been processed (step S211). If unprocessed pixels remain (No at step S211), the processing circuit 34 changes the setting of the XY coordinates (step S212), and proceeds to step S204 again. On the other hand, if all pixels have been processed (Yes at step S211), the processing circuit 34 determines that generation of the composite 2D image has been completed, and terminates the process.

As described above, according to the second embodiment, the memory 33 stores the weight distribution in the slice direction of the three-dimensional medical data for each lesion type. The detection function 342 also detects the location and type of lesion in multiple slices. In addition, the image processing function 343 determines the weight of each pixel in a plurality of slices by applying the weight distribution according to the type of lesion to the position of the lesion detected by the detection function 342, and determines the determined weight A composite 2D image is generated by combining multiple slices based on . Therefore, the medical image processing apparatus 30 according to the second embodiment can improve the visibility of lesions in a synthesized 2D image obtained by tomosynthesis imaging.

In addition, as described above, the memory 33 stores a distribution of weights set such that the smaller the size of the lesion, the greater the kurtosis. That is, the medical image processing apparatus 30 reflects information from a large number of slices in a synthesized 2D image for large-sized lesions such as tumors and architectural disturbances, and reflects a small number of slices for small-sized lesions such as calcifications. Information from the slices is reflected in the composite 2D image. Therefore, the medical image processing apparatus 30 according to the second embodiment can reflect lesion information contained in a plurality of slices in the synthesized 2D image in just the right amount, thereby improving the visibility of the lesion.

Further, as described above, when the detection function 342 detects a plurality of lesions at overlapping positions in the slice direction, the image processing function 343 calculates the distribution of weights according to the types of each of the plurality of lesions. A synthetic distribution is calculated by synthesizing according to the position, and a weight for each pixel in a plurality of slices is determined based on the synthetic distribution. Also, when the detection function 342 detects one lesion at an overlapping position in the slice direction, the image processing function 343 determines the weight for each pixel in a plurality of slices based on the distribution of weights according to the type of lesion. Therefore, the medical image processing apparatus 30 according to the second embodiment can generate a composite 2D image that appropriately represents each lesion even when a plurality of lesions overlap in the slice direction.

For example, in the XY coordinates (x6, y6) shown in FIGS. 14A and 14B, both the “tumor” lesion L6 and the “calcification” lesion L7 are weighted based on the composite distribution. be. Therefore, in the XY coordinates (x6, y6) of the synthesized 2D image (two-dimensional image data C202) to be generated, the "tumor" and the "calcification" are superimposed. Further, for example, in the XY coordinates (x7, y7), the lesion L6, which is a "tumor", is weighted based on the weight distribution of "tumor". Therefore, in the XY coordinates (x7, y7) of the two-dimensional image data C202, the "tumor" is expressed preferentially. That is, both "tumor" and "calcification" can be visually recognized in the two-dimensional image data C202.

(Third Embodiment)
In the first embodiment described above, a case has been described in which a synthetic 2D image is generated by synthesizing a plurality of slices included in three-dimensional medical data. On the other hand, in the third embodiment, a case will be described in which a synthetic 2D image is generated by performing image processing on a plurality of slices and synthesizing the slices after the image processing.

The medical image processing system 1 according to the third embodiment has the same configuration as the medical image processing system 1 according to the first embodiment shown in FIGS. part is different. The same reference numerals as in FIGS. 1 and 2 are assigned to the same configurations as those described in the first embodiment, and descriptions thereof are omitted.

First, the detection function 342 detects the position and type of lesion in multiple slices included in the three-dimensional medical data. As an example, a case where a lesion is detected in the slice B120 shown in FIG. 18 and the slice B130 shown in FIG. 18 among the slices B101 to B150 will be described below. Specifically, the detection function 342 detects a lesion L8, which is a "tumor", at the position shown in FIG. 18 in the slice B120. In addition, the detection function 342 detects a lesion L9 that is “calcification” and a lesion L10 that is “arrangement of architecture” at the positions shown in FIG. 18 in the slice B130. Note that FIG. 18 is a diagram for explaining generation of a synthesized 2D image according to the third embodiment.

