Medical image processing device, method for processing medical image, and x-ray ct apparatus

Abstract

To reduce an exposed dose.SOLUTION: A medical image processing device includes an acquisition part, an extraction part, a reconstruction part, and a derivation part. The acquisition part acquires first projection data obtained by non-contrast scanning a first imaging region of a subject and second projection data obtained by contrast scanning the first imaging region. The extraction part extract third projection data equivalent to a second imaging region of the first imaging region from the second projection data. The reconstruction part reconstructs a first reconstruction image on the basis of the first projection data, a second reconstruction image on the basis of the second projection data, and a third reconstruction image on the basis of a difference between the first projection data and the third projection data. The derivation part uses a model trained by training data that uses the first reconstruction image and the third reconstruction image as input data and uses the second reconstruction image as correct answer data to derive a second reconstruction image of a processing result from the first projection data and the third projection data that are processing objects.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The embodiments disclosed in this specification and the drawings relate to a medical image processing device, a medical image processing method, and an X-ray CT apparatus. [Background technology]

[0002] In order to evaluate the physiological functions of the subject's heart, the X-ray CT scanner performs dynamic imaging (perfusion CT imaging) by injecting a contrast agent into the subject. The X-ray CT scanner evaluates the blood flow rate of the myocardium or coronary artery by analyzing the time density curve (TDC) of the contrast agent that has reached the myocardium or coronary artery.

[0003] However, in dynamic imaging, the X-ray CT scanner continuously images the heart over multiple heartbeats, so the radiation dose is higher than in normal CT imaging. To reduce the radiation dose, methods such as low tube voltage imaging, dual energy imaging, iterative reconstruction, and deep learning-based reconstruction have been proposed. In addition to these methods, imaging or reconstruction methods with lower radiation doses are required. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2011-156402 A Summary of the Invention [Problem to be solved by the invention]

[0005] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to reduce radiation exposure dose. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described later can also be positioned as other problems. [Means for solving the problem]

[0006] A medical image processing apparatus according to an embodiment includes an acquisition unit, an extraction unit, a reconstruction unit, and a derivation unit. The acquisition unit acquires first projection data obtained by performing a non-contrast scan of a first imaging region of a subject, and second projection data obtained by performing a contrast scan of the first imaging region. The extraction unit extracts third projection data corresponding to a second imaging region of the first imaging region from the second projection data. The reconstruction unit reconstructs a first reconstructed image based on the first projection data, a second reconstructed image based on the second projection data, and a third reconstructed image based on a difference between the first projection data and the third projection data. The derivation unit uses the first reconstructed image and the third reconstructed image as input data, and derives the second reconstructed image as a processing result from the first projection data and the third projection data to be processed, using a model trained by training data in which the second reconstructed image is correct data. [Brief description of the drawings]

[0007] [Figure 1] 1 is a block diagram showing an example of the arrangement of a medical image processing system according to a first embodiment. [Diagram 2] 1 is a block diagram showing an example of the arrangement of a medical image processing apparatus according to a first embodiment. [Diagram 3] FIG. 2 is a block diagram showing an example of the configuration of an operation terminal according to the first embodiment. [Figure 4] 4 is a flowchart showing an example of training a model according to the first embodiment. [Diagram 5] FIG. 2 is a schematic diagram showing a training example of a model according to the first embodiment. [Figure 6] 5 is a flowchart showing an example of application of a model according to the first embodiment. [Figure 7] FIG. 4 is a schematic diagram showing an application example of the model according to the first embodiment. [Figure 8] 4 is a flowchart showing a reconstruction example of a reconstructed image according to the first embodiment. [Figure 9] FIG. 2 is a schematic diagram showing a reconstruction example of a reconstructed image according to the first embodiment. [Figure 10]FIG. 11 is a block diagram showing an example of the arrangement of an X-ray CT apparatus according to a second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, each embodiment will be described with reference to the drawings. In the following embodiments, parts with the same reference numerals perform similar operations, and duplicated descriptions will be omitted as appropriate.

[0009] (First embodiment) 1 is a block diagram showing an example of the configuration of a medical image processing system 100 according to the first embodiment. The medical image processing system 100 is a system that processes medical image data of a subject and the like. The medical image processing system 100 includes, as its components, a medical image processing device 1, multiple operation terminals 2, an X-ray CT device 3, and a medical image storage device 4. Each component is connected to each other so as to be able to communicate with each other via a bus (BUS) that is a common signal communication path.

[0010] The medical image processing device 1 is a device that processes medical image data of a subject, etc. The medical image processing device 1 functions as a server of a medical image processing system 100. The medical image processing device 1 may be a workstation capable of performing high-speed information processing. The medical image processing device 1 processes data transmitted from an operation terminal 2, an X-ray CT device 3, or a medical image storage device 4, and transmits the processed data to the operation terminal 2, the X-ray CT device 3, or the medical image storage device 4.

[0011] The operation terminal 2 is a terminal operated by a user. The operation terminal 2 functions as a client of the medical image processing system 100. The operation terminal 2 may be a desktop PC, a notebook PC, a smartphone, a tablet terminal, or a wearable terminal.

[0012] The X-ray CT device 3 is a device that performs X-ray CT imaging of a subject to generate a sinogram (an example of projection data) of the subject, CT image data (an example of medical image data), etc. The X-ray CT device 3 transmits the generated data to a medical image storage device 4 (see FIG. 10).

