JP2020096757A - Medical image processing device, medical image diagnostic device and image processing program - Google Patents

Abstract

To support diagnosis of circulatory disturbance.SOLUTION: A medical image processing device according to an embodiment includes: an acquisition unit which acquires first density change information indicating the density change of a contrast agent in a first period, and second density change information indicating the density change of the contrast agent in a second period different from the first period by executing scan using an X-ray with respect to a subject to which the contrast agent is injected once; and an estimation unit which estimates the presence/absence of the circulatory disturbance of the subject on the basis of the input data including the first density change information and the second density change information.SELECTED DRAWING: Figure 2

Description

Embodiments of the present invention relate to a medical image processing apparatus, a medical image diagnostic apparatus, and an image processing program.

Conventionally, a doctor has made a diagnosis of circulatory disorder. Examples of circulatory disorders include peripheral circulatory disorders in which blood does not normally circulate to the ends of the body such as limbs (that is, the whole body), and pulmonary circulatory disorders in which blood does not normally circulate in the lungs.

Diagnosis of peripheral circulatory disorders is performed, for example, by a doctor's examination of the color of the skin (e.g., patient) of a subject (e.g., patient) and the condition of cyanosis. However, for example, whether the skin color of the subject is blue is a subjective diagnosis of the doctor, and therefore depends on the experience and skill of the doctor. Moreover, chest CT (Computed Tomography) imaging, chest MR (Magnetic Resonance) imaging, and the like are used to diagnose pulmonary circulatory disorders, but it is difficult to diagnose an initial pulmonary circulatory disorder.

International Publication No. 2007/078012 JP 2007-284362A Japanese Patent Publication No. 2016-511032

The problem to be solved by the present invention is to support the diagnosis of circulatory disorders.

The medical image processing apparatus according to the embodiment performs a scan using X-rays on a subject into which the contrast agent has been injected once, thereby representing a change in the concentration of the contrast agent in the first period. Of the concentration change information and a second concentration change information indicating a concentration change of the contrast agent in a second period different from the first period, the first concentration change information and the acquisition unit. An estimation unit that estimates the presence or absence of a circulatory disorder in the subject based on the input data including the second concentration change information.

FIG. 1 is a diagram showing an example of the configuration of a medical image processing system according to the first embodiment. FIG. 2 is a diagram showing an example of the configuration of the X-ray CT apparatus according to the first embodiment. FIG. 3 is a flowchart showing a procedure of processing by the medical image processing system according to the first embodiment. FIG. 4 is a diagram for explaining an example of the ROI in the first embodiment, and is a diagram showing the configuration of the heart. FIG. 5 is a diagram for explaining an example of TDC in the first embodiment. FIG. 6 is a diagram for explaining an example of learning using AI in the first embodiment. FIG. 7 is a diagram for explaining an example of learning using AI in the first embodiment. FIG. 8 is a diagram for explaining an example of machine learning such as clustering according to the first embodiment. FIG. 9 is a diagram for explaining an example of the ROI in the second embodiment, and is a diagram showing the configuration of the heart. FIG. 10 is a diagram for explaining an example of TDC (RV) and TDC (LV) in the second embodiment. FIG. 11 is a diagram for explaining an example of learning using AI in the second embodiment. FIG. 12 is a diagram for explaining an example of learning using AI in the second embodiment. FIG. 13 is a diagram for explaining an example of machine learning such as clustering according to the second embodiment. FIG. 14 is a diagram for explaining an example of learning using AI in the modified examples of the first embodiment and the second embodiment. FIG. 15 is a diagram for explaining an example of learning using AI in the modified examples of the first embodiment and the second embodiment. FIG. 16 is a diagram showing an example of the configuration of the medical image processing apparatus according to the third embodiment.

Hereinafter, embodiments of a medical image processing apparatus and a medical image diagnostic apparatus will be described in detail with reference to the drawings. The medical image processing apparatus and the medical image diagnostic apparatus according to the present application are not limited to the embodiments described below.

(First embodiment)
First, the first embodiment will be described. In the first embodiment, an example in which the medical image diagnostic apparatus is applied to an X-ray computed tomography (CT) apparatus will be described. Hereinafter, a medical image processing system including an X-ray CT apparatus and a medical image processing apparatus will be described as an example.

FIG. 1 is a diagram showing an example of the configuration of a medical image processing system according to the first embodiment. As shown in FIG. 1, a medical image processing system 100 according to the first embodiment includes an X-ray CT (Computed Tomography) device 1, an image storage device 2, and a medical image processing device 3, and a network 200. Connected to each other via. The example illustrated in FIG. 1 is merely an example, and may be a case where various other devices (for example, a terminal device) are connected to the network 200.

For example, the X-ray CT apparatus 1 according to the first embodiment is connected to the image storage apparatus 2 and the medical image processing apparatus 3 via the network 200, as shown in FIG. The medical image processing system 100 is further connected to another medical image diagnostic apparatus such as an MRI (Magnetic Resonance Imaging) apparatus, an ultrasonic diagnostic apparatus and a PET (Positron Emission Tomography) apparatus, and a terminal apparatus via the network 200. May be connected. The network 200 may be a local network closed in the hospital or a network via the Internet.

The image storage device 2 stores image data collected by various medical image diagnostic devices. For example, the image storage device 2 is realized by a computer device such as a server device. In the present embodiment, the image storage device 2 acquires projection data or CT image data from the X-ray CT device 1 via the network 200, and the acquired projection data or CT image data is provided inside or outside the device. Store in the memory circuit.

The medical image processing apparatus 3 acquires image data from various medical image diagnostic apparatuses via the network 200 and processes the acquired image data. For example, the medical image processing apparatus 3 is realized by computer equipment such as a workstation. In the present embodiment, the medical image processing apparatus 3 acquires projection data and CT image data from the X-ray CT apparatus 1 or the image storage apparatus 2 via the network 200, and performs various kinds of processing on the acquired projection data and CT image data. Perform image processing. The medical image processing apparatus 3 displays CT image data before or after performing image processing on a display or the like.

The X-ray CT apparatus 1 collects CT image data of a subject. Specifically, the X-ray CT apparatus 1 swivels the X-ray tube and the X-ray detector about the subject to detect the X-rays transmitted through the subject and collects projection data. Then, the X-ray CT apparatus 1 generates CT image data based on the collected projection data.

FIG. 2 is a diagram showing an example of the configuration of the X-ray CT apparatus 1 according to the first embodiment. As shown in FIG. 2, the X-ray CT apparatus 1 according to the first embodiment includes a gantry device 10, a bed device 30, and a console device 40.

Here, in FIG. 2, the longitudinal direction of the rotary shaft of the rotary frame 13 or the top plate 33 of the bed apparatus 30 in the non-tilted state is the Z-axis direction. Further, an axial direction that is orthogonal to the Z-axis direction and is horizontal to the floor surface is the X-axis direction. Further, the Y-axis direction is defined as an axial direction that is orthogonal to the Z-axis direction and is perpendicular to the floor surface. 2 shows the gantry device 10 drawn from a plurality of directions for the sake of explanation, and shows a case where the X-ray CT apparatus 1 has one gantry device 10.

The gantry device 10 includes an X-ray tube 11, an X-ray detector 12, a rotating frame 13, an X-ray high voltage device 14, a controller 15, a wedge 16, a collimator 17, and a DAS (Data Acquisition System). 18 and.

The X-ray tube 11 is a vacuum tube having a cathode (filament) that generates thermoelectrons and an anode (target) that generates X-rays upon receiving collision of thermoelectrons. The X-ray tube 11 radiates thermoelectrons from the cathode to the anode by applying a high voltage from the X-ray high-voltage device 14 to generate X-rays for irradiating the subject P. For example, the X-ray tube 11 includes a rotary anode type X-ray tube that generates X-rays by irradiating a rotating anode with thermoelectrons.

The X-ray detector 12 has a plurality of detection elements that detect X-rays. Each detection element in the X-ray detector 12 detects the X-rays emitted from the X-ray tube 11 and passed through the subject P, and outputs a signal corresponding to the detected X-ray dose to the DAS 18. The X-ray detector 12 has, for example, a plurality of detection element arrays in which a plurality of detection elements are arranged in the channel direction (channel direction) along one arc centered on the focal point of the X-ray tube 11. The X-ray detector 12 has, for example, a structure in which a plurality of detection element rows in which a plurality of detection elements are arranged in the channel direction are arranged in the row direction (slice direction, row direction).

For example, the X-ray detector 12 is an indirect conversion type detector having a grid, a scintillator array, and a photosensor array. The scintillator array has a plurality of scintillators. The scintillator has a scintillator crystal that outputs a photon amount of light according to the incident X-ray dose. The grid has an X-ray shield plate arranged on the X-ray incident side surface of the scintillator array and absorbing scattered X-rays. The grid may be called a collimator (one-dimensional collimator or two-dimensional collimator). The optical sensor array has a function of converting into an electric signal according to the amount of light from the scintillator, and has, for example, an optical sensor such as a photodiode. The X-ray detector 12 may be a direct conversion type detector having a semiconductor element that converts an incident X-ray into an electric signal.

The rotating frame 13 is an annular frame that supports the X-ray tube 11 and the X-ray detector 12 so as to face each other, and causes the controller 15 to rotate the X-ray tube 11 and the X-ray detector 12. For example, the rotating frame 13 is a casting made of aluminum. In addition to the X-ray tube 11 and the X-ray detector 12, the rotating frame 13 can further support the X-ray high-voltage device 14, the wedge 16, the collimator 17, the DAS 18, and the like. Further, the rotating frame 13 can further support various configurations not shown in FIG.