Next, the image processing function 343 performs image processing according to the type of lesion on the pixels corresponding to the position of the lesion detected by the detection function 342 among the pixels in the plurality of slices. Note that the image processing function 343 may perform image processing only on pixels corresponding to the position of the lesion, or may perform image processing on pixels including pixels corresponding to the position of the lesion. .

Specifically, the image processing function 343 performs image processing T1 according to the lesion type “tumor” on the pixels corresponding to the position of the lesion L8 in the slice B120. Here, the image processing T1 is, for example, image processing that emphasizes low-frequency components. For example, the image processing function 343 performs filtering using a smoothing filter and processing for adding an emphasis component on the low frequency side as the image processing T1.

Further, the image processing function 343 performs image processing T2 corresponding to the type of lesion “calcification” on the pixels corresponding to the position of the lesion L9 in the slice B130. Here, the image processing T2 is, for example, image processing that emphasizes high frequency components. For example, the image processing function 343 performs filtering using an edge enhancement filter and processing for adding an enhancement component on the high frequency side as the image processing T2.

In addition, the image processing function 343 performs image processing T3 according to the type of lesion “construction disorder” on the pixels corresponding to the position of the lesion L10 in the slice B130. Here, the image processing T3 is, for example, image processing for enhancing contrast. For example, the image processing function 343 performs filtering using an edge enhancement filter and processing for extending the pixel value histogram of pixels corresponding to the position of the lesion L10 as the image processing T3.

The image processing function 343 then determines weights for each pixel in the multiple slices after image processing based on the location and type of the lesion. For example, the image processing function 343 determines the weight of the pixel of the slice in which the lesion is detected to be "1" for the XY coordinates in which one lesion is detected (such as the XY coordinates included in the position of the lesion L10). and determine the weights of pixels in other slices to be "0". In addition, the image processing function 343 uses the ratio set for each combination of lesion types for the XY coordinates at which multiple lesions are detected (such as the XY coordinates included in the position of the lesion L9) to obtain slices of each lesion. Weights are determined, and the weights of pixels in other slices are determined to be "0".

In another example, the image processing function 343 applies a distribution of weights according to the type of lesion to the locations of the lesions detected by the detection function 342 to determine weights for each pixel in multiple slices. do. Here, the image processing function 343 calculates a composite distribution by synthesizing distributions of weights corresponding to the types of each of the plurality of lesions for the XY coordinates at which a plurality of lesions are detected, and based on the synthesized distribution, Determine weights for each pixel in multiple slices.

Then, the image processing function 343 generates two-dimensional image data C301 (composite 2D image) shown in FIG. 18 by synthesizing a plurality of slices based on the determined weights. That is, the image processing function 343 performs image processing on slices B120 and B130 and other slices (slices B101 to B119, slices B121 to B129, and slices B131 to B150) based on the determined weights. , to generate two-dimensional image data C301. For example, the image processing function 343 executes weighted averaging processing based on the determined weight when pixels with determined weights exist in the slice direction, and when pixels with determined weights do not exist in the slice direction. generates two-dimensional image data C301 by executing minimum intensity projection processing.

Here, the visibility of each lesion is improved in the two-dimensional image data C301. For example, a region R3 in the two-dimensional image data C301 represents a "tumor" in which low-frequency components are emphasized. A region R4 in the two-dimensional image data C301 represents "calcification" in which high-frequency components are emphasized. Further, the region R5 in the two-dimensional image data C301 expresses the "disturbance of construction" in which the contrast is emphasized. That is, the medical image processing apparatus 30 can improve the visibility of lesions in the two-dimensional image data C301 obtained by tomosynthesis imaging.

Next, an example of the procedure of processing by the medical image processing apparatus 30 will be described using FIG. FIG. 19 is a flowchart for explaining a series of processing flows of the medical image processing apparatus 30 according to the third embodiment. Step S<b>301 is a step corresponding to the control function 341 . Step S302 is a step corresponding to the detection function 342. FIG. Steps S 303 , S 304 , S 305 , S 306 , S 307 , S 308 , S 309 , S 310 , S 311 , S 312 and S 313 are steps corresponding to the image processing function 343 . Note that FIG. 19 illustrates a case where the weight for each pixel is determined using the weight distribution in the slice direction as an example.