[0013] The medical image storage device 4 is a device that stores medical image data, etc. The medical image storage device 4 may be a storage medium (e.g., a magnetic storage medium, an electromagnetic storage medium, an optical storage medium, or a semiconductor memory) or a drive device that reads and writes data or information to and from the storage medium. The medical image storage device 4 communicates medical image data, etc. with the medical image processing device 1, the operation terminal 2, or the X-ray CT device 3.

[0014] 2 is a block diagram showing an example of the configuration of a medical image processing apparatus 1 according to the first embodiment. The medical image processing apparatus 1 includes, as its components, a processing circuit 11, a memory 12, and a communication IF 13. The components are communicatively connected to each other via a bus (BUS) which is a common signal communication path.

[0015] The processing circuit 11 is a circuit that controls the overall operation of the medical image processing device 1. The processing circuit 11 includes at least one processor. The processor refers to circuits such as a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)). When the processor is a CPU, the CPU realizes each function by reading and executing each program stored in the memory 12. When the processor is an ASIC, each function is directly incorporated as a logic circuit in the circuit of the ASIC. The processor may be configured as a single circuit, or may be configured by combining multiple independent circuits with each other. The processing circuitry 11 realizes each function (acquisition function 111, extraction function 112, reconstruction function 113, training function 114, derivation function 115, projection function 116, integration function 117, and system control function 118).

[0016] The acquisition function 111 is a function for acquiring various data or information. For example, the acquisition function 111 acquires first projection data P1 obtained by performing a non-contrast scan of a first imaging region A1 of a subject, and second projection data P2 obtained by performing a contrast scan of the first imaging region A1. The acquisition function 111 may acquire the first projection data P1 and third projection data P3 to be processed. The acquisition function 111 is an example of an acquisition unit.

[0017] The extraction function 112 is a function that extracts various data or information. For example, the extraction function 112 extracts third projection data P3 corresponding to the second imaging area A2 of the first imaging area A1 from the second projection data P2. The extraction function 112 may extract a difference between the first projection data P1 and the third projection data P3. The extraction function 112 is an example of an extraction unit.

[0018] The reconstruction function 113 is a function for reconstructing various data or information. For example, the reconstruction function 113 reconstructs a first reconstructed image R1 based on the first projection data P1, a second reconstructed image R2 based on the second projection data P2, and a third reconstructed image R3 based on the difference between the first projection data P1 and the third projection data P3. The reconstruction function 113 may reconstruct a fourth reconstructed image R4 corresponding to the second imaging area A2 based on the integrated projection data GP. The reconstruction function 113 is an example of a reconstruction unit.

[0019] The training function 114 is a function for training various models. For example, the training function 114 trains a model using training data T in which the first reconstructed image R1 and the third reconstructed image R3 are input data and the second reconstructed image R2 is correct answer data. The training function 114 is an example of a training unit.

[0020] The derivation function 115 is a function that derives various data or information. For example, the derivation function 115 derives a second reconstructed image R2 as a processing result from the first projection data P1 and the third projection data P3 to be processed, using a model trained by the training data T. The derivation function 115 is an example of a derivation unit.

[0021] The projection function 116 is a function that performs various projection processes. For example, the projection function 116 obtains estimated projection data EP by forward projecting the second reconstructed image R2 derived by the derivation function 115. The projection function 116 is an example of a projection unit.

[0022] The integration function 117 is a function that performs various integration processes. For example, the integration function 117 obtains integrated projection data GP by integrating the third projection data P3 to be processed with the estimated projection data EP. The integration function 117 is an example of an integration unit.

[0023] The system control function 118 is a function that controls various operations performed by the processing circuitry 11. For example, the system control function 118 provides an operating system (OS) for the processing circuitry 11 to realize each function (acquisition function 111, extraction function 112, reconstruction function 113, training function 114, derivation function 115, projection function 116, and integration function 117). The system control function 118 is an example of a system control unit.

[0024] The memory 12 is a device that stores various data or information. The memory 12 has a hardware configuration similar to that of the medical image storage device 4. The memory 12 stores each program that causes the processing circuit 11 to realize each function (acquisition function 111, extraction function 112, reconstruction function 113, training function 114, derivation function 115, projection function 116, integration function 117, and system control function 118). The memory 12 stores a model 120. The memory 12 is an example of a storage unit.

[0025] The model 120 may be any type of algorithm. The model 120 may be a machine learning model (e.g., regression, decision tree, random forest, support vector machine, naive Bayes, neural network). In particular, the model 120 may be a deep convolutional neural network (DCNN).

[0026] The communication IF 13 is an interface that communicates various data or information. For example, the communication IF 13 communicates various data or information with the operation terminal 2, the X-ray CT device 3, or the medical image storage device 4. The communication IF 13 is an example of a communication unit.

[0027] 3 is a block diagram showing an example of the configuration of the operation terminal 2 according to the first embodiment. The operation terminal 2 includes, as its components, a processing circuit 21, a memory 22, an input IF 23, a display IF 24, and a communication IF 25. Each component is communicatively connected to each other via a bus (BUS) which is a common signal communication path.

[0028] The processing circuitry 21 is a circuit that controls the overall operation of the operation terminal 2. The processing circuitry 21 of the operation terminal 2 has a hardware configuration similar to that of the processing circuitry 11 of the medical image processing device 1. The processing circuitry 21 realizes each function (a communication function 211, a display control function 212, and a system control function 213).

[0029] The communication function 211 is a function for communicating various data or information. For example, the communication function 211 communicates various data or information with the medical image processing device 1, the X-ray CT device 3, or the medical image storage device 4. The communication function 211 is an example of a communication unit.

