JP2020141808A - Medical processing device and medical diagnostic system - Google Patents

Abstract

To shorten the time for calibration.SOLUTION: A medical processing device includes a data acquisition unit and a data generation unit. The data acquisition unit acquires first calibration data corresponding to a first scan condition. The data generation unit generates second calibration data by inputting the first calibration data to a first learned model for generating the second calibration data corresponding to a second scan condition different from the first scan condition based on the first calibration data.SELECTED DRAWING: Figure 1

Description

Embodiments of the present invention relate to medical processing devices and medical diagnostic systems.

Conventionally, in medical diagnostic equipment such as X-ray CT equipment, deviations in measured values caused by individual differences in each unit of the equipment (differences in radiation quality, differences in the effects of scattered radiation, etc.), differences in installation environment, etc. Calibration data for correction is being collected. The collection of this calibration data is, for example, the tube voltage (kV) applied to the X-ray tube, the number of columns detected by the X-ray detector, the X-ray irradiation range (FOV: Field of View) at the time of photographing, It is carried out for each condition such as the rotation speed of the X-ray tube, the setting (gain) of the X-ray detector, and the focal size irradiated by the X-ray tube.

However, in the conventional technique, the more combinations of conditions for collecting calibration data, the longer the time required for collection. For this reason, for example, when installing an X-ray CT device or when replacing hardware parts such as a tube, it takes time to collect calibration data, which may delay the imaging of the subject. there were.

Japanese Unexamined Patent Publication No. 2001-70297

The problem to be solved by the present invention is to reduce the calibration time.

The medical processing apparatus of the embodiment includes a data acquisition unit and a data generation unit. The data acquisition unit acquires the first calibration data corresponding to the first scan condition. The data generation unit refers to the first trained model for generating the second calibration data corresponding to the second scan condition different from the first scan condition based on the first calibration data. Then, by inputting the first calibration data, the second calibration data is generated.

The block diagram of the X-ray CT apparatus 1 which concerns on 1st Embodiment. The figure which shows an example of the data stored in the memory 41 which concerns on 1st Embodiment. The figure which shows an example of the calibration data stored in the memory 41 which concerns on 1st Embodiment. The figure which shows an example of the functional structure of the calibration function 57 which concerns on 1st Embodiment. The flowchart which shows the series flow of the learning process of the processing circuit 50 which concerns on 1st Embodiment. The figure explaining the concept of the learning process by the learning function 58 which concerns on 1st Embodiment. The flowchart which shows the series flow of the calibration data generation processing by the processing circuit 50 which concerns on 1st Embodiment. The figure explaining the concept of the calibration data generation processing by the calibration function 57 which concerns on 1st Embodiment. The figure explaining another example of the calibration data generation processing by the calibration function 57 which concerns on 1st Embodiment. The figure explaining the concept of the learning process by the learning function 58 in the 2nd Embodiment. The figure explaining the concept of the calibration data generation processing by the calibration function 57 which concerns on 2nd Embodiment. The figure which shows an example of a state in which the X-ray CT apparatus 1 installed in the medical facility which concerns on embodiment and the server apparatus S on the cloud are connected via a network NW.

Hereinafter, the medical processing apparatus and the medical diagnostic system of the embodiment will be described with reference to the drawings. The medical processing apparatus is, for example, a medical diagnostic apparatus that diagnoses a subject by processing a medical image such as an X-ray CT (Computed Tomography) apparatus. In the following description, a case where the medical processing device is an X-ray CT device will be described as an example, but the present invention is not limited to this.

(First Embodiment)
FIG. 1 is a block diagram of the X-ray CT apparatus 1 according to the first embodiment. The X-ray CT device 1 includes, for example, a gantry device 10, a sleeper device 30, and a console device 40. In FIG. 1, for convenience of explanation, both a view of the gantry device 10 from the Z-axis direction and a view from the X-axis direction are shown, but in reality, the gantry device 10 is one. In the present embodiment, the rotation axis of the rotation frame 17 in the non-tilt state or the longitudinal direction of the top plate 33 of the sleeper device 30 is orthogonal to the Z-axis direction and the Z-axis direction, and the axis horizontal to the floor surface is X. The directions orthogonal to the axial direction and the Z-axis direction and perpendicular to the floor surface are defined as the Y-axis directions, respectively. The X-ray CT device 1 or the console device 40 is an example of a “medical processing device”.

The gantry device 10 includes, for example, an X-ray tube 11, a wedge 12, a collimator 13, an X-ray high voltage device 14, an X-ray detector 15, and a data acquisition system (hereinafter, DAS: Data Acquisition System) 16. , A rotating frame 17 and a control device 18.

The X-ray tube 11 generates X-rays by irradiating thermoelectrons from the cathode (filament) toward the anode (target) by applying a high voltage from the X-ray high voltage device 14. The X-ray tube 11 includes a vacuum tube. For example, the X-ray tube 11 is a rotating anode type X-ray tube that generates X-rays by irradiating a rotating anode with thermoelectrons.