First, the processing circuit 34 acquires three-dimensional medical data from the X-ray diagnostic apparatus 10 or the image storage apparatus 20 (step S301). Next, the processing circuitry 34 detects the positions and types of lesions in the breast P in a plurality of slices included in the three-dimensional medical data (step S302). Next, the processing circuit 34 performs image processing according to the type of lesion on the pixels corresponding to the position of the detected lesion among the pixels in the plurality of slices (step S303).

Next, the processing circuit 34 sets one of the XY coordinates included in the three-dimensional medical data (step S304), and determines whether a lesion exists in the slice direction (Z direction) at the set XY coordinates. Determine (step S305). That is, the processing circuit 34 determines whether or not there is a slice in which a lesion is detected at the set XY coordinates.

Here, if there is a lesion in the slice direction (Yes at step S305), the processing circuit 34 determines whether there are a plurality of lesions (step S306). If there are a plurality of lesions (Yes at step S306), the processing circuit 34 reads the distribution of weights corresponding to the types of each of the plurality of lesions from the memory 33, and converts the read distributions to the weight distributions of the plurality of lesions. A synthetic distribution is calculated by synthesizing according to the position (step S307).

If there are not multiple lesions (No at step S306), or after step S307, processing circuitry 34 applies the distribution to multiple slices (step S308). Here, if the number of lesions is not plural, the processing circuit 34 applies the distribution according to the type of lesion to plural slices. On the other hand, if there are multiple lesions, the processing circuitry 34 applies the composite distribution calculated in step S307 to multiple slices. Accordingly, processing circuitry 34 determines a weight for each pixel in a plurality of slices (step S309).

Processing circuitry 34 then performs a weighted average process based on the determined weights (step S310). Alternatively, if no lesion exists in the slice direction (No at step S305), the processing circuitry 34 executes minimum intensity projection processing (step S311). Step S310 or step S311 allows processing circuitry 34 to determine the pixel values of the pixels of the composite 2D image at the set XY coordinates.

Here, the processing circuit 34 determines whether or not all pixels in the three-dimensional medical data have been processed (step S312). If unprocessed pixels remain (No at step S312), the processing circuit 34 changes the XY coordinate settings (step S313), and proceeds to step S305 again. On the other hand, if all pixels have been processed (Yes at step S312), the processing circuit 34 determines that generation of the composite 2D image has been completed, and terminates the process.

As described above, according to the third embodiment, the detection function 342 detects the location and type of lesion in multiple slices included in the 3D medical data. Further, the image processing function 343 performs image processing according to the type of lesion on pixels corresponding to the position of the lesion detected by the detection function 342 among pixels in a plurality of slices. In addition, the image processing function 343 determines a weight for each pixel in a plurality of image-processed slices based on the position and type of the lesion, and synthesizes the plurality of slices based on the determined weights to obtain a synthesized 2D image. to generate Therefore, the medical image processing apparatus 30 according to the third embodiment can improve the visibility of lesions in a synthesized 2D image obtained by tomosynthesis imaging.

Further, according to the third embodiment, when architectural disturbance is detected as a lesion, the image processing function 343 applies contrast to pixels corresponding to the position of architectural disturbance among pixels in a plurality of slices. Perform image processing to enhance. This allows the image processing function 343 to represent architectural disturbances with high contrast even in composite 2D images based on multiple slices. Therefore, the medical image processing apparatus 30 can improve the visibility of structural disturbances in the synthesized 2D image.

(Fourth embodiment)
Now, although the first to third embodiments have been described so far, various different modes other than the above-described embodiments may be implemented.

For example, the medical image processing apparatus 30 may generate multiple composite 2D images from one piece of three-dimensional medical data. In one example, detection function 342 detects lesion L3 in slice B120, and lesions L4 and L5 in slice B130, as shown in FIGS. 9A and 9B. That is, the detection function 342 detects a plurality of lesions (lesion L3 and lesion L4) at overlapping positions in the slice direction. Here, the image processing function 343 generates a composite 2D image for each of multiple lesions.