The X-ray high-voltage device 14 has an electric circuit such as a transformer and a rectifier, and a high-voltage generator that generates a high voltage to be applied to the X-ray tube 11, and an X-ray generated by the X-ray tube 11. And an X-ray controller for controlling the output voltage according to the above. The high voltage generator may be of a transformer type or an inverter type. The X-ray high voltage device 14 may be provided on the rotating frame 13 or a fixed frame (not shown).

The control device 15 has a processing circuit having a CPU (Central Processing Unit) and the like, and a drive mechanism such as a motor and an actuator. The control device 15 receives an input signal from the input interface 43 and controls the operation of the gantry device 10 and the bed device 30. For example, the control device 15 controls the rotation of the rotary frame 13, the tilt of the gantry device 10, the operations of the bed device 30 and the top plate 33, and the like. As an example, the control device 15 rotates the rotating frame 13 about an axis parallel to the X-axis direction based on the input tilt angle (tilt angle) information as control for tilting the gantry device 10. The control device 15 may be provided in the gantry device 10 or the console device 40.

The wedge 16 is a filter for adjusting the X-ray dose emitted from the X-ray tube 11. Specifically, the wedge 16 transmits and attenuates the X-rays emitted from the X-ray tube 11 so that the X-rays emitted from the X-ray tube 11 to the subject P have a predetermined distribution. It is a filter to do. For example, the wedge 16 is a wedge filter or a bow-tie filter, and is a filter obtained by processing aluminum or the like so as to have a predetermined target angle and a predetermined thickness.

The collimator 17 is a lead plate or the like for narrowing down the irradiation range of the X-rays that have passed through the wedge 16, and a slit is formed by a combination of a plurality of lead plates or the like. The collimator 17 may be called an X-ray diaphragm. 2 shows the case where the wedge 16 is arranged between the X-ray tube 11 and the collimator 17, the case where the collimator 17 is arranged between the X-ray tube 11 and the wedge 16. Good. In this case, the wedge 16 transmits the X-rays which are emitted from the X-ray tube 11 and whose irradiation range is limited by the collimator 17, and attenuate the X-rays.

The DAS 18 collects X-ray signals detected by the respective detection elements included in the X-ray detector 12. For example, the DAS 18 includes an amplifier that amplifies an electric signal output from each detection element and an A/D converter that converts the electric signal into a digital signal, and generates detection data. The DAS 18 is realized by, for example, a processor.

The data generated by the DAS 18 is transmitted from a transmitter having a light emitting diode (LED) provided on the rotating frame 13 by optical communication to a non-rotating portion of the gantry device 10 (for example, a fixed frame. In FIG. 1). (Not shown in the figure), and is transmitted to the console device 40. Here, the non-rotating portion is, for example, a fixed frame or the like that rotatably supports the rotating frame 13. The method of transmitting data from the rotating frame 13 to the non-rotating portion of the gantry device 10 is not limited to optical communication, and any non-contact data transmission method may be adopted. You can use it.

The bed device 30 is a device for placing and moving the subject P to be imaged, and includes a base 31, a bed driving device 32, a top plate 33, and a support frame 34. The base 31 is a housing that supports the support frame 34 so as to be movable in the vertical direction. The bed driving device 32 is a drive mechanism that moves the top plate 33 on which the subject P is placed in the long axis direction of the top plate 33, and includes a motor, an actuator, and the like. The top plate 33 provided on the upper surface of the support frame 34 is a plate on which the subject P is placed. The bed driving device 32 may move the support frame 34 in the long axis direction of the top plate 33 in addition to the top plate 33.

The console device 40 has a memory 41, a display 42, an input interface 43, and a processing circuit 44. Although the console device 40 is described as a separate body from the gantry device 10, the gantry device 10 may include the console device 40 or a part of each component of the console device 40.

The memory 41 is realized 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. The memory 41 stores, for example, projection data and CT image data. Further, for example, the memory 41 stores a program for a circuit included in the X-ray CT apparatus 1 to realize its function. The memory 41 may be realized by a server group (cloud) connected to the X-ray CT apparatus 1 via a network. The memory 41 is an example of a storage unit.

The display 42 displays various information. For example, the display 42 displays various images generated by the processing circuit 44 and displays a GUI (Graphical User Interface) for receiving various operations from an operator. For example, the display 42 is a liquid crystal display or a CRT (Cathode Ray Tube) display. The display 42 may be a desktop type or a tablet terminal or the like that can wirelessly communicate with the console device 40 main body. The display 42 is an example of a display unit.

The input interface 43 receives various input operations from the operator, converts the received input operations into electric signals, and outputs the electric signals to the processing circuit 44. For example, the input interface 43 receives from the operator an input operation such as a reconstruction condition for reconstructing CT image data and an image processing condition for generating a post-processed image from CT image data.

For example, the input interface 43 includes a mouse, a keyboard, a trackball, a switch, a button, a joystick, a touch pad for performing an input operation by touching an operation surface, a touch screen in which a display screen and a touch pad are integrated, and an optical sensor. It is realized by the used non-contact input circuit, voice input circuit and the like. The input interface 43 may be provided in the gantry device 10. Further, the input interface 43 may be configured with a tablet terminal or the like that can wirelessly communicate with the console device 40 main body. Further, the input interface 43 is not limited to the one including physical operation parts such as a mouse and a keyboard. For example, an electric signal processing circuit that receives an electric signal corresponding to an input operation from an external input device provided separately from the console device 40 and outputs the electric signal to the processing circuit 44 is also an example of the input interface 43. included. The input interface 43 is an example of an input unit.

The processing circuit 44 controls the operation of the entire X-ray CT apparatus 1. For example, the processing circuit 44 executes the scan control function 441, the preprocessing function 442, the image reconstruction function 443, the acquisition function 444, the estimation function 445, and the display control function 446. The scan control function 441 is an example of a scan unit. The acquisition function 444 is an example of an acquisition unit. The estimation function 445 is an example of an estimation unit. The display control function 446 is an example of a display control unit.

In the X-ray CT apparatus 1 shown in FIG. 2, each processing function is stored in the memory 41 in the form of a program that can be executed by a computer. The processing circuit 44 is a processor that realizes a function corresponding to each program by reading the program from the memory 41 and executing the program. In other words, the processing circuit 44 in the state where each program is read has a function corresponding to the read program.

Note that, in FIG. 2, each processing function of the scan control function 441, the preprocessing function 442, the image reconstruction function 443, the acquisition function 444, the estimation function 445, and the display control function 446 is realized by the single processing circuit 44. Although the case is shown, the embodiment is not limited to this. For example, the processing circuit 44 may be configured by combining a plurality of independent processors, and each processor may realize each processing function by executing each program. Further, each processing function of the processing circuit 44 may be implemented by being appropriately dispersed or integrated into a single or a plurality of processing circuits.

The overall configuration of the X-ray CT apparatus 1 according to this embodiment has been described above. With such a configuration, the X-ray CT apparatus 1 according to the present embodiment executes the following processes to support the diagnosis of circulatory disorder. In the X-ray CT apparatus 1 according to this embodiment, the processing circuit 44 executes the scan using the X-rays on the subject P into which the contrast agent has been injected once, so that the contrast agent in the first period is increased. The first concentration change information indicating the change in the concentration and the second concentration change information indicating the change in the concentration of the contrast agent in the second period different from the first period are acquired. Then, the processing circuit 44 estimates the presence or absence of a circulatory disorder in the subject P based on the input data including the acquired first concentration change information and the acquired second concentration change information.

Next, each function of the scan control function 441, the preprocessing function 442, the image reconstruction function 443, the acquisition function 444, the estimation function 445, and the display control function 446 executed by the processing circuit 44 in the first embodiment will be described. ..

FIG. 3 is a flowchart showing a procedure of processing by the medical image processing system according to the first embodiment. For example, in the first embodiment, CT imaging of the heart of the subject P is performed to support diagnosis of peripheral circulatory disorders. CT imaging is performed at regular intervals when the contrast agent is injected once into the vein of the subject P.

Step S101 of FIG. 3 is a step in which the processing circuit 44 calls the program corresponding to the scan control function 441 from the memory 41 and is executed. In step S101, the scan control function 441 executes a scan on the subject P using X-rays. For example, the scan control function 441 controls the scan based on the input operation received from the operator via the input interface 43. Specifically, the scan control function 441 controls the output voltage from the high voltage generator by transmitting a control signal to the X-ray high voltage device 14 based on the input operation. Further, the scan control function 441 controls the data collection by the DAS 18 by transmitting a control signal to the DAS 18.

Step S102 of FIG. 3 is a step in which the processing circuit 44 calls the program corresponding to the preprocessing function 442 from the memory 41 and is executed. In step S102, the preprocessing function 442 performs projection processing on the X-ray detection data transmitted from the DAS 18 by performing logarithmic conversion processing, correction processing such as offset correction, sensitivity correction, and beam hardening correction. To generate. The data before the preprocessing (detection data) and the data after the preprocessing may be collectively referred to as projection data.

Step S103 of FIG. 3 is a step in which the processing circuit 44 calls the program corresponding to the image reconstruction function 443 from the memory 41 and is executed. In step S103, the image reconstruction function 443 reconstructs the projection data generated by the preprocessing function 442 by various reconstruction methods (for example, a backprojection method such as FBP (Filtered Back Projection) or a successive approximation method). The CT image data is generated by the configuration, and the generated CT image data is stored in the memory 41.