The wedge 12 is a filter for adjusting the X-ray dose applied to the subject P from the X-ray tube 11. The wedge 12 attenuates the X-rays that pass through itself so that the distribution of the X-ray dose radiated from the X-ray tube 11 to the subject P becomes a predetermined distribution. The wedge 12 is also called a wedge filter or a bow-tie filter. The wedge 12 is made of aluminum processed so as to have a predetermined target angle and a predetermined thickness, for example.

The collimator 13 is a mechanism for narrowing down the irradiation range of X-rays transmitted through the wedge 12. The collimator 13 narrows down the X-ray irradiation range (FOV) by forming a slit, for example, by combining a plurality of lead plates. The collimator 13 is sometimes called an X-ray diaphragm.

The X-ray high voltage device 14 includes, for example, a high voltage generator and an X-ray control device. The high voltage generator has an electric circuit including a transformer, a rectifier, and the like. The high voltage generator generates a high voltage applied to the X-ray tube 11. The X-ray control device controls the output voltage of the high voltage generator according to the X-ray dose to be generated in the X-ray tube 11. The high voltage generator may be one that boosts the voltage by the transformer described above, or may be one that boosts the voltage by an inverter. The X-ray high-voltage device 14 may be provided on the rotating frame 17, or may be provided on the side of the fixed frame (not shown) of the gantry device 10.

The X-ray detector 15 detects the intensity of X-rays generated by the X-ray tube 11 and passed through the subject P and incident. The X-ray detector 15 outputs an electric signal (or an optical signal or the like) corresponding to the intensity of the detected X-ray to the DAS 16. The X-ray detector 15 has, for example, a plurality of X-ray detection element sequences. Each of the plurality of X-ray detection element trains is an array of a plurality of X-ray detection elements in the channel direction along an arc centered on the focal point of the X-ray tube 11. The plurality of X-ray detection element sequences are arranged in the slice direction (column direction, row direction).

The X-ray detector 15 is, for example, an indirect detector having a grid, a scintillator array, and an optical sensor array. The scintillator array has a plurality of scintillators. Each scintillator has a scintillator crystal. The scintillator crystal emits an amount of light according to the intensity of the incident X-rays. The grid is arranged on the surface of the scintillator array where X-rays are incident, and has an X-ray shielding plate having a function of absorbing scattered X-rays. The grid may also be called a collimator (one-dimensional collimator or two-dimensional collimator). The optical sensor array includes, for example, an optical sensor such as a photomultiplier tube (PMT). The optical sensor array outputs an electric signal according to the amount of light emitted by the scintillator. The X-ray detector 15 may be a direct conversion type detector having a semiconductor element that converts incident X-rays into an electric signal.

The DAS 16 has, for example, an amplifier, an integrator, and an A / D converter. The amplifier performs amplification processing on the electric signal output by each X-ray detection element of the X-ray detector 15. The integrator integrates the amplified electrical signal over the view period (described later). The A / D converter converts an electric signal indicating the integration result into a digital signal. The DAS 16 outputs the detection data based on the digital signal to the console device 40. The detection data is a digital value of the X-ray intensity identified by the channel number, the column number, and the view number indicating the collected view of the X-ray detection element of the generator. The view number is a number that changes according to the rotation of the rotation frame 17, for example, a number that is incremented according to the rotation of the rotation frame 17. Therefore, the view number is information indicating the rotation angle of the X-ray tube 11. The view period is a period within a period from the rotation angle corresponding to a certain view number to the arrival at the rotation angle corresponding to the next view number. The DAS 16 may detect the change of view by a timing signal input from the control device 18, an internal timer, or a signal acquired from a sensor (not shown). .. When X-rays are continuously exposed by the X-ray tube 11 in the case of performing a full scan, the DAS 16 collects the detection data group for the entire circumference (360 degrees). When X-rays are continuously exposed by the X-ray tube 11 in the case of performing a half scan, the DAS 16 collects detection data for a half circumference (180 degrees).

The rotating frame 17 is an annular member that supports the X-ray tube 11, the wedge 12, the collimator 13, and the X-ray detector 15 so as to face each other. The rotating frame 17 is rotatably supported by the fixed frame around the subject P introduced inside. The rotating frame 17 further supports the DAS 16. The detection data output by the DAS 16 has a photodiode provided in a non-rotating portion (for example, a fixed frame) of the gantry device 10 by optical communication from a transmitter having a light emitting diode (LED) provided in the rotating frame 17. It is transmitted to the receiver and transferred to the console device 40 by the receiver. The method of transmitting the detection data from the rotating frame 17 to the non-rotating portion is not limited to the method using the above-mentioned optical communication, and any non-contact type transmitting method may be adopted. The rotating frame 17 is not limited to an annular member as long as it can support and rotate the X-ray tube 11 and the like, and may be a member such as an arm.