[0030] The display control function 212 is a function for displaying various data or information. For example, the display control function 212 displays various data or information obtained by the communication function 211 on the display IF 24. The display control function 212 is an example of a display control unit.

[0031] The system control function 213 is a function that controls various operations performed by the processing circuitry 21. For example, the system control function 213 provides an operating system (OS) for the processing circuitry 21 to realize each function (communication function 211, display control function 212). The system control function 213 is an example of a system control unit.

[0032] The memory 22 is a device that stores various data or information. The memory 22 of the operation terminal 2 has a hardware configuration similar to that of the memory 12 of the medical image processing device 1. The memory 22 stores each program that causes the processing circuitry 21 to realize each function (communication function 211, display control function 212, system control function 213). The memory 22 is an example of a storage unit.

[0033] The input IF 23 is an interface that accepts various input operations from a user. The input IF 23 converts the various input operations accepted from the user into electrical signals and transmits the electrical signals to the processing circuit 21. The input IF 23 may be a mouse, a keyboard, a button, a panel switch, a slider switch, a trackball, an operation panel, or a touch panel. The input IF 23 is an example of an input unit.

[0034] The display IF 24 is an interface that displays various data or information. The display IF 24 may be a liquid crystal display, a plasma display, an organic EL display, or an LED display. The display IF 24 may be a touch panel display that also functions as the input IF 23. The display IF 24 is an example of a display unit.

[0035] The communication IF 25 is an interface that communicates various data or information. For example, the communication IF 25 communicates various data or information with the medical image processing device 1, the X-ray CT device 3, or the medical image storage device 4. The communication IF 25 is an example of a communication unit.

[0036] 4 is a flowchart showing an example of training the model 120 according to the first embodiment. According to this example, the medical image processing apparatus 1 trains the model 120 based on the first projection data P1 and the second projection data P2.

[0037] (Step S11) First, the medical image processing apparatus 1 acquires the first projection data P1 and the second projection data P2 by the acquisition function 111. Specifically, the acquisition function 111 accesses the medical image storage device 4 to acquire the first projection data P1 and the second projection data P2 from the medical image storage device 4.

[0038] (Step S12) Next, the medical image processing apparatus 1 extracts the third projection data P3 from the second projection data P2 by the extraction function 112. Specifically, the extraction function 112 extracts the third projection data P3, which is a part of the second projection data P2, from the second projection data P2 acquired in step S11.

[0039] (Step S13) Next, the medical image processing apparatus 1 extracts the difference between the first projection data P1 and the third projection data P3 by the extraction function 112. Specifically, the extraction function 112 extracts the difference between the first projection data P1 acquired in step S11 and the third projection data P3 extracted in step S12.

[0040] (Step S14) Next, the medical image processing apparatus 1 reconstructs a reconstruction image by the reconstruction function 113. First, the reconstruction function 113 reconstructs a first reconstruction image R1 based on the first projection data P1 acquired in step S11. Second, the reconstruction function 113 reconstructs a second reconstruction image R2 based on the second projection data P2 acquired in step S11. Third, the reconstruction function 113 reconstructs a third reconstruction image R3 based on the difference extracted in step S13.

[0041] (Step S15) Finally, the medical image processing apparatus 1 trains the model 120 by the training function 114. Specifically, the training function 114 generates training data T using the first reconstructed image R1 and the third reconstructed image R3 reconstructed in step S14 as input data and the second reconstructed image R2 reconstructed in step S14 as correct answer data. The training function 114 trains the model 120 by the generated training data T. After step S15, the medical image processing apparatus 1 ends a series of operations.

[0042] The medical image processing device 1 may repeatedly train the model 120 using a plurality of training data sets. That is, the medical image processing device 1 may sequentially generate training data sets including the first projection data P1 and the second projection data P2, and repeatedly train the model 120 using each of the generated training data sets. This optimizes the parameters of the model 120.

[0043] 5 is a schematic diagram showing a training example of the model 120 according to the first embodiment. According to this example, the medical image processing apparatus 1 trains a pre-training model 120A based on a non-contrast sinogram 510 (an example of the first projection data P1) and a contrast sinogram 520 (an example of the second projection data P2). The pre-training model 120A is the model 120 before being trained.

[0044] First, the medical image processing device 1 acquires the non-contrast sinogram 510 and the contrast sinogram 520 by the acquisition function 111. Specifically, the acquisition function 111 accesses the medical image storage device 4 to acquire the non-contrast sinogram 510 and the contrast sinogram 520 from the medical image storage device 4 (see step S11).

[0045] The non-contrast sinogram 510 is projection data obtained by a method of imaging the first imaging region A1 of the subject without injecting a contrast agent into the subject (non-contrast scan). The non-contrast sinogram 510 is obtained by the X-ray CT device 3 performing a simple scan for calcium score measurement on the body axial cross section including the heart of the subject.

[0046] The contrast sinogram 520 is projection data obtained by a method of injecting a contrast agent into the subject and imaging the first imaging region A1 of the subject (contrast scan). The contrast sinogram 520 is obtained by the X-ray CT device 3 performing dynamic imaging for measuring a time-density curve of the body axial cross section including the heart of the subject.

[0047] Next, the medical image processing apparatus 1 extracts, by the extraction function 112, a contrast sinogram 530 (an example of the third projection data P3) from the contrast sinogram 520. Specifically, the extraction function 112 extracts, from the contrast sinogram 520 related to the first photographing region A1, a contrast sinogram 530 related to the second photographing region A2, which is a part of the first photographing region A1 (see step S12).