In step S104 of FIG. 3, the processing circuit 44 calls the program corresponding to the acquisition function 444 and the display control function 446 from the memory 41 and is executed. In step S104, the acquisition function 444 acquires the CT image data stored in the memory 41. Alternatively, the acquisition function 444 acquires the CT image data generated by the image reconstruction function 443. Here, in the acquired CT image data, a region of interest of the subject P (Region Of Interest, hereinafter referred to as ROI) is designated. For example, the display control function 446 causes the display 42 to display the CT image data acquired by the acquisition function 444, and the ROI is designated in the CT image data on the display 42.

FIG. 4 is a diagram for explaining an example of the ROI in the first embodiment, and is a diagram showing the configuration of the heart. As shown in FIG. 4, in the first embodiment, for example, a region including the aorta near the heart of the subject P is designated as the ROI.

The acquisition function 444 calculates the CT value regarding the ROI of the acquired CT image data. The CT value corresponds to the concentration of the contrast agent flowing in the subject P and is also called HU (Hounsfield Unit). Then, the acquisition function 444 generates a time density curve (hereinafter, referred to as TDC) in which CT values regarding the ROI are plotted in time series. For example, the display control function 446 causes the display 42 to display the TDC generated by the acquisition function 444.

FIG. 5 is a diagram for explaining an example of TDC in the first embodiment. In FIG. 5, the horizontal axis of TDC represents the elapsed time after the contrast agent is injected into the subject P, and the vertical axis represents the CT value which is the concentration of the contrast agent flowing in the subject P.

Here, blood flows in the order of vena cava, heart (right atrium, right ventricle), pulmonary artery, lung, pulmonary vein, heart (left atrium, left ventricle), aorta, systemic organs and tissues. Therefore, the contrast agent passes through the ROI after being injected into the subject P, passes through the entire body of the subject P, and then passes through the ROI again. Therefore, as shown in FIG. 5, in TDC, when the contrast agent injected into the subject P passes through the ROI, the CT value reaches a peak value (first peak value), and then the CT value changes. Gradually lower. Then, when the contrast agent that has passed through the whole body of the subject P passes through the ROI again after a certain period of time, the CT value becomes the peak value (the second peak value lower than the first peak value) again. The CT value gradually decreases after that.

As a result, the acquisition function 444 acquires the first concentration change information indicating the concentration change of the contrast agent when the contrast agent passes through the ROI of the subject P in the first period, and after the first period. In the second period of, the second concentration change information indicating the concentration change of the contrast agent when the contrast agent passes through the ROI is acquired. Specifically, the acquisition function 444 acquires the first density change information and the second density change information using TDC as described below.

For example, as shown in FIG. 5, the first concentration change information includes the first peak value Pk1 of the CT value and the time t1 when the CT value reaches the first peak value Pk1 in TDC. .. Further, the first concentration change information is the passage time Δt1 from when the CT value exceeds the threshold value (CT value Pth) to when it reaches the first peak value Pk1 and then below the threshold value (CT value Pth) at TDC. It includes an area Ar1 representing the relationship with the CT value. The first peak value Pk1 of the CT value corresponds to the peak value of the concentration of the contrast agent when the contrast agent first flows into the ROI. The time t1 corresponds to the time when the concentration of the contrast agent reaches the peak value when the contrast agent first flows into the ROI. The area Ar1 corresponds to the amount of the contrast agent when the contrast agent first flows into the ROI. The passage time Δt1 corresponds to the above-described first period.

For example, as shown in FIG. 5, the second density change information includes the second peak value Pk2 of the CT value and the time t2 when the CT value reaches the second peak value Pk2 in TDC. .. Further, the second density change information is the passage time Δt2 from the time when the CT value exceeds the threshold value (CT value Pth) to the second peak value Pk2 after the CT value exceeds the threshold value (CT value Pth) in TDC. It includes an area Ar2 representing the relationship with the CT value. The second peak value Pk2 of the CT value corresponds to the peak value of the concentration of the contrast agent when the contrast agent next flows into the ROI. The time t2 corresponds to the time when the concentration of the contrast agent reaches the peak value when the contrast agent next flows into the ROI. The area Ar2 corresponds to the amount of the contrast agent when the contrast agent next flows into the ROI. The passage time Δt2 corresponds to the above-described second period.

The terminal circulation disorder is caused by the inhibition of blood circulation in the whole body. Therefore, the first density change information (first peak value Pk1 of CT value, time t1, area Ar1) and the second density change information (second peak value Pk2 of CT value, time t2, area Ar2) Is an index for diagnosis of peripheral circulatory disorders.

Step S105 of FIG. 3 is a step in which the processing circuit 44 calls a program corresponding to the estimation function 445 and the display control function 446 from the memory 41 and is executed. In step S105, the estimation function 445 causes the first concentration change information (first peak value Pk1 of CT value, time t1, area Ar1) acquired by the acquisition function 444 and second concentration change information (CT value of CT value). Based on the input data including the second peak value Pk2, time t2, and area Ar2), the presence/absence of an end circulation disorder in the subject P is estimated. For example, the display control function 446 causes the display 42 to display the result estimated by the estimation function 445.

For example, the estimation function 445 is acquired by the acquisition function 444 by learning using the first concentration change information, the second concentration change information obtained in the past, and the diagnosis result indicating the presence or absence of a peripheral circulation disorder. From the input data including the first concentration change information and the second concentration change information, the presence/absence of an end circulation disorder in the subject P is estimated. Examples of the learning include learning using AI (Artificial Intelligence) based on a neural network, machine learning, and the like. In addition, the estimation function 445 refers to a table that correlates the first concentration change information, the second concentration change information obtained in the past, and the diagnosis result indicating the presence or absence of a peripheral circulation disorder, and then inputs the data from the input data. The presence/absence of a terminal circulation disorder in the sample P may be estimated.

Here, in the first embodiment, the presence/absence of a terminal circulatory disorder in the subject P is estimated by learning using AI to estimate the presence/absence of a terminal circulatory disorder in the subject P and by machine learning such as clustering. A case and a case will be described.

First, a case will be described in which the presence/absence of an end circulation disorder in the subject P is estimated by learning using AI. When estimating the presence/absence of a peripheral circulatory disorder in the subject P, in learning using AI, a process of generating a learned model and a process of using the learned model are executed.

In the first embodiment, the learned model is generated by an apparatus (not shown) (hereinafter, referred to as a learned model generating apparatus) and stored in the X-ray CT apparatus 1 (for example, stored in the memory 41). ). Here, the learned model generation device may be realized by an in-hospital device (for example, the medical image processing device 3 in FIG. 1) or an external device. Alternatively, the learned model may be generated by the X-ray CT apparatus 1 (for example, the estimation function 445) and stored in the X-ray CT apparatus 1. Hereinafter, a case where the learned model is generated by the learned model generation device and stored in the X-ray CT apparatus 1 will be described as an example.

6 and 7 are diagrams for explaining an example of learning using AI in the first embodiment. 6 and 7 are block diagrams assuming deep learning using AI.

In the process of generating the learned model, as shown in FIG. 6, first, the learned model generating device acquires the data set 501 obtained when the scan using the X-ray was performed on the subject in the past. input. The data set 501 includes first density change information (first peak value Pk1 of CT value, time t1, area Ar1) acquired in the past and second density change information (second peak of CT value). The value Pk2, the time t2, and the area Ar2) are included. Furthermore, the data set 501 includes, for example, a diagnosis result indicating the presence or absence of a peripheral circulation disorder acquired in the past, information regarding the subject, and the like. The information on the subject includes the subject's sex, weight, age, cardiac function index, and the like. Examples of the cardiac function index include left ventricular ejection fraction (EF), stroke volume (SV), cardiac output (CO), and the like. Since the algorithms such as AI are learned from experience, the data sets used for learning do not have to be the same subject.

Then, the learned model generation device processes the data set 501 into the learning data set 502. The learning data set 502 includes, for example, a time difference Δt (Δt=t2-t1) between the time t1 included in the first density change information and the time t2 included in the second density change information, and the first density change information. Ratio Rp (Rp=Pk2/Pk1) between the peak value Pk1 of the CT value included in the first density change information and the peak value Pk2 of the CT value included in the second density change information, and the area Ar1 included in the first density change information The ratio Ra (Ra=Ar2/Ar1) with the area Ar2 included in the concentration change information of No. 2, the diagnosis result indicating the presence or absence of a peripheral circulation disorder, and the information about the subject are included.

The learned model generation device has a learning program 503 as an AI algorithm. The learned model generation apparatus uses the learning program 503 to learn from the learning data set 502 patterns such as the presence/absence of a terminal circulatory disorder, and receives the input data 505 described later to determine the presence/absence of a terminal circulatory disorder. Generate a trained model 504 that is functionalized to output output data 506, which will be described below. For example, the generated learned model 504 is stored in the X-ray CT apparatus 1 (stored in the memory 41). The learned model 504 in the memory 41 can be read by the estimating function 445, for example.

In the process using the learned model, as shown in FIG. 7, the estimation function 445 inputs the input data 505 obtained when the scan using the X-ray is performed on the subject P. The input data 505 includes the first concentration change information (first peak value Pk1 of CT value, time t1, area Ar1) acquired by the acquisition function 444, and second concentration change information (second CT value second). Peak value Pk2, time t2, area Ar2). Further, the input data 505 includes, for example, information regarding the subject P input by the input interface 43. The information on the subject P includes the subject P's sex, weight, age, and cardiac function index (EF, SV, CO, etc.).

Then, the estimation function 445 uses the learned model 504 read from the memory 41 to estimate the presence/absence of a terminal circulation disorder of the subject P from the input data 505, and outputs the estimated result as output data 506. For example, the display control function 446 causes the output data 506 to be displayed on the display 42.