The X-ray CT apparatus 1 is, for example, a Rotate / Rotate-Type X-ray CT apparatus (third) in which both the X-ray tube 11 and the X-ray detector 15 are supported by the rotating frame 17 and rotate around the subject P. Generation CT), but not limited to this, in Stationary / Rotate-Type in which a plurality of X-ray detection elements arranged in an annular shape are fixed to a fixed frame and the X-ray tube 11 rotates around the subject P. It may be an X-ray CT apparatus (4th generation CT).

The control device 18 includes, for example, a processing circuit having a processor such as a CPU (Central Processing Unit) and a drive mechanism including a motor, an actuator, and the like. The control device 18 receives an input signal from the input interface 43 attached to the console device 40 or the gantry device 10 and controls the operation of the gantry device 10 and the sleeper device 30.

The control device 18 rotates, for example, the rotating frame 17, tilts the gantry device 10, and moves the top plate 33 of the sleeper device 30. When tilting the gantry device 10, the control device 18 rotates the rotating frame 17 around an axis parallel to the Z-axis direction based on the tilt angle (tilt angle) input to the input interface 43. The control device 18 grasps the rotation angle of the rotation frame 17 by the output of a sensor (not shown) or the like. Further, the control device 18 provides the processing circuit 50 with the rotation angle of the rotation frame 17 at any time. The control device 18 may be provided in the gantry device 10 or in the console device 40.

The sleeper device 30 is a device on which the subject P to be scanned is placed, moved, and introduced into the rotating frame 17 of the gantry device 10. The sleeper device 30 has, for example, a base 31, a sleeper drive device 32, a top plate 33, and a support frame 34. The base 31 includes a housing that movably supports the support frame 34 in the vertical direction (Y-axis direction). The sleeper drive device 32 includes a motor and an actuator. The sleeper drive device 32 moves the top plate 33 on which the subject P is placed in the longitudinal direction (Z-axis direction) of the top plate 33 along the support frame 34. The top plate 33 is a plate-shaped member on which the subject P is placed.

The sleeper drive device 32 may move not only the top plate 33 but also the support frame 34 in the longitudinal direction of the top plate 33. Further, contrary to the above, the gantry device 10 may be movable in the Z-axis direction, and the rotation frame 17 may be controlled to come around the subject P by the movement of the gantry device 10. Further, both the gantry device 10 and the top plate 33 may be movable. Further, the X-ray CT device 1 may be a device in which the subject P is scanned in a standing or sitting position. In this case, the X-ray CT device 1 has a subject support mechanism instead of the sleeper device 30, and the gantry device 10 rotates the rotating frame 17 about an axial direction perpendicular to the floor surface.

The console device 40 has, for example, a memory 41, a display 42, an input interface 43, a network connection circuit 44, and a processing circuit 50. In the first embodiment, the console device 40 will be described as a separate body from the gantry device 10, but the gantry device 10 may include a part or all of each component of the console device 40.

The memory 41 is realized by, for example, a RAM (Random Access Memory), a semiconductor memory element such as a flash memory, a hard disk, an optical disk, or the like. The memory 41 stores, for example, detection data, projection data, a reconstructed image (CT image), and the like. These data may be stored in an external memory with which the X-ray CT apparatus 1 can communicate, instead of the memory 41 (or in addition to the memory 41). The external memory is controlled by the cloud server, for example, when the cloud server that manages the external memory receives a read / write request. The external memory is realized by, for example, a system called PACS (Picture Archiving and Communication Systems). PACS is a system that systematically stores images and the like taken by various diagnostic imaging devices.

FIG. 2 is a diagram showing an example of data stored in the memory 41. As shown in FIG. 2, in the memory 41, for example, the shooting condition 41-1, the detection data 41-2 output by the DAS 16, the projection data 41-3 generated by the processing circuit 50, and the reconstructed image 41- 4. Information such as calibration data 41-5 and trained model 41-6 is stored. The trained model 41-6 will be described later.

The calibration data 41-5 refers to the deviation of the measured values caused by the individual difference of each unit of the X-ray CT apparatus 1 (difference in radiation quality, difference in influence of scattered radiation, etc.), difference in installation environment, etc. This is the data for correction. For example, when the X-ray CT apparatus 1 is installed or when hardware parts such as a tube are replaced, the calibration data is collected before the main imaging of the X-ray CT apparatus 1. This calibration data is prepared for each imaging condition of the X-ray CT apparatus 1, and is stored in the memory 41 in association with the imaging condition. For example, this calibration data includes the tube voltage (kV) applied to the X-ray tube 11, the number of columns detected by the X-ray detector 15, and the X-ray irradiation range (FOV: Field of View) at the time of photographing. Prepared for each condition (for each combination of conditions) such as the rotation speed of the X-ray tube 11, the setting (gain) of the X-ray detector 15, and the focal size irradiated by the X-ray tube 11, and the memory 41 is associated with the shooting conditions. Stored in.

For example, at the time of the main shooting, the deviation of the measured value is corrected by performing a process of subtracting the calibration data associated with the same condition from the detection data 41-2 detected by the shooting under a certain shooting condition. That is, the data obtained by scanning the subject under the first scan condition is corrected by using the first calibration data corresponding to the first scan condition, and the second scan condition is used. The deviation of the measured value is corrected by correcting the data obtained by scanning the subject using the second calibration data corresponding to the second scan condition.