For example, the image processing function 343 generates a composite 2D image with improved visibility of the lesion L3 and a composite 2D image with improved visibility of the lesion L4. In one example, image processing function 343 first determines weights for each pixel in multiple slices such that pixels at lesion L3 are weighted more than pixels at lesion L4. do. Then, the image processing function 343 generates two-dimensional image data C401 by synthesizing a plurality of slices based on the determined weights. Here, the two-dimensional image data C401 preferentially represents the lesion L3. That is, the visibility of the lesion L3 is improved in the two-dimensional image data C401.

The image processing function 343 also determines weights for each pixel in the multiple slices such that pixels at the location of lesion L4 are weighted more heavily than pixels at the location of lesion L3. Then, the image processing function 343 generates two-dimensional image data C402 by synthesizing a plurality of slices based on the determined weights. Here, the two-dimensional image data C402 preferentially represents the lesion L4. That is, the visibility of the lesion L4 is improved in the two-dimensional image data C402.

The display control function 344 then displays the generated two-dimensional image data. For example, the display control function 344 displays the two-dimensional image data C401 and the two-dimensional image data C402 side by side on the display 32 . Further, for example, the display control function 344 switches and displays the two-dimensional image data C401 and the two-dimensional image data C402 according to the input operation by the user.

Also, in the above-described embodiment, the case where the processing circuit 34 in the medical image processing apparatus 30 generates a composite 2D image has been described. However, embodiments are not so limited. For example, processing circuitry 114 in X-ray diagnostic apparatus 10 may generate a composite 2D image.

For example, as shown in FIG. 20, the processing circuit 114 further has a detection function 114c and an image processing function 114d in addition to the control function 114a and the display control function 114b. Here, the detection function 114 c is a function corresponding to the detection function 342 . The image processing function 114 d is a function corresponding to the image processing function 343 . Note that FIG. 20 is a block diagram showing an example of the processing circuit 114 according to the fourth embodiment.

In the case shown in FIG. 20, the control function 114a first acquires three-dimensional medical data by performing tomosynthesis imaging on the breast P of the subject. Next, the detection function 114c detects the positions of lesions in the breast P in multiple slices included in the three-dimensional medical data. The image processing function 114d then generates a composite 2D image by combining multiple slices based on the locations of the detected lesions. For example, the image processing function 114d determines weights for each pixel in multiple slices based on the location of the lesion and combines the multiple slices based on the determined weights to generate a composite 2D image.

In addition, the detection function 114c may further detect lesion types in multiple slices. In this case, the image processing function 114d determines the weight of each pixel in a plurality of slices by, for example, assigning weights according to the types of lesions to the positions of lesions detected by the detection function 114c. The image processing function 114d then generates a synthesized 2D image by synthesizing the multiple slices based on the determined weights.

Alternatively, the image processing function 114d may use the weight distribution in the slice direction of the three-dimensional medical data to determine the weight for each pixel in multiple slices. Specifically, the memory 112 stores the weight distribution in the slice direction of the three-dimensional medical data. In addition, the image processing function 114d determines the weight of each pixel in a plurality of slices by applying weight distribution according to the type of lesion to the position of the lesion detected by the detection function 114c. The image processing function 114d then generates a synthesized 2D image by synthesizing the multiple slices based on the determined weights.

Further, the image processing function 114d may perform image processing according to the type of lesion on pixels corresponding to the position of the lesion detected by the detection function 114c among pixels in a plurality of slices. In this case, the image processing function 114d determines a weight for each pixel in the plurality of image-processed slices based on the position and type of the lesion, and combines the plurality of slices based on the determined weights to obtain a synthesized 2D image. Generate an image.

Each component of each device according to the first to fourth embodiments is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to the illustrated one, and all or part of them can be functionally or physically distributed and integrated in arbitrary units according to various loads and usage conditions. Can be integrated and configured. Furthermore, all or any part of each processing function performed by each device can be implemented by a CPU and a program analyzed and executed by the CPU, or implemented as hardware based on wired logic.