The learned model generation device generates a learned model 504 for each subject information, and the estimation function 445 uses the generated learned model 504 to calculate the end of the subject P from the input data 505. The presence or absence of circulatory disorders may be estimated. For example, the learned model generation device generates a learned model 504 for males and females for each gender of the subject, and the estimation function 445 uses the generated learned model 504 to calculate from the input data 505. Alternatively, the presence or absence of a terminal circulation disorder in the subject P may be estimated. Further, for example, the learned model generation device generates the learned model 504 for each of the subject's weight, age, and cardiac function index as well as the subject's sex, and the estimation function 445 uses the generated learned model. 504 may be used to estimate the presence or absence of a peripheral circulation disorder of the subject P from the input data 505.

In the present embodiment, the learned model generation apparatus inputs the input data 505 including the first density change information and the second density change information acquired from the TDC, and uses the input data 505 as the learning data set 502. And learning patterns 503 are learned from the learning data set 502 using the learning program 503 to generate the learned model 504. However, it is not limited to this. For example, the learned model generation device may learn the TDC shape itself. That is, the learned model generation apparatus inputs the input data 505 including TDC, learns the pattern of the presence or absence of a terminal circulatory disorder from the input data 505 using the learning program 503, and outputs the learned model 504. May be generated.

Next, a case will be described in which the presence/absence of a peripheral circulation disorder in the subject P is estimated by machine learning such as clustering. When estimating the presence/absence of a terminal circulatory disorder in the subject P, in machine learning such as clustering, the first concentration change information obtained in the past, the second concentration change information, and a diagnosis indicating the presence/absence of a terminal circulatory disorder A process of generating correspondence information associated with the result and a process of using the correspondence information are executed.

In the first embodiment, the correspondence information is generated by a device (not shown) (hereinafter, referred to as correspondence information generation device) and stored in the X-ray CT apparatus 1 (for example, stored in the memory 41). Here, the correspondence information generation device may be realized by an in-hospital device (for example, the medical image processing device 3 in FIG. 1) or an external device. Alternatively, the correspondence information may be generated by the X-ray CT apparatus 1 (for example, the estimation function 445) and stored in the X-ray CT apparatus 1. Hereinafter, a case where the correspondence information is generated by the correspondence information generation device and stored in the X-ray CT apparatus 1 will be described as an example.

FIG. 8 is a diagram for explaining an example of machine learning such as clustering according to the first embodiment.

In the process of generating the correspondence information, first, the correspondence information generation device inputs a data set obtained when a scan using X-rays was previously performed on the subject. The input data set is, for example, similar to the data set 501 of FIG. 6, the first concentration change information (first peak value Pk1 of CT value, time t1, area Ar1) acquired in the past, and 2 density change information (second peak value Pk2 of CT value, time t2, area Ar2). Further, the input data set is, for example, similar to the data set 501 of FIG. 6, a diagnosis result indicating the presence or absence of a peripheral circulatory disorder acquired in the past, and information regarding the subject (sex, weight, age of the subject). , Cardiac function indicators, etc.) etc. Note that the data sets used for machine learning do not have to be the same subject.

Then, as shown in FIG. 8, the correspondence information generating device uses the input data set to associate the first concentration change information, the second concentration change information, and the diagnosis result indicating the presence or absence of peripheral circulatory disorder. The corresponding information 510 is generated. For example, the generated correspondence information 510 is stored in the X-ray CT apparatus 1 (stored in the memory 41). The correspondence information 510 in the memory 41 can be read by the estimation function 445, for example.

In the example illustrated in FIG. 8, the horizontal axis of the correspondence information 510 is the ratio Rp between the peak value Pk1 of the CT value included in the first density change information and the peak value Pk2 of the CT value included in the second density change information. (Rp=Pk2/Pk1), and the vertical axis represents the time difference Δt (Δt=t2-t1) between the time t1 included in the first density change information and the time t2 included in the second density change information. .. The horizontal axis of the correspondence information 510 is not limited to the ratio Rp described above, and may be the ratio of the area Ar1 included in the first density change information and the area Ar2 included in the second density change information.

Here, as shown in FIG. 8, the diagnosis result indicating the presence or absence of a terminal circulation disorder is plotted in the correspondence information 510 in association with the ratio Rp and the time difference Δt. For example, in FIG. 8, the diagnosis result is represented by points such as “◯” and “×”, and is plotted in the correspondence information 510 in association with the ratio Rp and the time difference Δt. In the correspondence information 510, the input data set includes a set of data (hereinafter, referred to as a cluster 511) including a diagnosis result (points indicated by “◯” in FIG. 8) indicating that there is a terminal circulatory disorder, and a terminal end. It is divided into a data set (hereinafter referred to as a cluster 512) including a diagnosis result (a point represented by “x” in FIG. 8) indicating that there is no circulatory disorder.

In the process using the correspondence information, the estimation function 445 inputs the input data obtained when the scan using the X-ray is performed on the subject P. The input data is, for example, similar to the input data 505 in FIG. 7, the first density change information (first peak value Pk1 of CT value Pk1, time t1, area Ar1) acquired by the acquisition function 444, and 2 density change information (second peak value Pk2 of CT value, time t2, area Ar2). Further, the input data includes, for example, similar to the input data 505 in FIG. 7, information regarding the subject P input by the input interface 43 (sex, weight, age, cardiac function index, etc. of the subject).

Then, the estimation function 445 refers to the correspondence information 510 read from the memory 41, estimates the presence/absence of a terminal circulation disorder of the subject P from the input data, and outputs the estimated result as output data. For example, the display control function 446 causes the output data to be displayed on the display 42.

For example, the estimation function 445 refers to the correspondence information 510, and when the input data is included in the cluster 511, outputs the estimation result indicating that there is a peripheral circulatory disorder as output data, and the input data is included in the cluster 512. In this case, the estimation result indicating that there is no end circulation disorder is output as output data. Specifically, the estimation function 445 obtains the ratio Rp (Rp=Pk2/Pk1) and the time difference Δt (Δt=t2-t1) from the input data, and associates the obtained ratio Rp with the time difference Δt, for example. Prediction points for estimating the presence or absence of peripheral circulation disorder are plotted in the correspondence information 510 of FIG. Then, the estimation function 445 refers to the correspondence information 510, and when the prediction point is included in the cluster 511, outputs the estimation result indicating that there is a peripheral circulation disorder as output data, and the prediction point is included in the cluster 512. In this case, the estimation result indicating that there is no end circulation disorder is output as output data.

The correspondence information generation device generates correspondence information 510 for each subject information, and the estimation function 445 refers to the generated correspondence information 510 and refers to the input data to determine whether or not there is a terminal circulation disorder of the subject P. May be estimated. For example, the correspondence information generation device generates the correspondence information 510 for men and women for each sex of the subject, and the estimation function 445 refers to the generated correspondence information 510 and refers to the input data to obtain the subject P. The presence/absence of peripheral circulatory disorder of the above may be estimated. Further, for example, the correspondence information generation device generates correspondence information 510 not only for the sex of the subject but also for each weight, age, and cardiac function index of the subject, and the estimation function 445 refers to the generated correspondence information 510. Then, the presence/absence of a terminal circulation disorder of the subject P may be estimated from the input data.

As described above, in the X-ray CT apparatus 1 according to the first embodiment, CT imaging of the heart of the subject P can be performed to support diagnosis of a peripheral circulatory disorder. For example, the scan control function 441 executes a scan using X-rays on the subject P into which the contrast agent has been injected once. By the execution of the scan, the acquisition function 444 causes the first concentration change information (first peak value Pk1 of CT value, time t1, area Ar1) indicating the concentration change of the contrast agent in the first period, and The second concentration change information (the second peak value Pk2 of the CT value, the time t2, the area Ar2) representing the concentration change of the contrast agent in the second period different from the first period is acquired. Here, the acquisition function 444 acquires the first concentration change information indicating the concentration change of the contrast agent in the region of interest (ROI) of the subject P in the first period, and then acquires the second concentration change information after the first period. During the period, the second concentration change information indicating the concentration change of the contrast agent in the ROI is acquired. Specifically, a region including the aorta near the heart of the subject P is designated as the ROI, and the acquisition function 444 causes the concentration concentration of the contrast agent for the ROI of the subject P to be represented in time series as a time concentration curve (TDC). Is generated, and the first density change information and the second density change information are acquired using TDC. The estimation function 445 estimates the presence/absence of a terminal circulation disorder in the subject P based on the input data including the acquired first concentration change information and second concentration change information.

As described above, the terminal end circulatory disorder is caused by the inhibition of blood circulation in the whole body. Therefore, the first concentration change information and the second concentration change information acquired using TDC are It serves as an index for diagnosis of circulatory disorders. As described above, in the X-ray CT apparatus 1 according to the first embodiment, CT imaging of the heart of the subject P can be performed to support diagnosis of peripheral circulatory disorders.

(Second embodiment)
In the above-described first embodiment, a case has been described in which CT imaging of the heart is performed to support diagnosis of peripheral circulatory disorders. In the second embodiment, a case will be described in which CT imaging of the heart is performed to support diagnosis of pulmonary circulation disorders. The X-ray CT apparatus 1 according to the second embodiment has the same configuration as that of the X-ray CT apparatus 1 shown in FIG. 2, and a part of the processing by the acquisition function 444 and the estimation function 445 is different. Hereinafter, points having the same configurations as those described in the first embodiment will be assigned the same reference numerals as those in FIG. 2, and description thereof will be omitted.