FIG. 3 is a diagram showing an example of calibration data stored in the memory 41 for each shooting condition. As shown in the figure, for example, the first condition (condition 1 (tube voltage kV) is "120 kV", condition 2 (number of columns) is "80", condition 3 (FOV) is "500 mm") and the calibration data " It is stored in the memory 41 in association with "CD1".

The display 42 displays various information. For example, the display 42 displays a medical image (CT image) generated by the processing circuit 50, a GUI (Graphical User Interface) image that accepts various operations by the operator, and the like. The display 42 is, for example, a liquid crystal display, a CRT (Cathode Ray Tube), an organic EL (Electroluminescence) display, or the like. The display 42 may be provided on the gantry device 10. The display 42 may be a desktop type or a display device (for example, a tablet terminal) capable of wirelessly communicating with the main body of the console device 40.

The input interface 43 accepts various input operations by the operator and outputs an electric signal indicating the contents of the received input operations to the processing circuit 50. For example, the input interface 43 is an input operation such as a collection condition when collecting detection data or projection data, a reconstruction condition when reconstructing a CT image, and an image processing condition when generating a post-processed image from a CT image. Accept. For example, the input interface 43 is realized by a mouse, a keyboard, a touch panel, a trackball, a switch, a button, a joystick, a camera, an infrared sensor, a microphone, and the like. The input interface 43 may be provided on the gantry device 10. Further, the input interface 43 may be realized by a display device (for example, a tablet terminal) capable of wireless communication with the main body of the console device 40.

The network connection circuit 44 includes, for example, a network card having a printed circuit board, a wireless communication module, and the like. The network connection circuit 44 implements an information communication protocol according to the form of the network to be connected. The network includes, for example, a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, a cellular network, a dedicated line, and the like.

The processing circuit 50 controls the overall operation of the X-ray CT apparatus 1. The processing circuit 50 executes, for example, a system control function 51, a preprocessing function 52, a reconstruction processing function 53, an image processing function 54, a scan control function 55, a display control function 56, a calibration function 57, a learning function 58, and the like. .. The processing circuit 50 realizes these functions by, for example, the hardware processor executing a program stored in the memory 41. The learning function 58 is an example of a “learning unit”.

The hardware processor 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 (SPLD)) or It means a circuit (circuitry) such as a complex programmable logic device (CPLD) and a field programmable gate array (FPGA). Instead of storing the program in the memory 41, the program may be configured to be directly embedded in the circuit of the hardware processor. In this case, the hardware processor realizes the function by reading and executing the program embedded in the circuit. The hardware processor is not limited to one configured as a single circuit, and may be configured as one hardware processor by combining a plurality of independent circuits to realize each function. Further, a plurality of components may be integrated into one hardware processor to realize each function.

Each component of the console device 40 or the processing circuit 50 may be distributed and implemented by a plurality of hardware. The processing circuit 50 may be realized not by the configuration of the console device 40 but by a processing device capable of communicating with the console device 40. The processing device is, for example, a workstation connected to one X-ray CT device, or a device connected to a plurality of X-ray CT devices and collectively executing processing equivalent to the processing circuit 50 described below (for example,). , Cloud server). That is, the configuration of the present embodiment can be realized as an X-ray CT system (medical diagnostic system) in which an X-ray CT apparatus and another processing apparatus are connected via a network.

The system control function 51 controls various functions of the processing circuit 50, for example, based on the input operation received by the input interface 43.

The preprocessing function 52 preprocesses the detection data 41-2 output by the DAS 16 before logarithmic conversion processing, offset correction processing, interchannel sensitivity correction processing, beam hardening correction, correction processing using calibration data, and the like. The processing is performed to generate projection data 41-3, and the generated projection data 41-3 is stored in the memory 41. The correction process using the calibration data may be performed by the reconstruction process function 53.

The reconstruction processing function 53 performs reconstruction processing on the projection data 41-3 generated by the preprocessing function 52 by a filter correction back projection method, a successive approximation reconstruction method, or the like to obtain the reconstruction image 41-4. The generated reconstructed image 41-4 is stored in the memory 41.

The image processing function 54 converts the reconstructed image 41-4 into a three-dimensional image or cross-sectional image data of an arbitrary cross section by a known method based on the input operation received by the input interface 43. The conversion to a three-dimensional image may be performed by the preprocessing function 52.

The scan control function 55 controls the collection process of the detection data in the gantry device 10 by instructing the X-ray high voltage device 14, the DAS 16, the control device 18, and the sleeper drive device 32. The scan control function 55 controls the operation of each part at the time of capturing the alignment data, the main captured image, and the image used for diagnosis, and collecting the calibration data.

The display control function 56 controls the display mode of the display 42. For example, the display control function 56 controls the display 42 to display a CT image generated by the processing circuit 50, a GUI image that accepts various operations by the operator, and the like.