Further, the medical image processing methods described in the first to fourth embodiments can be realized by executing a prepared medical image processing program on a computer such as a personal computer or workstation. This medical image processing program can be distributed via a network such as the Internet. In addition, this medical image processing program is recorded on a non-transitory computer-readable recording medium such as a hard disk, flexible disk (FD), CD-ROM, MO, DVD, etc., and is read from the recording medium by a computer. can also be run by

According to at least one embodiment described above, it is possible to improve the visibility of a lesion in a synthesized 2D image obtained by tomosynthesis imaging.

While several embodiments of the invention have been described, these embodiments have been presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, as well as the scope of the invention described in the claims and equivalents thereof.

1 medical image processing system 10 X-ray diagnostic apparatus 112 memory 114 processing circuit 114a control function 114b display control function 114c detection function 114d image processing function 30 medical image processing apparatus 33 memory 34 processing circuit 341 control function 342 detection function 343 image processing function 344 Display control function

Claims (7)

a detection unit that detects the position and type of the lesion in the breast in a plurality of tomographic images included in the three-dimensional medical data obtained by tomosynthesis imaging the breast of the subject;
an image processing unit that generates two-dimensional image data by synthesizing the plurality of tomographic images based on the detected position of the lesion;
a storage unit that stores a weight distribution in the slice direction of the three-dimensional medical data for each type of lesion;
The image processing unit determines a weight for each pixel in the plurality of tomographic images by applying the distribution corresponding to the type of the lesion to the position of the lesion detected by the detection unit, and determines the weight A medical image processing apparatus that generates the two-dimensional image data by synthesizing the plurality of tomographic images based on . When the detection unit detects a plurality of lesions at positions overlapping in the slice direction, the image processing unit converts the distribution corresponding to the type of each of the plurality of lesions to the position of each of the plurality of lesions. 2. The medical image processing apparatus according to claim 1 , wherein a synthetic distribution is calculated by synthesizing, and a weight for each pixel in the plurality of tomographic images is determined based on the synthetic distribution. 3. The medical image processing apparatus according to claim 1 , wherein said storage unit stores said distribution set such that kurtosis increases as said lesion size decreases. The image processing unit executes weighted averaging processing based on the determined weight when pixels with determined weights exist in the slice direction, and when pixels with determined weights do not exist in the slice direction, the minimum The medical image processing apparatus according to any one of claims 1 to 3 , wherein the two-dimensional image data is generated by executing value projection processing. further comprising a display control unit for displaying the two-dimensional image data,
The image processing unit generates the two-dimensional image data for each of the plurality of lesions when the detection unit detects a plurality of the lesions at overlapping positions in the slice direction,
The medical image processing apparatus according to any one of claims 1 to 4 , wherein said display control unit displays said plurality of generated two-dimensional image data. a control unit that collects three-dimensional medical data by performing tomosynthesis imaging on the breast of a subject;
a detection unit that detects the position of the lesion of the breast and the type of the lesion in a plurality of tomographic images included in the three-dimensional medical data;
an image processing unit that generates two-dimensional image data by synthesizing the plurality of tomographic images based on the detected position of the lesion;
a storage unit that stores a weight distribution in the slice direction of the three-dimensional medical data for each type of lesion;
The image processing unit determines a weight for each pixel in the plurality of tomographic images by applying the distribution corresponding to the type of the lesion to the position of the lesion detected by the detection unit, and determines the weight X-ray diagnostic apparatus that generates the two-dimensional image data by synthesizing the plurality of tomographic images based on . Detecting the position of a lesion in the breast and the type of the lesion in a plurality of tomographic images included in three-dimensional medical data obtained by tomosynthesis imaging the breast of a subject,
a distribution of weights in the slice direction of the three-dimensional medical data stored for each type of the lesion, wherein the distribution corresponding to the type of the lesion is applied to the position of the detected lesion, Two-dimensional image data is generated by determining a weight for each pixel in a plurality of tomographic images and synthesizing the plurality of tomographic images based on the weight.
A medical image processing program that causes a computer to execute each process.

Images