In step S101 of FIG. 3, the scan control function 441 executes a scan on the subject P using X-rays. For example, the scan control function 441 controls the scan based on the input operation received from the operator via the input interface 43. Specifically, the scan control function 441 controls the output voltage from the high voltage generator by transmitting a control signal to the X-ray high voltage device 14 based on the input operation. Further, the scan control function 441 controls the data collection by the DAS 18 by transmitting a control signal to the DAS 18.

In step S102 of FIG. 3, the preprocessing function 442 performs logarithmic conversion processing, correction processing such as offset correction, sensitivity correction, and beam hardening correction on the X-ray detection data transmitted from the DAS 18, Generate projection data.

In step S103 of FIG. 3, the image reconstruction function 443 reconstructs the projection data generated by the preprocessing function 442 by various projection methods such as FBP, a backprojection method, and a successive approximation method. By doing so, CT image data is generated, and the generated CT image data is stored in the memory 41.

In step S104 of FIG. 3, the acquisition function 444 acquires the CT image data stored in the memory 41. Alternatively, the acquisition function 444 acquires the CT image data generated by the image reconstruction function 443. Here, in the acquired CT image data, the first ROI and the second ROI of the subject P are designated. For example, the display control function 446 causes the display 42 to display the CT image data acquired by the acquisition function 444, and in the CT image data on the display 42, the first ROI and the second ROI are designated.

FIG. 9 is a diagram for explaining an example of the ROI in the second embodiment, and is a diagram showing the configuration of the heart. As shown in FIG. 9, in the second embodiment, for example, a region including the right ventricle (RV) of the heart of the subject P is designated as the first ROI and a region including the left ventricle (LV) of the heart. Is designated as the second ROI. Here, in the second embodiment, the first ROI and the second ROI are described as ROI (RV) and ROI (LV), respectively.

The acquisition function 444 calculates a CT value related to ROI (RV) and ROI (LV) of the acquired CT image data. Then, the acquisition function 444 generates a time concentration curve (TDC) in which CT values relating to ROI (RV) and ROI (LV) are plotted in time series. Here, in the second embodiment, TDCs related to ROI (RV) and ROI (LV) are described as TDC (RV) and TDC (LV), respectively. For example, the display control function 446 causes the display 42 to display the TDC(RV) and TDC(LV) generated by the acquisition function 444.

FIG. 10 is a diagram for explaining an example of TDC (RV) and TDC (LV) in the second embodiment. In FIG. 10, the horizontal axes of TDC (RV) and TDC (LV) represent the elapsed time after the contrast agent is injected into the subject P, and the vertical axis represents the concentration of the contrast agent flowing in the subject P. It represents a CT value.

Here, blood flows in the order of vena cava, heart (right atrium, right ventricle), pulmonary artery, lung, pulmonary vein, heart (left atrium, left ventricle), aorta, systemic organs and tissues. In the pulmonary circulation, blood flows in the order of right atrium, right ventricle (RV), pulmonary artery, lung, pulmonary vein, left atrium, left ventricle (LV), and aorta. Therefore, the contrast agent passes through the ROI (RV) after being injected into the subject P, and passes through the ROI (LV) after passing through the lungs of the subject P. Therefore, as shown in FIG. 10, in the TDC (RV), the contrast agent passes through the ROI (RV) after being injected into the subject P, so that the CT value reaches the peak value and then the CT value gradually increases. Go down. Then, in the TDC (LV), after a certain period of time, the contrast agent that has passed through the lungs of the subject P passes through the ROI (LV), so that the CT value reaches a peak value and then the CT value gradually decreases.

As a result, the acquisition function 444 acquires the first concentration change information indicating the concentration change of the contrast agent when the contrast agent passes through the ROI (RV) of the subject P in the first period, and the first concentration change information is acquired. In the second period after the period, the second concentration change information indicating the concentration change of the contrast agent when the contrast agent passes through the ROI (LV) of the subject P is acquired. Specifically, the acquisition function 444 acquires the first concentration change information using TDC(RV) and acquires the second concentration change information using TDC(LV) as follows. ..

For example, as shown in FIG. 10, the first concentration change information includes the peak value PkRV of the CT value and the time tRV when the CT value reaches the peak value PkRV in TDC(RV). Further, the first concentration change information is the passage time ΔtRV from the time when the CT value exceeds the threshold value (CT value Pth) to the time when it reaches the peak value PkRV and then falls below the threshold value (CT value Pth) in TDC (RV). It includes an area ArRV representing the relationship with the CT value. The peak value PkRV of the CT value corresponds to the peak value of the concentration of the contrast agent when the contrast agent flows into the ROI (RV). The time tRV corresponds to the time when the concentration of the contrast agent reaches the peak value when the contrast agent flows into the ROI (RV). The area ArRV corresponds to the amount of the contrast agent when the contrast agent flows into the ROI (RV). The passage time ΔtRV corresponds to the above-described first period.

For example, as shown in FIG. 10, the second concentration change information includes the peak value PkLV of the CT value and the time tLV when the CT value reaches the peak value PkLV in TDC (LV). Further, the second concentration change information is the passing time ΔtLV from the time when the CT value exceeds the threshold value (CT value Pth) to the peak value PkLV after the CT value exceeds the threshold value (CT value Pth) in TDC (LV). It includes an area ArLV representing the relationship with the CT value. The peak value PkLV of the CT value corresponds to the peak value of the concentration of the contrast agent when the contrast agent flows into the ROI (LV). The time tLV corresponds to the time when the concentration of the contrast agent reaches the peak value when the contrast agent flows into the ROI (LV). The area ArLV corresponds to the amount of the contrast agent when the contrast agent flows into the ROI (LV). The passage time ΔtLV corresponds to the above-described second period.

Pulmonary circulatory disorders result from the inhibition of blood circulation in the lungs. Therefore, the first concentration change information (CT value peak value PkRV, time tRV, area ArRV) and the second concentration change information (CT value peak value PkLV, time tLV, area ArLV) are pulmonary circulation disorders. It will be an index for diagnosis.

In step S105 in FIG. 3, the estimation function 445 uses the first concentration change information (peak value PkRV of CT value, time tRV, area ArRV) and second concentration change information (CT value of the CT value acquired by the acquisition function 444. Based on the input data including the peak value PkLV, the time tLV, and the area ArLV), the presence or absence of the pulmonary circulation disorder of the subject P is estimated. For example, the display control function 446 causes the display 42 to display the result estimated by the estimation function 445.

For example, the estimation function 445 is acquired by the acquisition function 444 by learning using the first concentration change information, the second concentration change information obtained in the past, and the diagnosis result indicating the presence or absence of pulmonary circulation disorder. The presence or absence of the pulmonary circulation disorder of the subject P is estimated from the input data including the first concentration change information and the second concentration change information. Examples of the learning include learning using AI based on a neural network and machine learning. In addition, the estimation function 445 refers to a table that correlates the first concentration change information, the second concentration change information obtained in the past, and the diagnosis result indicating the presence or absence of a peripheral circulation disorder, and then inputs the data from the input data. The presence or absence of pulmonary circulation disorder in the sample P may be estimated.

Here, in the second embodiment, a case where the presence or absence of pulmonary circulation disorder of the subject P is estimated by learning using AI, and a case where the presence or absence of pulmonary circulation disorder of the subject P is estimated by machine learning such as clustering. And will be described.

First, a case will be described in which the presence or absence of pulmonary circulation disorder in the subject P is estimated by learning using AI. When estimating the presence/absence of pulmonary circulation disorder in the subject P, in learning using AI, a process of generating a learned model and a process of using the learned model are executed.

In the second embodiment, the learned model is generated by an apparatus (not shown) (hereinafter, referred to as a learned model generation apparatus) and stored in the X-ray CT apparatus 1 (for example, stored in the memory 41). ). Here, the learned model generation device may be realized by an in-hospital device (for example, the medical image processing device 3 in FIG. 1) or an external device. Alternatively, the learned model may be generated by the X-ray CT apparatus 1 (for example, the estimation function 445) and stored in the X-ray CT apparatus 1. Hereinafter, a case where the learned model is generated by the learned model generation device and stored in the X-ray CT apparatus 1 will be described as an example.

11 and 12 are diagrams for explaining an example of learning using AI in the second embodiment. 11 and 12 are block diagrams assuming deep learning using AI.

In the process of generating the learned model, as shown in FIG. 11, first, the learned model generation device acquires the data set 601 obtained when the scan using the X-ray was performed on the subject in the past. input. The data set 601 includes first concentration change information (peak value PkRV of CT value, time tRV, area ArRV) acquired in the past, and second concentration change information (peak value PkLV of CT value, time tLV, Area ArLV). Furthermore, the data set 601 includes, for example, a diagnosis result indicating the presence or absence of pulmonary circulation disorder acquired in the past, information regarding the subject, and the like. The information on the subject includes the subject's sex, weight, age, cardiac function index, and the like. Examples of the cardiac function index include left ventricular ejection fraction (EF), stroke volume (SV), cardiac output (CO), and the like. Since the algorithms such as AI are learned from experience, the data sets used for learning do not have to be the same subject.

Then, the learned model generation device processes the data set 601 into the learning data set 602. The learning data set 602 includes, for example, a time difference Δt (Δt=tLV-tRV) between the time tRV included in the first density change information and the time tLV included in the second density change information, the first density change information. Ratio Rp (Pp=PkLV/PkRV) between the peak value PkRV of the CT value included in the first concentration change information and the peak value PkLV of the CT value included in the second concentration change information, and the area ArRV and the first concentration change information included in the first concentration change information. The ratio Ra (Ra=ArLV/ArRV) with respect to the area ArLV included in the concentration change information of No. 2, the diagnosis result indicating the presence or absence of pulmonary circulation disorder, and the information about the subject are included.