FIG. 4 is a diagram showing an example of the functional configuration of the calibration function 57. The calibration function 57 has, for example, a data acquisition function 57-1 and a data generation function 57-2. The data acquisition function 57-1 collects calibration data (hereinafter referred to as “reference calibration data”) from the gantry device 10 that operates under a certain reference condition, and stores it in the memory 41 as the reference calibration data BD. The data generation function 57-2 calibrates conditions different from the reference conditions based on the reference calibration data BD acquired by the data acquisition function 57-1 and the trained model 41-6 stored in the memory 41. Generation data (hereinafter referred to as "generation calibration data"). The data acquisition function 57-1 is an example of a “data acquisition unit”. The data generation function 57-2 is an example of a “data generation unit”.

The reference calibration data BD is detected by the gantry device 10 that operates under the reference conditions with the calibration phantom (for example, a water phantom) placed on the top plate 33 under the control of the scan control function 55. It is the detection data (actual data) to be performed. The reference condition is determined based on, for example, the operation of the input interface 43 by the operator. On the other hand, the generated calibration data GD is data (estimated data) output when the reference calibration data BD is input to the trained model 41-6.

The learning function 58 learns the relationship between the first calibration data corresponding to the first scan condition and the second calibration data corresponding to the second scan condition different from the first scan condition. , Generate a trained model. The learning function 58 generates a trained model for generating the second calibration data based on the first calibration data. For example, the learning function 58 sets various parameters in the model by SGD so that the difference between the data output when the first calibration data is input to the model and the second calibration data becomes small. (Stochastic Gradient Descent), Momentum SGD, AdaGrad, RMSprop, AdaDelta, Adam (Adaptive moment estimation) and other gradient methods are used for learning. The learning function 58 may use a model generated by any machine learning, such as a neural network, logistic regression analysis, or a technique based on a support vector machine. The details of the learning function 58 will be described later.

With the above configuration, the X-ray CT apparatus 1 scans the subject P in a manner such as a helical scan, a conventional scan, and a step-and-shoot. The helical scan is a mode in which the rotating frame 17 is rotated while the top plate 33 is moved to spirally scan the subject P. The conventional scan is a mode in which the rotating frame 17 is rotated while the top plate 33 is stationary to scan the subject P in a circular orbit. Perform a conventional scan. The step-and-shoot is a mode in which the position of the top plate 33 is moved at regular intervals to perform a conventional scan in a plurality of scan areas.

[Processing flow (learning stage)]
Hereinafter, the processing flow of the processing circuit 50 in the first embodiment will be described. The process of the processing circuit 50 includes a learning process of generating a trained model and a generation process of generating generation calibration data using the trained model. In the following, first, the learning process by the processing circuit 50 will be described. FIG. 5 is a flowchart showing a series of flow of learning processing of the processing circuit 50. FIG. 6 is a diagram for explaining the concept of learning processing by the learning function 58. In the processing of the flowchart shown in FIG. 5, for example, at the time of shipment of the X-ray CT device 1 (before installation in the usage environment), the operator of the X-ray CT device 1 operates the input interface 43 to perform the learning process. It is started when the start is instructed. In the following, it is assumed that the calibration data for each shooting condition (for example, for each tube voltage (kV)), which is the learning data, has already been collected and stored in the memory 41. The calibration data for each shooting condition is one or a plurality of data (data sets). For example, the calibration data corresponding to a certain scan condition (tube voltage is “120 kV”) may be a plurality of data (data sets).

First, the learning function 58 acquires a set of learning data from the calibration data (detection data) stored in the memory 41 (step S100). This set of training data includes, for example, a first calibration data corresponding to the first scan condition and a second calibration data corresponding to a second scan condition different from the first scan condition. It is a group. In the example shown in FIG. 6, for example, the learning function 58 has the first calibration data D1 corresponding to the first scan condition (tube voltage is “120 kV”) and the second scan condition (tube voltage is “80 kV”). ”), And the second calibration data D2 corresponding to the above is acquired. That is, the first scan condition and the second scan condition each include the same type of parameter (tube voltage), and the values of the same type of parameter are different between the first scan condition and the second scan condition. different.

Next, the learning function 58 inputs the first calibration data corresponding to the first scan condition into the model and obtains the processing result (output result) (step S102). In the example shown in FIG. 6, the learning function 58 inputs the first calibration data D1 to the model M1 and obtains the processing result thereof.

Next, the learning function 58 calculates the difference between the processing result data and the second calibration data D2 (step S104).

Next, the learning function 58 updates the internal parameters of the model M1 based on the calculated difference (step S106). For example, the learning function 58 updates the internal parameters of the model M1 so that the calculated difference becomes small. The first model M1 whose internal parameters have been updated in this way becomes the first trained model. The learning function 58 stores the first trained model in the memory 41 (step S108).