[0048] The contrast sinogram 530 can be said to be projection data obtained by a method of injecting a contrast agent into the subject and imaging the second imaging region A2 of the subject (contrast scan). Alternatively, the contrast sinogram 530 can be said to be projection data obtained by the X-ray CT device 3 performing dynamic imaging for measuring a time-density curve of the subject's heart. The contrast sinogram 530 can also be said to be projection data obtained by truncating the contrast sinogram 520 in the channel direction (horizontal direction).

[0049] Next, the medical image processing device 1 uses the extraction function 112 to extract the difference between the non-contrast sinogram 510 and the contrast sinogram 530. Specifically, the extraction function 112 extracts the difference between the data region of the non-contrast sinogram 510 and the data region of the contrast sinogram 530 to extract the difference sinogram 540 (see step S13).

[0050] The difference sinogram 540 can be said to be projection data showing a part of the second imaging region A2 of the subject that is enhanced by the contrast agent, or the difference sinogram 540 can be said to be projection data in which the contrast effect of the heart of the subject is enhanced.

[0051] Next, the medical image processing device 1 reconstructs a non-contrast image 560, a contrast image 570, and a difference image 580 by the reconstruction function 113. First, the reconstruction function 113 reconstructs the non-contrast image 560 (an example of the first reconstructed image R1) based on the non-contrast sinogram 510. Second, the reconstruction function 113 reconstructs the contrast image 570 (an example of the second reconstructed image R2) based on the contrast sinogram 520. Third, the reconstruction function 113 reconstructs the difference image 580 (an example of the third reconstructed image R3) based on the difference sinogram 540 (see step S14).

[0052] Finally, the medical image processing apparatus 1 trains the pre-trained model 120A by the training function 114. Specifically, the training function 114 generates training data T using the non-contrast image 560 and the difference image 580 as input data and the contrast image 570 as correct answer data. The training function 114 trains the pre-trained model 120A in the framework of supervised learning using the generated training data T (see step S15).

[0053] 6 is a flowchart showing an application example of the model 120 according to the first embodiment. According to this example, the medical image processing apparatus 1 derives a second reconstructed image R2 as a processing result based on the first projection data P1 and the third projection data P3 to be processed.

[0054] (Step S21) First, the medical image processing device 1 acquires the first projection data P1 and the third projection data P3 to be processed by the acquisition function 111. Specifically, the acquisition function 111 accesses the X-ray CT device 3 to acquire the first projection data P1 and the third projection data P3 to be processed from the X-ray CT device 3.

[0055] (Step S22) Next, the medical image processing apparatus 1 extracts a difference between the first projection data P1 and the third projection data P3 to be processed by the extraction function 112. Specifically, the extraction function 112 extracts a difference between the first projection data P1 and the third projection data P3 acquired in step S21.

[0056] (Step S23) Next, the medical image processing apparatus 1 reconstructs a reconstruction image by the reconstruction function 113. First, the reconstruction function 113 reconstructs a first reconstruction image R1 based on the first projection data P1 to be processed acquired in step S21. Second, the reconstruction function 113 reconstructs a third reconstruction image R3 based on the difference extracted in step S22.

[0057] (Step S24) Next, the medical image processing apparatus 1 derives the second reconstructed image R2 of the processing result by the derivation function 115 using the trained model 120. Specifically, the derivation function 115 derives the second reconstructed image R2 of the processing result from the first reconstructed image R1 and the third reconstructed image R3 reconstructed in step S23 using the model 120 trained in step S15. After step S24, the process proceeds to step S25 (see FIG. 8).

[0058] 7 is a schematic diagram showing an application example of the model 120 according to the first embodiment. According to this example, the medical image processing apparatus 1 derives a contrast image 670 (an example of the second reconstructed image R2) as a processing result based on a non-contrast sinogram 610 (an example of the first projection data P1) and a contrast sinogram 620 (an example of the third projection data P3) to be processed and the post-training model 120B. The post-training model 120B is the model 120 after being trained.

[0059] First, the medical image processing device 1 acquires the non-contrast sinogram 610 and the contrast sinogram 620 by the acquisition function 111. Specifically, the acquisition function 111 accesses the X-ray CT device 3 to acquire the non-contrast sinogram 610 and the contrast sinogram 620 from the X-ray CT device 3 (see step S21).

[0060] The non-contrast sinogram 610 is projection data obtained by a method of imaging the first imaging region A1 of the subject without injecting a contrast agent into the subject (non-contrast scan). In other words, the non-contrast sinogram 610 is similar to the non-contrast sinogram 510.

[0061] The contrast sinogram 620 is projection data obtained by a method of injecting a contrast agent into the subject and imaging the second imaging region A2 of the subject (contrast scan). The contrast sinogram 620 is obtained by the X-ray CT device 3 performing dynamic imaging for measuring a time-density curve of the subject's heart. The difference is that the contrast sinograms 620 and 530 are projection data related to the second imaging region A2, and the contrast sinogram 520 is projection data related to the first imaging region A1.

[0062] Next, the medical image processing apparatus 1 uses the extraction function 112 to extract the difference between the non-contrast sinogram 610 and the contrast sinogram 620. Specifically, the extraction function 112 extracts the difference between the data region of the non-contrast sinogram 610 and the data region of the contrast sinogram 620 to extract the difference sinogram 630 (see step S22).