The learned model generation device has a learning program 603 as an AI algorithm. The learned model generation apparatus learns a pattern of presence/absence of pulmonary circulation disorder from the learning data set 602 using the learning program 603, and receives an input of input data 605 described later to estimate the presence/absence of pulmonary circulation disorder. A trained model 604 that is functionalized to output output data 606, which will be described later, is generated. For example, the generated learned model 604 is stored in the X-ray CT apparatus 1 (stored in the memory 41). The learned model 604 in the memory 41 can be read by the estimation function 445, for example.

In the process using the learned model, as shown in FIG. 12, the estimation function 445 inputs the input data 605 obtained when the scan using the X-ray is performed on the subject P. The input data 605 includes the first concentration change information (peak value PkRV of CT value, time tRV, area ArRV) acquired by the obtaining function 444, and second concentration change information (peak value PkLV of CT value PkLV, time). tLV, area ArLV). Further, the input data 605 includes, for example, information regarding the subject P input by the input interface 43. The information on the subject P includes the sex, weight, age, cardiac function index (EF, SV, CO, etc.) of the subject P.

Then, the estimation function 445 estimates the presence or absence of the pulmonary circulation disorder of the subject P from the input data 605 using the learned model 604 read from the memory 41, and outputs the estimated result as output data 606. For example, the display control function 446 causes the output data 606 to be displayed on the display 42.

The trained model generation device generates a trained model 604 for each subject information, and the estimation function 445 uses the trained model 604 thus generated to calculate the pulmonary circulation of the subject P from the input data 605. The presence or absence of a fault may be estimated. For example, the learned model generation device generates a learned model 604 for males and females for each gender of the subject, and the estimation function 445 uses the generated learned model 604 to calculate from the input data 605. The presence or absence of pulmonary circulation disorder in the subject P may be estimated. Further, for example, the learned model generation device generates the learned model 604 for each of the subject's weight, age, and cardiac function index as well as the subject's sex, and the estimation function 445 uses the generated learned model. The presence/absence of pulmonary circulation disorder of the subject P may be estimated from the input data 605 using 604.

Further, the position of the peak of TDC described above, for example, the peak value PkRV of the CT value included in the first concentration change information and the time tRV, and the peak value PkLV of the CT value included in the second concentration change information and the time tLV. May differ not only in the subject information described above, but also in the subject's lung volume. Therefore, for example, the learned model generation device generates a learned model 604 for each lung volume of the subject, and the estimation function 445 uses the generated learned model 604 to relate the lung volume of the subject P. The presence or absence of pulmonary circulation disorder in the subject P may be estimated from the input data 605 that further includes information.

In the present embodiment, the learned model generation device inputs the input data 605 including the first density change information and the second density change information acquired from the TDC, and inputs the input data 605 as the learning data set 602. Then, the learning program 603 is used to learn from the learning data set 602 the patterns such as the presence/absence of pulmonary circulation disorder, and the learned model 604 is generated. However, it is not limited to this. For example, the learned model generation device may learn the TDC shape itself. That is, the learned model generation device inputs the input data 605 including TDC, learns the pattern of the presence or absence of pulmonary circulation disorder from the input data 605 using the learning program 603, and generates the learned model 604. You may.

Next, a case will be described in which the presence or absence of a pulmonary circulation disorder of the subject P is estimated by machine learning such as clustering. When estimating the presence/absence of pulmonary circulation disorder in the subject P, in machine learning such as clustering, the first concentration change information obtained in the past, the second concentration change information, and the diagnostic result indicating the presence/absence of pulmonary circulation disorder are obtained. A process of generating the corresponding correspondence information and a process of using the correspondence information are executed.

In the second embodiment, the correspondence information is generated by a device (not shown) (hereinafter, referred to as correspondence information generation device) and stored in the X-ray CT apparatus 1 (for example, stored in the memory 41). Here, the correspondence information generation device may be realized by an in-hospital device (for example, the medical image processing device 3 in FIG. 1) or an external device. Alternatively, the correspondence information may be generated by the X-ray CT apparatus 1 (for example, the estimation function 445) and stored in the X-ray CT apparatus 1. Hereinafter, a case where the correspondence information is generated by the correspondence information generation device and stored in the X-ray CT apparatus 1 will be described as an example.

FIG. 13 is a diagram for explaining an example of machine learning such as clustering according to the second embodiment.

In the process of generating the correspondence information, first, the correspondence information generation device inputs a data set obtained when a scan using X-rays was previously performed on the subject. The input data set is, for example, similar to the data set 601 in FIG. 11, the first density change information (peak value PkRV of CT value, time tRV, area ArRV) acquired in the past, and the second density. Change information (peak value PkLV of CT value, time tLV, area ArLV) is included. Further, the input data set is, for example, similar to the data set 601 of FIG. 11, a diagnosis result indicating the presence or absence of pulmonary circulation disorder acquired in the past, and information about the subject (sex, weight, age of the subject, Cardiac function index etc.) etc. are included. The data sets used for machine learning may not be the same subject.

Then, as shown in FIG. 13, the correspondence information generation apparatus uses the input data set to associate the first concentration change information, the second concentration change information, and the diagnosis result indicating the presence or absence of pulmonary circulation disorder. Corresponding correspondence information 610 is generated. For example, the generated correspondence information 610 is stored in the X-ray CT apparatus 1 (stored in the memory 41). The correspondence information 610 in the memory 41 can be read by the estimation function 445, for example.

In the example shown in FIG. 13, the horizontal axis of the correspondence information 610 indicates the ratio Rp between the peak value PkRV of the CT value included in the first density change information and the peak value PkLV of the CT value included in the second density change information. (Rp=PkLV/PkRV), and the vertical axis represents the time difference Δt (Δt=tLV-tRV) between the time tRV included in the first density change information and the time tLV included in the second density change information. .. The horizontal axis of the correspondence information 610 is not limited to the ratio Rp described above, and may be the ratio of the area ArRV included in the first density change information and the area ArLV included in the second density change information.

Here, as shown in FIG. 13, the diagnosis result indicating the presence or absence of the pulmonary circulation disorder is plotted in the correspondence information 610 in association with the ratio Rp and the time difference Δt. For example, in FIG. 13, the diagnosis result is represented by points such as “◯” and “×”, and is plotted in the correspondence information 610 in association with the ratio Rp and the time difference Δt. In the correspondence information 610, the input data set includes a set of data (hereinafter, referred to as a cluster 611) including a diagnosis result (points indicated by “◯” in FIG. 13) indicating that there is a pulmonary circulation disorder, and a pulmonary circulation disorder. Data set (hereinafter, referred to as cluster 612) including a diagnosis result (a point represented by “x” in FIG. 13) indicating that there is no error.

In the process using the correspondence information, the estimation function 445 inputs the input data obtained when the scan using the X-ray is performed on the subject P. The input data is, for example, similar to the input data 605 in FIG. 12, the first concentration change information (CT value peak value PkRV, time tRV, area ArRV) acquired by the acquisition function 444, and the second density. Change information (peak value PkLV of CT value, time tLV, area ArLV) is included. Further, the input data includes, for example, information about the subject P input by the input interface 43 (sex, weight, age, cardiac function index, etc. of the subject) similar to the input data 605 in FIG.

Then, the estimation function 445 refers to the correspondence information 610 read from the memory 41, estimates the presence or absence of the pulmonary circulation disorder of the subject P from the input data, and outputs the estimated result as output data. For example, the display control function 446 causes the output data to be displayed on the display 42.

For example, when the input data is included in the cluster 611, the estimation function 445 outputs the estimation result indicating that there is a pulmonary circulation disorder as output data when the input data is included in the cluster 611, and the input data is included in the cluster 612. , The estimation result indicating that there is no pulmonary circulation disorder is output as output data. Specifically, the estimation function 445 obtains the ratio Rp (Rp=PkLV/PkRV) and the time difference Δt (Δt=tLV-tRV) from the input data, and associates the obtained ratio Rp with the time difference Δt, Prediction points for estimating the presence or absence of pulmonary circulation disorder are plotted in the correspondence information 610 in FIG. When the prediction point is included in the cluster 611, the estimation function 445 outputs the estimation result indicating that the pulmonary circulation disorder is present as output data when the prediction point is included in the cluster 611, and the prediction point is included in the cluster 612. , The estimation result indicating that there is no pulmonary circulation disorder is output as output data.

The correspondence information generation device generates correspondence information 610 for each piece of subject information, and the estimation function 445 refers to the generated correspondence information 610 to determine whether or not there is a pulmonary circulation disorder of the subject P from the input data. It may be estimated. For example, the correspondence information generation device generates the correspondence information 610 for men and women for each sex of the subject, and the estimation function 445 refers to the generated correspondence information 610 and refers to the input data to examine the subject P from the input data. The presence/absence of pulmonary circulation disorder may be estimated. Further, for example, the correspondence information generation device generates correspondence information 610 not only for the sex of the subject but also for each weight, age, and cardiac function index of the subject, and the estimation function 445 refers to the generated correspondence information 610. Then, the presence or absence of the pulmonary circulation disorder of the subject P may be estimated from the input data.