Next, the learning function 58 determines whether or not the processing has been completed for all the sets of learning data (step S110). In the example shown in FIG. 6, the first calibration data D1 and the third calibration corresponding to the third scan condition (tube voltage is “100 kV”) different from each of the first and second scan conditions. The process for the set with the data D3, and the first calibration data D1 and the fourth scan condition (tube voltage is "135 kV") different from each of the first to third scan conditions. The process for the set with the calibration data D4 of 4 is not completed. In this case, the learning function 58 repeats the same processing after step S100 for these sets as well. As a result, the second trained model and the third trained model, which are different from the first trained model, are stored in the memory 41.

On the other hand, when it is determined that the processing is completed for all the learning data, the learning function 58 ends the processing of this flowchart.

After the X-ray CT device 1 is installed in the usage environment, the trained model may be generated using the calibration data acquired by the X-ray CT device 1. As a result, it is possible to generate a highly accurate trained model in consideration of the situation (installation environment) in which the X-ray CT apparatus 1 performs main imaging.

When a general-purpose trained model generated by a model generation device different from the X-ray CT device 1 is stored in the memory 41 of the X-ray CT device 1 and shipped, the learning function 58 described above is used. The learning process does not have to be performed. Alternatively, the learning function 58 may further learn the calibration data acquired by the X-ray CT apparatus 1 with respect to this general-purpose trained model. As a result, it is possible to generate a highly accurate trained model in consideration of the individual difference of the X-ray CT apparatus 1 (individual difference of each unit).

[Processing flow (calibration data generation processing)]
Next, the calibration data generation process by the processing circuit 50 will be described. FIG. 7 is a flowchart showing a series of flow of calibration data generation processing by the processing circuit 50. FIG. 8 is a diagram for explaining the concept of the calibration data generation process by the calibration function 57. In the processing of this flowchart, for example, when the X-ray CT device 1 is installed or after the hardware parts are replaced, the operator of the X-ray CT device 1 operates the input interface 43 to generate calibration data. It is started when the start is instructed.

First, the calibration function 57 (data acquisition function 57-1) provides reference calibration data in the X-ray CT apparatus 1 in which a calibration phantom (for example, a water phantom) is placed on the top plate 33. Acquire BD1 (step S200). The reference condition (for example, tube voltage (kV)) for acquiring the reference calibration data is determined based on, for example, the operation of the input interface 43 by the operator. In the example shown in FIG. 8, the tube voltage is set to 120 kV as a reference condition.

Next, the calibration function 57 (data generation function 57-2) has been trained to output the required calibration data from the plurality of trained models 41-6 stored in the memory 41. Models 41-6 are selected (step S202). The trained model 41-6 may be selected based on, for example, the operation of the input interface 43 by the operator. Further, the preset trained models 41-6 may be selected, or all the trained models 41-6 stored in the memory 41 may be selected. In the example shown in FIG. 8, the first trained model TM1, the second trained model TM2, and the third trained model TM3 are selected.

Next, the calibration function 57 inputs the acquired reference calibration data BD to each of the selected trained models 41-6, and generates the generated calibration data GD which is the output of the model (step S204). ..

In the example shown in FIG. 8, the first trained model TM1 is trained to generate generated calibration data (tube voltage 80 kV) based on the reference calibration data BD (tube voltage 120 kV). .. The calibration function 57 generates the first generated calibration data GD1 (tube voltage is 80 kV) by inputting the reference calibration data BD into the first trained model TM1. Further, the second trained model TM2 is trained to generate generation calibration data (tube voltage is 100 kV) based on the reference calibration data BD (tube voltage is 120 kV). The calibration function 57 generates the second generated calibration data GD2 (tube voltage is 100 kV) by inputting the reference calibration data BD into the second trained model TM2. Further, the third trained model TM3 is trained to generate generated calibration data (tube voltage is 135 kV) based on the reference calibration data BD (tube voltage is 120 kV). The calibration function 57 generates the third generated calibration data GD3 (tube voltage is 135 kV) by inputting the reference calibration data BD into the third trained model TM3.

Next, the calibration function 57 stores the generated generated calibration data GD in the memory 41 (step S206). As described above, the calibration function 57 ends the process of this flowchart.

According to the X-ray CT apparatus 1 of the first embodiment described above, the calibration time can be shortened by generating a plurality of generated calibration data based on the reference calibration data. In addition, it is possible to generate optimum calibration data according to the state (installation environment) when the X-ray CT apparatus 1 after installation performs imaging.

In the above, an example of individually generating trained models (generating a plurality of trained models) for each shooting condition has been described, but the present invention is not limited to this. FIG. 9 is a diagram illustrating another example of the calibration data generation process by the calibration function 57. As shown in the figure, the learning function 58 is one that outputs a plurality of generated calibration data GDs (GD1 to GD3) associated with a plurality of shooting conditions when the reference calibration data BD is input. The trained model TM4 may be generated. The calibration function 57 can generate a plurality of generated calibration data GDs (GD1 to GD3) at once by inputting the reference calibration data BD to the trained model TM4.