[0063] The difference sinogram 630 can be said to be projection data indicating a portion of the second imaging region A2 of the subject that has been enhanced by the contrast agent.

[0064] Next, the medical image processing device 1 reconstructs a non-contrast image 650 and a difference image 660 by the reconstruction function 113. First, the reconstruction function 113 reconstructs the non-contrast image 650 (an example of the first reconstructed image R1) based on the non-contrast sinogram 610. Second, the reconstruction function 113 reconstructs the difference image 660 (an example of the third reconstructed image R3) based on the difference sinogram 630 (see step S23).

[0065] Next, the medical image processing device 1 derives a contrast image 670 (an example of the second reconstructed image R2) as a processing result using the post-training model 120B by the derivation function 115. Specifically, the derivation function 115 inputs the non-contrast image 650 and the difference image 660 to the post-training model 120B to derive the contrast image 670 from the post-training model 120B (see step S24).

[0066] 8 is a flowchart showing a reconstruction example of a reconstructed image according to the first embodiment. According to this example, the medical image processing apparatus 1 reconstructs a fourth reconstructed image R4 based on the derived second reconstructed image R2.

[0067] (Step S25) Next, the medical image processing apparatus 1 forward projects the derived second reconstructed image R2 by the projection function 116. Specifically, the projection function 116 forward projects the second reconstructed image R2 derived in step S24 into the projection space to obtain estimated projection data EP.

[0068] (Step S26) Next, the medical image processing apparatus 1 integrates the third projection data P3 to be processed with the estimated projection data EP by the integration function 117. Specifically, the integration function 117 integrates the third projection data P3 to be processed acquired in step S21 with the estimated projection data EP obtained in step S25 to obtain integrated projection data GP.

[0069] (Step S27) Finally, the medical image processing apparatus 1 reconstructs a reconstructed image based on the integrated projection data GP by the reconstruction function 113. Specifically, the reconstruction function 113 reconstructs a fourth reconstructed image R4 corresponding to the second imaging area A2 based on the integrated projection data GP obtained in step S26. After step S27, the medical image processing apparatus 1 ends a series of operations.

[0070] 9 is a schematic diagram showing a reconstruction example of a reconstructed image according to the first embodiment. According to this example, the medical image processing apparatus 1 reconstructs an enlarged contrast image 900 (an example of the fourth reconstructed image R4) based on the contrast image 670 derived from the post-training model 120B.

[0071] First, the medical image-processing apparatus 1 forward projects the contrast image 670 using the projection function 116 to obtain an estimated contrast sinogram 700 (an example of estimated projection data EP) (see step S25).

[0072] The estimated contrast sinogram 700 has a data region 700A on one end side in the channel direction, a data region 700B on the center side, and a data region 700C on the other end side. The data regions 700A, 700B, and 700C show projection data related to the first imaging region A1. The data region 700B shows projection data related to the second imaging region A2.

[0073] Next, the medical image processing apparatus 1 uses the integration function 117 to integrate the estimated contrast sinogram 700 with the contrast sinogram 620 to obtain an integrated contrast sinogram 800 (an example of integrated projection data GP) (see step S26).

[0074] The contrast sinogram 620 has a data region 620B on the central side in the channel direction. The data region 620B indicates projection data relating to the second imaging region A2.

[0075] The integrated contrast sinogram 800 has a data region 800A on one end side in the channel direction, a data region 800B on the center side, and a data region 800C on the other end side. The data regions 800A, 800B, and 800C show projection data related to the first imaging region A1. The data region 800B shows projection data related to the second imaging region A2.

[0076] First, integration function 117 integrates contrast sinogram 620 with estimated contrast sinogram 700 such that the projection data in data region 800A approximates the projection data in data region 700A. Second, integration function 117 integrates contrast sinogram 620 with estimated contrast sinogram 700 such that the projection data in data region 800B approximates the projection data in data region 620B. Third, integration function 117 integrates contrast sinogram 620 with estimated contrast sinogram 700 such that the projection data in data region 800C approximates the projection data in data region 700C.

[0077] That is, the integrated contrast sinogram 800 may be projection data in which the data regions 800A and 800C are replaced with the data regions 700A and 700C, respectively, and the data region 800B is replaced with the data region 620B. Such projection data may also be referred to as "hybrid projection data."

[0078] The integration function 117 may apply feathering processing to the boundary between the data regions 800A and 800B, or between the data regions 800B and 800C. In the integrated contrast sinogram 800 after processing, the values ​​of the projection data vary smoothly (seamlessly) in the channel direction, so that the image quality of the enlarged contrast image 900 reconstructed from the integrated contrast sinogram 800 is improved.

[0079] Finally, the medical image processing apparatus 1 reconstructs, by the reconstruction function 113, an enlarged contrast image 900 corresponding to the second imaging region A2 (that is, the data region 800B) based on the integrated contrast sinogram 800 (see step S27).

[0080] The reconstruction function 113 may reconstruct a large field of view reconstructed image corresponding to the first imaging region A1 (i.e., data regions 800A, 800B, and 800C) based on the integrated contrast sinogram 800. The reconstruction function 113 may apply a masking process to a portion of the reconstructed image that does not correspond to the second imaging region A2 (i.e., data regions 800A and 800C). The reconstructed image after the process has a field of view similar to that of the enlarged contrast image 900.