Further, as described above, the position of the TDC peak, for example, the peak value PkRV and the time tRV of the CT value included in the first concentration change information, and the peak value PkLV of the CT value included in the second concentration change information. And time tLV may differ not only in the subject's information but also in the subject's lung volume. Therefore, for example, the learned model generation device generates the correspondence information 610 for each lung volume of the subject, and the estimation function 445 refers to the generated correspondence information 610 to obtain the information regarding the lung volume of the subject P. The presence/absence of a pulmonary circulation disorder of the subject P may be estimated from the input data further included.

As described above, the X-ray CT apparatus 1 according to the second embodiment can support the diagnosis of pulmonary circulation disorder by performing CT imaging of the heart of the subject P. For example, the scan control function 441 executes a scan using X-rays on the subject P into which the contrast agent has been injected once. By performing the scan, the acquisition function 444 causes the first concentration change information (peak value PkRV of CT value, time tRV, area ArRV) indicating the concentration change of the contrast agent in the first period, and the first period. The second concentration change information (peak value PkLV of CT value, time tLV, area ArLV) showing the concentration change of the contrast agent in the second period different from is acquired. Here, the acquisition function 444 acquires the first concentration change information representing the concentration change of the contrast agent in the first region of interest (ROI(RV)) of the subject P in the first period, and acquires the first concentration change information. In the second period after the period, the second concentration change information indicating the concentration change of the contrast agent in the second region of interest (ROI(LV)) of the subject P is acquired. Specifically, the region including the right ventricle (RV) of the heart of the subject P is designated as ROI (RV), and the region including the left ventricle (LV) of the heart is designated as ROI (LV), and the acquisition function 444, for ROI (RV) and ROI (LV) of the subject P, TDC (RV) and TDC (LV) which are time-density curves (TDC) representing the concentration of the contrast agent in time series are generated. , TDC(RV) is used to acquire the first density change information, and TDC(LV) is used to acquire the second density change information. The estimation function 445 estimates the presence or absence of the pulmonary circulation disorder of the subject P based on the input data including the acquired first concentration change information and the acquired second concentration change information.

As described above, since the pulmonary circulation disorder is caused by the inhibition of blood circulation in the lungs, the first concentration change information acquired by using TDC (RV) and TDC (LV) are used. The second concentration change information acquired as a result serves as an index for diagnosis of pulmonary circulation disorder. As described above, in the X-ray CT apparatus 1 according to the second embodiment, CT imaging of the heart of the subject P can support the diagnosis of pulmonary circulation disorders.

(Modifications of the first and second embodiments)
In the above-described first and second embodiments, a case has been described in which CT imaging of the heart is performed to support diagnosis of circulatory disorders such as peripheral circulation disorder and pulmonary circulation disorder. Here, in the modified examples of the first embodiment and the second embodiment, not only the presence or absence of circulatory disorder (terminal circulatory disorder or pulmonary circulatory disorder), but also lesions (disease, disorder, etc.) that co-occur with the circulatory disorder, and , Learning is also performed in consideration of lesions (illnesses, disorders, etc.) that occur after the circulatory disorder. As a result, in the modification, not only is the presence/absence of a circulatory disorder in the subject P estimated from the first concentration change information and the second concentration change information, but when the subject P has a circulatory disorder, It is also possible to estimate the presence or absence of lesions that occur together with and the presence or absence of lesions that occur after circulatory disorders.

In the modification, for example, the estimation function 445 uses the learning described in the first embodiment and the second embodiment to extract the subject P from the input data including the first concentration change information and the second concentration change information. The presence/absence of circulatory disorder is estimated, and when the subject P has a circulatory disorder, the presence/absence of a lesion coexisting with the circulatory disorder and the presence/absence of a lesion occurring after the circulatory disorder are estimated.

Here, in a modified example, a case will be described in which the presence or absence of a circulatory disorder in the subject P is estimated by learning using AI. Also in this case, in the learning using AI, the process of generating the learned model and the process of using the learned model are executed.

In the modified example, the learned model is generated by an apparatus (not shown) (hereinafter, referred to as a learned model generation apparatus) and stored in the X-ray CT apparatus 1 (for example, stored in the memory 41). Here, the learned model generation device may be realized by an in-hospital device (for example, the medical image processing device 3 in FIG. 1) or an external device. Alternatively, the learned model may be generated by the X-ray CT apparatus 1 (for example, the estimation function 445) and stored in the X-ray CT apparatus 1. Hereinafter, a case where the learned model is generated by the learned model generation device and stored in the X-ray CT apparatus 1 will be described as an example.

14 and 15 are diagrams for explaining an example of learning using AI in the modified examples of the first embodiment and the second embodiment. 14 and 15 are block diagrams assuming deep learning using AI.

In the process of generating the learned model, as shown in FIG. 14, first, the learned model generation apparatus acquires the data set 701 obtained when the scan using the X-ray was previously performed on the subject. input. The data set 701 includes the first density change information and the second density change information acquired in the past. Further, the data set 701 includes, for example, a diagnostic result indicating the presence or absence of a circulatory disorder (terminal circulatory disorder or pulmonary circulatory disorder) acquired in the past, information regarding the subject, and the like. The information about the subject includes the subject's sex, weight, age, and cardiac function index. In addition, the information regarding the subject includes a disease or disorder immediately before the circulatory disorder. In addition, the information regarding the subject includes a lesion (disease, disorder, etc.) that is accompanied by the circulatory disorder. Further, the information regarding the subject includes a lesion (illness, disorder, etc.) that has occurred after the circulatory disorder. Since the algorithms such as AI are learned from experience, the data sets used for learning do not have to be the same subject.

Then, the learned model generation device processes the data set 701 into the learning data set 702. The learning data set 702 includes, for example, a time difference Δt between the time included in the first density change information and the time included in the second density change information, and the peak value of the CT value included in the first density change information. The ratio Rp of the CT value included in the second density change information to the peak value, the ratio Ra of the area included in the first density change information to the area included in the second density change information, the circulatory disorder (end circulation Diagnostic result indicating the presence or absence of a disorder or pulmonary circulation disorder, and information regarding the subject.

The learned model generation device has a learning program 703 as an AI algorithm. The learned model generation apparatus uses the learning program 703 to learn from the learning data set 702 patterns such as the presence or absence of circulatory disorder (terminal circulatory disorder or pulmonary circulatory disorder), and inputs input data 705 described below. A trained model 704 is generated that is functionalized to output output data 706, which will be described later and that estimates the presence or absence of circulatory disorders. For example, the generated learned model 704 is stored in the X-ray CT apparatus 1 (stored in the memory 41). The learned model 704 in the memory 41 can be read by the estimation function 445, for example.

In the process using the learned model, the estimation function 445 inputs the input data 705 obtained when the scan using the X-ray is performed on the subject P, as illustrated in FIG. 15. The input data 705 includes the first density change information (peak value of CT value, time, area) and the second density change information (peak value of CT value, time, area) acquired by the acquisition function 444. Including. Further, the input data 705 includes information regarding the subject P input by the input interface 43, for example. The information on the subject P includes the sex, weight, age, cardiac function index (EF, SV, CO, etc.) of the subject P. Further, the information regarding the subject P includes information indicating the immediately preceding illness or disorder.

Then, the estimation function 445 uses the learned model 704 read from the memory 41 to estimate the presence/absence of a circulatory disorder (terminal circulatory disorder or pulmonary circulatory disorder) of the subject P from the input data 705, and to the subject P. When there is a circulatory disorder, the presence/absence of a lesion coexisting with the circulatory disorder and the presence/absence of a lesion occurring after the circulatory disorder are estimated. The estimation function 445 outputs the estimation result as output data 706. For example, the display control function 446 causes the output data 706 to be displayed on the display 42.

The learned model generation device generates a learned model 704 for each information of the subject, and the estimation function 445 uses the generated learned model 704 to circulate the subject P from the input data 705. In addition to estimating the presence/absence of a disorder, when the subject P has a circulatory disorder, the presence/absence of a lesion coexisting with the circulatory disorder and the presence/absence of a lesion occurring after the circulatory disorder may be estimated. For example, the learned model generation device generates a learned model 704 for males and females for each gender of the subject, and the estimation function 445 uses the generated learned model 704 to calculate from the input data 705. , Estimating the presence/absence of circulatory disorder in the subject P, and estimating the presence/absence of a lesion that occurs with the circulatory disorder when the subject P has a circulatory disorder and the presence/absence of a lesion occurring after the circulatory disorder. May be. Further, for example, the learned model generation device generates the learned model 704 for each of the subject's weight, age, and cardiac function index as well as the subject's sex, and the estimation function 445 uses the generated learned model. 704 is used to estimate the presence/absence of a circulatory disorder in the subject P from the input data 705, and in the case where the subject P has a circulatory disorder, the presence/absence of a lesion concomitant with the circulatory disorder and the circulatory disorder The presence or absence of lesions that occur after the procedure may be estimated.

In the modified example, the learned model generation apparatus inputs the input data 705 including the first density change information and the second density change information acquired from the TDC, and inputs the input data 705 into the learning data set 702. After processing, the learning program 703 is used to learn the pattern of the presence or absence of circulatory disorder from the learning data set 702, and the learned model 704 is generated. However, it is not limited to this. For example, the learned model generation device may learn the TDC shape itself. That is, the learned model generation device inputs the input data 705 including the TDC, uses the learning program 703 to learn the pattern of the presence or absence of circulatory disorder from the input data 705, and generates the learned model 704. You may.

(Third Embodiment)
Although the first and second embodiments have been described so far, the present invention may be implemented in various different forms other than the first and second embodiments described above.

In the above-described embodiment, the X-ray CT apparatus 1 is described as an example of the modality apparatus, but the present invention can be applied not only to CT imaging but also to a modality apparatus such as an MRI apparatus that images using a contrast agent.