Also, for example, by using a trained model trained to generate generated calibration data (320 columns) based on the reference calibration data BD (80 columns), the number of columns There is no need to prepare a phantom for shooting conditions in which is "320 rows". As a result, the size of the calibration phantom can be reduced. Further, the calibration data obtained by taking a picture of air without using the phantom (that is, the calibration data obtained by taking a picture with nothing on the top plate 33). By generating a trained model that has learned the above, it is possible to eliminate the use of the phantom during calibration.

(Second Embodiment)
Hereinafter, the second embodiment will be described. In the first embodiment described above, attention is paid to one imaging condition (one parameter, for example, tube voltage (kV)), and from the reference calibration data (for example, tube voltage 120 kV) under a certain reference condition, another An example of generating generation calibration data (for example, a tube voltage of 80 kV) associated with the above conditions has been described. On the other hand, in the second embodiment, attention is paid to a plurality of imaging conditions (two or more parameters, for example, tube voltage (kV) and the number of columns), and reference calibration is performed under certain reference conditions (two or more parameters). Calibration data associated with other conditions (two or more parameters) is generated from the voltage data. Therefore, for the configuration and the like, the drawings and related descriptions described in the first embodiment will be referred to, and detailed description will be omitted.

[Learning process]
FIG. 10 is a diagram illustrating the concept of learning processing by the learning function 58 in the second embodiment. As shown in the figure, the learning function 58 includes a first calibration data PD1 corresponding to the first scan condition (tube voltage is “120 kV”, number of columns is “320 columns”) and a second scan condition (tube). The first trained model is generated by learning the relationship with the second calibration data PD2 corresponding to the voltage “80 kV” and the number of columns “80 columns”). Further, the learning function 58 has a first calibration data PD1 corresponding to the first scan condition and a third scan condition (tube voltage is “80 kV”, number of rows is “160 rows”) corresponding to the first calibration data PD1. A second trained model is generated by learning the relationship with the calibration data PD3. Further, the learning function 58 corresponds to the first calibration data PD1 corresponding to the first scan condition and the fourth scan condition (tube voltage is “80 kV”, number of rows of detection elements is “320 rows”). By learning the relationship with the third calibration data PD3 to be performed, a third trained model is generated.

As described above, in the learning data learned by the learning function 58, each of the first scan condition to the fourth scan condition includes a plurality of parameters. Further, the value of at least one of the plurality of parameters is different between the first scan condition and each of the second to fourth scan conditions. The first scan condition and the second scan condition are conditions related to the X-ray CT system. At least one parameter is the tube voltage applied to the X-ray tube 11 included in the X-ray CT system, the rotation speed of the X-ray tube 11, the focal size of the X-ray emitted from the X-ray tube 11, and the quality of the X-ray. Includes the type of filter (eg, wedge 12) for adjusting, the number of columns that detect X-rays with the X-ray detector 15 included in the X-ray CT system, or the gain of the X-ray detector 15.

[Calibration data generation process]
FIG. 11 is a diagram illustrating a concept of calibration data generation processing by the calibration function 57 in the second embodiment. As shown in the figure, the calibration function 57 inputs the reference calibration data PBD (tube voltage is "120 kV", number of columns is "320 columns") into the first trained model TPM1 to obtain the first one. Generation Calibration data PGD1 (tube voltage is "80 kV", number of columns is "80 columns") is generated. Further, the calibration function 57 inputs the reference calibration data BD to the second trained model TPM2 to generate the second generated calibration data PGD1 (tube voltage is “80 kV” and the number of columns is “160 columns”. ") Is generated. Further, the calibration function 57 inputs the reference calibration data BD to the third trained model TPM3 to generate the third generated calibration data PGD3 (tube voltage is “80 kV” and the number of columns is “320 columns”. ") Is generated.

According to the X-ray CT apparatus 1 of the second embodiment described above, the calibration time can be shortened by generating a plurality of generated calibration data based on the reference calibration data. In addition, it is possible to generate optimum calibration data according to the state (installation environment) when the X-ray CT apparatus 1 after installation performs imaging. In addition, by focusing on multiple shooting conditions and generating calibration data associated with other conditions from the reference calibration data of a certain reference condition, calibration data corresponding to various shooting conditions can be generated in a short time. Can be prepared.

In the above embodiment, an example in which a trained model is generated in the X-ray CT apparatus 1 at the time of shipment or the trained model is changed according to each X-ray CT apparatus 1 has been described, but the present invention is limited to this. I can't. For example, as shown in FIG. 12, the X-ray CT device 1 installed in each medical facility H1 to HN (N is an arbitrary natural number) and the server device S on the cloud are connected via the network NW. The trained model may be provided from the server device S to the X-ray CT device 1. Further, the X-ray CT device 1 transmits the reference calibration data acquired by the X-ray CT device 1 after installation to the server device S, and the server device uses the trained model stored in the server device S. The generated calibration data may be generated in S and transmitted to the X-ray CT apparatus 1. The combination of the X-ray CT apparatus 1 and the server apparatus S is an example of a "medical processing apparatus" or a "medical diagnostic system".