[0081] According to the first embodiment described above, the medical image processing apparatus 1 inputs the contrast sinogram 620 obtained by contrast scanning of a small field of view related to the cardiac region of the subject to the post-training model 120B, thereby deriving a contrast image 670 of a large field of view including the cardiac region. At this time, the X-ray CT apparatus 3 only needs to perform a contrast scan limited to the cardiac region of the subject to obtain the contrast sinogram 620, and does not need to perform a contrast scan on an imaging region larger than the cardiac region. Therefore, the medical image processing apparatus 1 can indirectly reduce the radiation dose of the subject.

[0082] Furthermore, the medical image processing device 1 forward projects the contrast image 670 to obtain an estimated contrast sinogram 700. The medical image processing device 1 integrates the contrast sinogram 620 with the estimated contrast sinogram 700 to obtain an integrated contrast sinogram 800. The medical image processing device 1 reconstructs an enlarged contrast image 900 of the cardiac region of the subject based on the integrated contrast sinogram 800.

[0083] Here, the integrated contrast sinogram 800 includes actual measurement projection data of the subject's cardiac region derived from the contrast sinogram 620. Therefore, the degree of reproduction of the subject's cardiac region is higher in the enlarged contrast image 900 than in the contrast image 670. That is, the medical image processing apparatus 1 can provide a reconstructed image of the subject's cardiac region with a higher degree of reproduction.

[0084] In addition, the medical image processing device 1 does not directly reconstruct a reconstructed image from the contrast sinogram 620 truncated in the channel direction. In general, a reconstructed image based on truncated projection data has cupping artifacts and the like. Therefore, the medical image processing device 1 can reduce artifacts by using the non-contrast sinogram 610 obtained by a non-contrast scan of a large field of view including the cardiac region of the subject as prior information for reconstruction.

[0085] Second embodiment 10 is a block diagram showing an example of the configuration of an X-ray CT apparatus 3 according to the second embodiment. The X-ray CT apparatus 3 includes, as its respective modules, a gantry device 30, a bed device 40, and a console device 50. The respective modules are communicatively connected to each other via a bus (BUS) which is a common signal communication path.

[0086] The gantry device 30 is a module for performing X-ray CT imaging of the subject P. The gantry device 30 includes, as its components, an X-ray tube 31, a wedge 32, a collimator 33, a high-voltage device 34, a detector 35, a DAS 36, a rotating frame 37, and a control device 38. Each component is accommodated in a housing (not shown) in which a bore 39 is formed.

[0087] The X-ray tube 31 is a vacuum tube that irradiates the subject P with X-rays. The X-ray tube 31 generates X-rays by colliding thermal electrons from a cathode filament with an anode target using a high voltage and a filament current supplied from a high voltage device 34. The generated X-rays are irradiated onto the subject P through a wedge 32 and a collimator 33.

[0088] The wedge 32 is a filter that adjusts the amount of X-rays emitted from the X-ray tube 31. The wedge 32 has a concave lens shape that is thin in the center and thick on the periphery, and is made of aluminum or the like. The wedge 32 can be switched to a different wedge 32 under the control of the control device 38.

[0089] The collimator 33 is a window plate that limits the irradiation range of the X-rays irradiated from the X-ray tube 31. The collimator 33 has a slit in the center and is made of lead or the like. The slit of the collimator 33 is adjusted to an arbitrary size according to the control of the control device 38. The irradiation range of the X-rays is adjusted to an arbitrary size by the X-rays passing through the slit of the collimator 33.

[0090] The high voltage device 34 is a device that generates a high voltage and a filament current. The high voltage device 34 is a transformer type or an inverter type. The high voltage device 34 generates a high voltage and a filament current and supplies them to the X-ray tube 31 according to the control of the control device 38.

[0091] The detector 35 is a device that detects X-rays transmitted through the subject P. The detector 35 generates an electrical signal corresponding to the detected X-rays and transmits the generated electrical signal to the DAS 36. The detector 35 may have a one-dimensional array structure in which a plurality of detection elements are arranged in the channel direction (X-axis direction; row direction). Alternatively, the detector 35 may have a two-dimensional array structure in which a plurality of detection elements are arranged in the channel direction and slice direction (Z-axis direction; column direction).

[0092] The DAS 36 is a device that generates projection data (sinogram) based on an electrical signal transmitted from the detector 35. When the detector 35 is an energy integrating detector (EID), the DAS 36 generates projection data indicating an X-ray dose detected by the detector 35. When the detector 35 is a photon counting detector (PCD), the DAS 36 generates projection data indicating a count value of X-ray photons detected by the detector 35. The DAS 36 generates projection data by reading out an electrical signal from the detector 35 under the control of the control device 38, and transmits the generated projection data to the console device 50. The DAS 36 is an abbreviation for Data Acquisition System.

[0093] The rotating frame 37 is a frame that supports the X-ray tube 31, the wedge 32, the collimator 33, the high voltage device 34, the detector 35, and the DAS 36 rotatably about a rotation axis Z. The rotating frame 37 rotates about the rotation axis Z at a predetermined angular velocity in response to control by a control device 38.

[0094] The control device 38 is a device that controls the wedge 32, the collimator 33, the high voltage device 34, the DAS 36, and the rotating frame 37 in response to control by the console device 50. The control device 38 has a processing circuit and a driving mechanism such as a motor and an actuator.

[0095] The bed device 40 is a module that moves the subject P in any axial direction (X-axis direction, Y-axis direction, Z-axis direction). The bed device 40 is installed facing the front surface of the gantry device 30. The bed device 40 includes a base 41, a driving mechanism 42, a support frame 43, and a top plate 44 as its respective components.