Moreover, in the above-mentioned embodiment, the case where the X-ray CT apparatus 1 executes various processes has been described. However, the embodiment is not limited to this, and may be a case where various processes are executed in the medical image processing apparatus 3, for example. FIG. 16 is a diagram showing an example of the configuration of the medical image processing apparatus 3 according to the third embodiment.

For example, as shown in FIG. 16, the medical image processing apparatus 3 has an input interface 310, a display 320, a storage circuit 330, and a processing circuit 340.

The input interface 310 includes a trackball for performing various settings, a switch button, a mouse, a keyboard, a touch pad for performing an input operation by touching an operation surface, a touch screen in which a display screen and a touch pad are integrated, It is realized by a non-contact input circuit using an optical sensor, a voice input circuit, and the like.

The input interface 310 is connected to the processing circuit 340, converts the input operation received from the operator into an electric signal, and outputs the electric signal to the processing circuit 340. In the present specification, the input interface 310 is not limited to one including physical operation parts such as a mouse and a keyboard. For example, a processing circuit that receives an electric signal corresponding to an input operation from an external input device provided separately from the device and outputs the electric signal to the processing circuit 340 is included in the example of the input interface.

The display 320 is connected to the processing circuit 340 and displays various information and various image data output from the processing circuit 340. For example, the display 320 is realized by a liquid crystal monitor, a CRT (Cathode Ray Tube) monitor, a touch panel, or the like. For example, the display 320 displays a GUI (Graphical User Interface) for accepting an operator's instruction, various display images, and various processing results by the processing circuit 350 (for example, a scatter diagram or an analysis image).

The storage circuit 330 is connected to the processing circuit 340 and stores various data. For example, the storage circuit 330 is realized by a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, a hard disk, an optical disk, or the like. For example, the storage circuit 330 stores CT image data or the like received from the X-ray CT apparatus 1 or the image storage apparatus 2. The storage circuit 330 also stores a program corresponding to each processing function executed by the processing circuit 340.

The processing circuit 340 controls each component included in the medical image processing apparatus 3 according to an input operation received from an operator via the input interface 310. For example, the processing circuit 340 is implemented by a processor. In the present embodiment, the processing circuit 340 stores the CT image data output from the input interface 310 in the storage circuit 330. Further, the processing circuit 340 reads the CT image data from the storage circuit 330 and displays it on the display 320.

Then, the processing circuit 350 executes a control function 341, an image reconstruction function 343, an acquisition function 344, an estimation function 345, and a display control function 346, as shown in FIG. The control function 341 controls the entire medical image processing apparatus 300. The image reconstruction function 343 executes the same processing as the image reconstruction function 443 described above. The acquisition function 344 executes the same processing as the acquisition function 444 described above. The estimation function 345 executes the same processing as the estimation function 445 described above. The display control function 346 executes the same processing as the display control function 446 described above.

In the embodiment described above, the independent processing by the X-ray CT apparatus 1 and the medical image processing apparatus 3 has been described. However, the embodiment is not limited to this, and may be a case where the X-ray CT apparatus 1 and the medical image processing apparatus 3 function as a medical image processing system that cooperates. In such a case, the CT image data collected by the X-ray CT apparatus 1 is transferred to the medical image processing apparatus 3, and the medical image processing apparatus 3 executes various processes.

The word "processor" used in the above description is, for example, a CPU, a GPU (Graphics Processing Unit), an application specific integrated circuit (ASIC), a programmable logic device (for example, a simple programmable logic device ( A circuit such as a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor implements the function by reading and executing the program stored in the memory 41 or the storage circuit 330.

Note that, in FIG. 2, the single memory 41 has been described as storing a program corresponding to each processing function. Further, in FIG. 16, the single storage circuit 330 has been described as storing a program corresponding to each processing function. However, the embodiment is not limited to this. For example, the plurality of memories 41 may be arranged in a distributed manner, and the processing circuit 44 may read the corresponding program from the individual memories 41. Further, for example, the plurality of storage circuits 330 may be arranged in a distributed manner, and the processing circuit 340 may read the corresponding program from the individual storage circuits 330. Further, instead of storing the program in the memory 41 or the memory circuit 330, the program may be directly incorporated in the circuit of the processor. In this case, the processor realizes the function by reading and executing the program incorporated in the circuit.

Further, the processing circuit 44 and the processing circuit 340 may implement the functions by utilizing the processor of the external device connected via the network. For example, the processing circuit 44 reads out and executes a program corresponding to each function from the memory 41, and uses an external workstation or cloud connected to the X-ray CT apparatus 1 via the network NW as a computing resource. This realizes each function shown in FIG. Further, for example, the processing circuit 340 reads out and executes a program corresponding to each function from the storage circuit 330, and uses an external workstation or cloud connected to the medical image processing apparatus 3 via the network NW as a calculation resource. By using it, each function shown in FIG. 16 is realized.

Each component of each device according to the above-described embodiment is functionally conceptual, and does not necessarily have to be physically configured as illustrated. That is, the specific form of distribution/integration of each device is not limited to that shown in the figure, and all or a part of the device may be functionally or physically distributed in arbitrary units according to various loads or usage conditions. It can be integrated and configured. Furthermore, all or arbitrary parts of the processing functions performed by each device may be realized by a CPU and a program that is analyzed and executed by the CPU, or may be realized as hardware by a wired logic.

Further, the control method described in the above embodiment can be realized by executing a control program prepared in advance on a computer such as a personal computer or a workstation. This control program can be distributed via a network such as the Internet. The control program can also be executed by being recorded in a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, or a DVD, and being read from the recording medium by the computer.

According to at least one embodiment described above, diagnosis of circulatory disorders can be supported.

Although some embodiments of the present invention 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, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and modifications thereof are included in the invention described in the claims and equivalents thereof, as well as included in the scope and spirit of the invention.

1 X-ray CT apparatus 3 Medical image processing apparatus 344 Acquisition function 345 Estimating function 346 Display control function 441 Scan control function 444 Acquisition function 445 Estimating function 446 Display control function

Claims (11)

By performing a scan using X-rays on a subject into which the contrast agent has been injected once, first concentration change information indicating a concentration change of the contrast agent in a first period, and the first concentration change information. An acquisition unit that acquires second concentration change information indicating a concentration change of the contrast agent in a second period different from the period of
An estimation unit that estimates the presence or absence of a circulatory disorder in the subject based on input data including the first concentration change information and the second concentration change information;
A medical image processing apparatus comprising: The acquisition unit acquires the first concentration change information indicating a concentration change of the contrast agent in a region of interest of the subject in the first period, and the second concentration change information after the first period. Acquiring the second concentration change information representing the concentration change of the contrast agent in the region of interest during the period;
The estimation unit estimates the presence or absence of a terminal circulatory disorder as a circulatory disorder of the subject based on the input data including the first concentration change information and the second concentration change information.
The medical image processing apparatus according to claim 1. The acquisition unit generates a time-density curve that represents the concentration of the contrast agent in time series for the region of interest of the subject, and uses the time-density curve to generate the first concentration change information and the second concentration change information. The concentration change information of
The medical image processing apparatus according to claim 2. The acquisition unit acquires the first concentration change information indicating the concentration change of the contrast agent in the first region of interest of the subject in the first period, and the acquisition unit after the first period. In the second period, acquiring the second concentration change information indicating the concentration change of the contrast agent in the second region of interest of the subject,
The estimation unit estimates the presence or absence of pulmonary circulation disorder as the circulatory disorder of the subject based on the input data including the first concentration change information and the second concentration change information.
The medical image processing apparatus according to claim 1. The acquisition unit includes a first time-concentration curve and a second time-concentration curve that represent the concentration of the contrast agent in time series for the first region of interest and the second region of interest of the subject, respectively. Is generated, the first concentration change information is acquired using the first time concentration curve, and the second concentration change information is acquired using the second time concentration curve,
The medical image processing apparatus according to claim 4. The first region of interest is a region including the right ventricle of the subject's heart, and the second region of interest is a region including the left ventricle of the heart.
The medical image processing apparatus according to claim 4. The input data further includes information about the lung capacity of the subject,
The medical image processing apparatus according to any one of claims 4 to 6. The estimation unit uses the input data to perform the learning by using the first concentration change information, the second concentration change information, and a diagnosis result indicating the presence or absence of circulatory disorder obtained in the past, from the input data to the subject. The presence or absence of circulatory disorders in
The medical image processing apparatus according to claim 1. The estimation unit estimates the presence or absence of a circulatory disorder in the subject from the input data by the learning, and determines the presence or absence of a lesion coexisting with the circulatory disorder, and the presence or absence of a lesion occurring after the circulatory disorder. presume,
The medical image processing apparatus according to claim 8. A scan unit that performs a scan using X-rays on a subject into which the contrast agent has been injected once;
By performing the scan, the first concentration change information indicating the concentration change of the contrast agent in the first period and the concentration change of the contrast agent in the second period different from the first period are displayed. An acquisition unit that acquires the second density change information that represents
An estimation unit that estimates the presence or absence of a circulatory disorder in the subject based on input data including the first concentration change information and the second concentration change information;
A medical image diagnostic apparatus comprising: By performing a scan using X-rays on a subject into which the contrast agent has been injected once, first concentration change information indicating a concentration change of the contrast agent in a first period, and the first concentration change information. Acquiring second concentration change information representing a concentration change of the contrast agent in a second period different from the period of
Estimating the presence/absence of a circulatory disorder in the subject based on the input data including the first concentration change information and the second concentration change information,
An image processing program that causes a computer to execute processing.

Images