At the time of shipment of the X-ray CT device 1, a general-purpose learned model generated by a model generation device separate from the X-ray CT device 1 is stored in the memory 41 of the X-ray CT device 1. In the case of the embodiment, by comparing the training data used for model generation in another model generation device with the data of the reference conditions obtained from the X-ray CT device 1 after installation, the individual X-ray CT devices It is possible to grasp the degree of variation in data. For example, by grasping the variation of the tube, it is possible to grasp the variation of the radiation quality, so that it can be expanded to the correction for making the CT value uniform.

In addition, it is known that focus shift occurs due to OLP (Over Load Protection) of the tube, but by learning the correlation between OLP and data, calibration is performed from the difference between OLP at the time of calibration and OLP at the time of actual shooting. The data can be corrected. In addition, since the quality of the radiation changes depending on the tube alignment, the parameters can be optimized from the data at the time of alignment. In addition, by learning the learning data associated with each temperature not only for the tube but also for changes on the detection element side such as the temperature drift of the detection element, calibration data according to the temperature can be prepared in a short time. can do. In addition, calibration data corresponding to various shooting conditions such as extraction of ring components and steps between packs can be prepared in a short time.

The embodiment described above can be expressed as follows.
The memory to store the program and
With a processor,
By executing the program, the processor
Acquire the first calibration data corresponding to the first scan condition,
For the first trained model for generating the second calibration data corresponding to the second scan condition different from the first scan condition based on the first calibration data. By inputting the first calibration data, the second calibration data is generated.
Medical processing equipment.

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 embodiments, and various mitigations, replacements, and modifications can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention as well as the invention described in the claims and the equivalent scope thereof.

1 ... X-ray CT device, 10 ... gantry device, 11 ... X-ray tube, 12 ... wedge, 13 ... collimeter, 14 ... X-ray high voltage device, 15 ... X-ray detector, 16 ... data acquisition system, 17 ... rotation Frame, 18 ... control device, 30 ... sleeper device, 31 ... base, 32 ... sleeper drive device, 33 ... top plate, 34 ... support frame, 40 ... console device, 50 ... processing circuit, 51 ... system control function, 52 ... preprocessing function, 53 ... reconstruction processing function, 54 ... image processing function, 55 ... scan control function, 56 ... display control function, 57 ... calibration function, 58 ... learning function

Claims (12)

A data acquisition unit that acquires the first calibration data corresponding to the first scan condition, and
For the first trained model for generating the second calibration data corresponding to the second scan condition different from the first scan condition based on the first calibration data. A data generation unit that generates the second calibration data by inputting the first calibration data, and
A medical processing device equipped with. Based on the first calibration data, the data generation unit corresponds to a third scan condition corresponding to the first scan condition and a third scan condition different from the second scan condition. By inputting the first calibration data to the second trained model different from the first trained model for generating the calibration data of 3, the third calibration The medical processing apparatus according to claim 1, which generates data. Based on the first calibration data, the first trained model is set to a third scan condition corresponding to the first scan condition and a third scan condition different from the second scan condition. It is a model for further generating the corresponding third calibration data.
The data generation unit generates the second calibration data and the third calibration data by inputting the first calibration data into the first trained model. Item 1. The medical processing apparatus according to item 1. The first scan condition and the second scan condition each include a plurality of parameters.
The value of at least one of the plurality of parameters differs between the first scan condition and the second scan condition.
The medical processing device according to claim 1. The first scan condition and the second scan condition are conditions related to an X-ray CT system.
The at least one parameter includes the tube voltage applied to the X-ray tube included in the X-ray CT system, the rotation speed of the X-ray tube, the focal size of the X-ray emitted from the X-ray tube, and the X-ray. The medical use according to claim 4, which includes the type of filter for adjusting the radiation quality, the number of columns for detecting X-rays by the X-ray detector included in the X-ray CT system, or the gain of the X-ray detector. Processing equipment. By correcting the data obtained by scanning the subject under the first scan condition using the first calibration data and scanning the subject under the second scan condition. The medical processing apparatus according to claim 1, wherein the obtained data is corrected by using the second calibration data. The first scan condition and the second scan condition each include the same type of parameters.
The values of the same type of parameters differ from each other between the first scan condition and the second scan condition.
The medical processing device according to claim 1. A learning unit that generates the first trained model by learning the relationship between the first calibration data and the second calibration data is further provided.
The medical processing device according to any one of claims 1 to 7. The learning unit has learned the first learning by learning the relationship between the first calibration data and the second calibration data acquired from the medical processing device before installation in the usage environment. Generate a model,
The medical processing device according to claim 8. The learning unit learns the relationship between the first calibration data and the second calibration data acquired from the medical processing device installed in the usage environment, thereby learning the first learned model. To generate,
The medical processing device according to claim 8. The first trained model is provided by a server device connected to the medical processing device via a network.
The medical processing device according to any one of claims 1 to 10. A medical diagnostic system comprising the medical processing apparatus according to any one of claims 1 to 11.

Images