[0096] The base 41 is a structure that supports the drive mechanism 42, the support frame 43, and the top plate 44. The base 41 is installed on the floor surface. The drive mechanism 42 is installed on the upper part of the base 41, the support frame 43 is installed on the upper part of the drive mechanism 42, and the top plate 44 is installed on the upper part of the support frame 43.

[0097] The drive mechanism 42 is a mechanism for moving the support frame 43 and the tabletop 44 in any axial direction. The drive mechanism 42 moves the support frame 43 and the tabletop 44 in accordance with the control of the control device 38 or the console device 50. The drive mechanism 42 has a motor such as a direct drive motor or a servo motor.

[0098] The support frame 43 is a structure that supports the top plate 44. The support frame 43 is made of a rigid body such as metal.

[0099] The top board 44 is a flat structure on which the subject P is placed. The top board 44 is made of an elastic material such as urethane.

[0100] The console device 50 is a module that controls the overall operation of the X-ray CT apparatus 3. The console device 50 includes, as its components, a processing circuit 51, a memory 52, an input IF 53, a display IF 54, and a communication IF 55. Each component is communicatively connected to each other via a bus (BUS) that is a common signal communication path.

[0101] The processing circuitry 51 is a circuit that controls the overall operation of the console device 50. The processing circuitry 51 has a hardware configuration similar to that of the processing circuitry 11 of the medical image processing device 1. The processing circuitry 51 realizes each function (imaging control function 511, preprocessing function 512, reconstruction function 513, display control function 514, and system control function 515).

[0102] The imaging control function 511 is a function for controlling X-ray CT imaging by the gantry device 30. For example, the imaging control function 511 controls the control device 38 of the gantry device 30 to control the X-ray tube 31, the wedge 32, the collimator 33, the high voltage device 34, the detector 35, the DAS 36, and the rotating frame 37. The imaging control function 511 is an example of an imaging control unit.

[0103] First, the imaging control function 511 may obtain the first projection data P1 to be processed by irradiating X-rays with a limited irradiation range to the first imaging region A1 of the subject P. Second, the imaging control function 511 may obtain the third projection data P3 to be processed by irradiating X-rays with a limited irradiation range to the second imaging region A2 of the subject P.

[0104] The imaging control function 511 may limit the irradiation range of the X-rays to the first imaging region A1 or the second imaging region A2 by adjusting the size of the slit of the collimator 33. In particular, the imaging control function 511 may limit the irradiation range of the X-rays to the second imaging region A2 by collimating the X-rays in the channel direction. The collimation can directly reduce the radiation dose of the subject P.

[0105] First, the imaging control function 511 performs a non-contrast scan for calcium score measurement on an imaging region including a cardiac region of the subject P. By the non-contrast scan, the imaging control function 511 can determine the position or range of the field of view. Furthermore, the imaging control function 511 can use the determined position or range of the field of view for a subsequent contrast scan.

[0106] Next, the imaging control function 511 performs a contrast scan for measuring a time-density curve on a small region based on the cardiac region of the subject P. As the contrast scan, the imaging control function 511 may perform (1) stressed myocardial perfusion imaging, (2) non-stressed coronary CT angiography, or (3) myocardial delayed contrast imaging.

[0107] The pre-processing function 512 is a function that performs various pre-processing. For example, the pre-processing function 512 performs pre-processing such as logarithmic conversion, offset correction, sensitivity correction, and beam hardening correction on the projection data transmitted from the DAS 36 to generate corrected projection data. The pre-processing function 512 is an example of a pre-processing unit.

[0108] The reconstruction function 513 is a function that performs various types of reconstruction. For example, the reconstruction function 513 performs reconstruction processing such as filtered back projection (FBP) and iterative reconstruction on the corrected projection data to generate CT image data. The reconstruction function 513 is an example of a reconstruction unit.

[0109] The display control function 514 is a function for displaying various data or information. For example, the display control function 514 displays a CT image based on CT image data on the display IF 54. The display control function 514 is an example of a display control unit.

[0110] The system control function 515 is a function that controls various operations performed by the processing circuitry 51. For example, the system control function 515 provides an operating system (OS) for the processing circuitry 51 to realize each function (imaging control function 511, preprocessing function 512, reconstruction function 513, and display control function 514). The system control function 515 is an example of a system control unit.

[0111] The processing circuitry 51 of the X-ray CT device 3 may realize each function (acquisition function 111, extraction function 112, reconstruction function 113, training function 114, derivation function 115, projection function 116, integration function 117, and system control function 118) similar to that of the processing circuitry 11 of the medical image processing device 1. In other words, the X-ray CT device 3 may execute the same operation as that of the medical image processing device 1.

[0112] The memory 52 is a device that stores various data or information. The memory 52 has a hardware configuration similar to that of the memory 12 of the medical image processing device 1. For example, the memory 52 stores projection data, CT image data, a control program, etc. The memory 52 is an example of a storage unit.

[0113] The input IF 53 is an interface that accepts various input operations from a user. The input IF 53 has a hardware configuration similar to that of the input IF 23 of the operation terminal 2. For example, the input IF 53 accepts input operations such as collection conditions of projection data and reconstruction conditions of CT image data. The input IF 53 is an example of an input unit.

[0114] The display IF 54 is an interface that displays various data or information. The display IF 54 has a hardware configuration similar to that of the display IF 24 of the operation terminal 2. For example, the display IF 54 displays a CT image, a GUI (Graphical User Interface), etc. The display IF 54 is an example of a display unit.

[0115] The communication IF 55 is an interface that communicates various data or information. For example, the communication IF 55 communicates various data or information between the medical image processing device 1, the operation terminal 2, or the medical image storage device 4. The communication IF 55 is an example of a communication unit.

[0116] According to the second embodiment described above, the X-ray CT apparatus 3 acquires the first projection data P1 and the third projection data P3 to be processed by adjusting the X-ray irradiation range. In particular, the X-ray CT apparatus 3 limits the X-ray irradiation range in the contrast scan to the second imaging area A2. Therefore, the X-ray CT apparatus 3 can reduce the radiation dose of the subject P compared to the case where the first imaging area A1 is irradiated with X-rays.

[0117] According to at least one of the embodiments described above, it is possible to reduce the radiation dose.

[0118] Although some embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope of the invention and its equivalents as described in the claims, as well as in the scope and spirit of the invention. [Explanation of symbols]

[0119] 1 Medical image processing device 2 Operation terminal 3 X-ray CT device 4 Medical image storage device 11,21,51 Processing circuit 12,22,52 Memory 13,25,55 Communication IF 23,53 Input IF 24,54 Display IF 30 Mounting device 31 X-ray tube 32 Wedge 33 Collimator 34 High Voltage Equipment 35 Detector 36 DAS 37 Rotating Frame 38 Control device 39 Bore 40 Sleeping device 41 Foundation 42 Drive mechanism 43 Support Frame 44 Tabletop 50 Console device 100 Medical image processing system 111 Acquisition Function 112 Extraction function 113,513 Reconfiguration function 114 Training Function 115 Derived Functions 116 Projection function 117 Integration Features 118,213,515 System Control Functions 120 Model 120A Pre-training model 120B Post-training model 211 Communication Functions 212,514 Display control functions 510,610 Non-contrast sinogram 511 Shooting control function 512 Pre-processing function 520,530,620 Contrast sinogram 540,630 Difference Sinogram 560,650 Non-contrast images 570,670 contrast images 580,660 difference images 620B, 700A, 700B, 700C, 800A, 800B, 800C Data area 700 Estimated contrast sinogram 800 Integrated Contrast Sinogram 900 Enlarged contrast images

Claims

1. an acquisition unit that acquires first projection data obtained by performing a non-contrast scan on a first imaging region of a subject and second projection data obtained by performing a contrast scan on the first imaging region; an extracting unit that extracts third projection data corresponding to a second imaging region of the first imaging region from the second projection data; a reconstruction unit configured to reconstruct a first reconstructed image based on the first projection data, a second reconstructed image based on the second projection data, and a third reconstructed image based on a difference between the first projection data and the third projection data; a derivation unit that uses a model trained with training data in which the first reconstructed image and the third reconstructed image are input data and the second reconstructed image is correct data, to derive the second reconstructed image as a processing result from the first projection data and the third projection data to be processed; A medical image processing device comprising:

2. a projection unit for obtaining estimated projection data by forward projecting the derived second reconstruction image; an integration unit that integrates the third projection data to be processed with the estimated projection data to obtain integrated projection data; Further comprising: The reconstruction unit reconstructs a fourth reconstruction image corresponding to the second imaging region based on the integrated projection data. The medical image processing device according to claim 1 .

3. A training unit for training the model using the training data. The medical image processing device according to claim 1 .

4. Obtaining first projection data obtained by performing a non-contrast scan of a first imaging region of a subject and second projection data obtained by performing a contrast scan of the first imaging region; extracting third projection data corresponding to a second imaging region of the first imaging region from the second projection data; reconstructing a first reconstructed image based on the first projection data, a second reconstructed image based on the second projection data, and a third reconstructed image based on a difference between the first projection data and the third projection data; deriving the second reconstructed image as a processing result from the first projection data and the third projection data to be processed using a model trained with training data in which the first reconstructed image and the third reconstructed image are input data and the second reconstructed image is correct data; A medical image processing method comprising:

5. an acquisition unit that acquires first projection data obtained by performing a non-contrast scan on a first imaging region of a subject and second projection data obtained by performing a contrast scan on the first imaging region; an extracting unit that extracts third projection data corresponding to a second imaging region of the first imaging region from the second projection data; a reconstruction unit configured to reconstruct a first reconstructed image based on the first projection data, a second reconstructed image based on the second projection data, and a third reconstructed image based on a difference between the first projection data and the third projection data; a derivation unit that uses a model trained with training data in which the first reconstructed image and the third reconstructed image are input data and the second reconstructed image is correct data, to derive the second reconstructed image as a processing result from the first projection data and the third projection data to be processed; An X-ray CT apparatus comprising:

6. an X-ray tube for irradiating the subject with X-rays; a collimator for limiting an irradiation range of the X-rays emitted from the X-ray tube; a detector that detects, as an electrical signal, X-rays that are irradiated from the X-ray tube through the collimator and transmitted through the subject; a data acquisition system for generating projection data based on the electrical signals; a control unit that controls the X-ray tube, the collimator, the detector, and the data acquisition system; Further comprising: the control unit obtains the first projection data of the processing target by irradiating X-rays to the first imaging region of the subject while limiting the irradiation range, and obtains the third projection data of the processing target by irradiating X-rays to the second imaging region of the subject while limiting the irradiation range.

6. The X-ray CT apparatus according to claim 5.

Images