Medical image diagnostic device, medical image diagnostic method, and program

Abstract

To more efficiently make a determination of different diseases based on an image photographed by a medical image diagnostic device.SOLUTION: A medical image diagnostic device of an embodiment has an acquisition unit, a first derivation unit, a selection unit, and a second derivation unit. The acquisition unit acquires first subject information, a first photographing condition, and first photographing data. The first derivation unit derives the probability of detection of a plurality of diseases, by inputting the first subject information and the first photographing condition to each of a plurality of first models that, upon input of the subject information and the photographing condition, output the probability of detection of a corresponding disease. The selection unit selects a specific disease, from among the plurality of diseases, on the basis of the probability of detection of the plurality of diseases. The second derivation unit derives an index related to the specific disease, by inputting the first photographing data to a second model that is for the specific disease selected by the selection unit and that, upon input of the photographing data, outputs a second index related to the corresponding disease.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The embodiments disclosed in the present specification and drawings relate to a medical image diagnostic apparatus, a medical image diagnostic method, and a program. [Background technology]

[0002] In the field of medical image diagnosis, medical image diagnostic devices, such as X-ray CT (Computed Tomography) devices, have been used to make diagnoses using images of the inside of a subject (patient). In a diagnosis using a medical image diagnostic device, a diagnostician such as a doctor requests imaging by specifying the area of ​​the patient to be imaged. Then, a user of the medical image diagnostic device, such as a radiologist or technician, confirms that there are no problems with the reconstructed image generated based on the projection data obtained by imaging the patient, then performs post-processing such as image shaping to generate a CT image, and transfers the generated CT image to the diagnostician who requested the patient imaging. The diagnostician then diagnoses the patient based on the patient information and the transferred CT image.

[0003] Incidentally, projection data from a patient's scans also includes images of the area surrounding the region designated by the diagnostician. These images may be useful for detecting diseases other than the target disease. However, if the user discards the surrounding images from the CT image through post-processing, it becomes impossible to detect the disease other than the target disease. For this reason, in recent years, deep learning, a type of machine learning function based on artificial intelligence (AI), has been used to estimate whether the surrounding images in the projection data contain diseases other than the target disease. However, to improve the accuracy of diagnoses, the prevailing trained models used for estimation are models trained only for the specific disease, i.e., models trained specifically for the specific disease to be diagnosed. Therefore, when performing estimation on the surrounding images, many trained models corresponding to candidate diseases (diseases other than the target disease) are required. This raises concerns about the processing load and processing time required to obtain estimation results that identify the presence or absence of all diseases. In other words, there is a concern that estimation on the surrounding images will result in high processing costs. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Republished WO2020 / 110214 Summary of the Invention [Problem to be solved by the invention]

[0005] The problem to be solved by the embodiments disclosed in this specification and the drawings is to more efficiently diagnose different diseases based on images captured by a medical image diagnostic device. However, the problem to be solved by the embodiments disclosed in this specification and the drawings is not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0006] A medical image diagnostic apparatus according to an embodiment includes an acquisition unit, a first derivation unit, a selection unit, and a second derivation unit. The acquisition unit acquires first subject information about a subject, first imaging conditions for imaging the subject, and first imaging data obtained by imaging the subject. The first derivation unit derives detection probabilities for multiple diseases by inputting the first subject information and the first imaging conditions into each of multiple first models that output a detection probability for a corresponding disease in response to input of the subject information and the imaging conditions. The selection unit selects a specific disease from the multiple diseases based on the detection probabilities for the multiple diseases. The second derivation unit derives an index for the specific disease by inputting the first imaging data into a second model for the specific disease selected by the selection unit that outputs a second index for the corresponding disease in response to input of the imaging data. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a diagram showing an example of the arrangement of a medical image diagnostic apparatus according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an example of the functional configuration of a determination function included in the medical image diagnostic apparatus according to the embodiment. [Figure 3] FIG. 2 is a diagram schematically showing an example of a method for generating a probability calculation model used in a detection probability calculation function included in the medical image diagnostic apparatus according to the embodiment. [Figure 4] 10 is a flowchart showing an example of a processing flow for determining the presence or absence of different diseases in a determination function included in the medical image diagnostic apparatus according to the embodiment. [Figure 5]FIG. 10 is a diagram schematically showing an example of a method for calculating the detection probability of different diseases in the detection probability calculation function included in the medical image diagnostic apparatus according to the embodiment. [Figure 6] FIG. 2 is a diagram schematically showing an example of a method for diagnosing different diseases in a disease diagnosing function included in the medical image diagnostic apparatus according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, a medical image diagnostic apparatus, a medical image diagnostic method, and a program according to embodiments will be described with reference to the drawings. In the following description, an example will be given in which the medical image diagnostic apparatus is an X-ray CT (Computed Tomography) apparatus for CT diagnosis, and the medical image is a CT image.

[0009] FIG. 1 is a diagram illustrating an example of the configuration of a medical image diagnostic apparatus (X-ray CT apparatus) according to an embodiment. The X-ray CT apparatus 1 is a medical image diagnostic apparatus that irradiates a patient P, who is an object, with X-rays and detects the X-rays that have passed through the patient P. The X-ray CT apparatus 1 generates and displays a reconstructed image corresponding to the detected X-rays. This allows a user of the X-ray CT apparatus, such as a radiologist or technician, to visually check whether the reconstructed image has been captured correctly. Based on the result, the X-ray CT apparatus 1 performs post-processing on the reconstructed image to generate a CT image and transmits the CT image to a diagnostician. Here, post-processing refers to image processing performed on the reconstructed image, such as creating a three-dimensional (3D) object model, rotating the image, cropping areas other than the target area, and creating a multiplanar reconstruction (MPR) image. This allows the diagnostician to visually check whether the patient P has a lesion.

[0010] The X-ray CT apparatus 1 includes, for example, a gantry 10, a bed 30, and a console 40. For convenience of explanation, FIG. 1 shows the gantry 10 viewed from both the Z-axis direction and the X-axis direction, but in reality, the X-ray CT apparatus 1 includes only one gantry 10. In this embodiment, the central axis of the rotating frame 17 in a non-tilted state or the longitudinal direction of the tabletop 33 of the bed 30 (the body axis direction of the patient P) is defined as the Z-axis direction, an axis perpendicular to the Z-axis direction and horizontal to the floor surface is defined as the X-axis direction, and a direction perpendicular to the Z-axis direction and perpendicular to the floor surface is defined as the Y-axis direction. The X-ray CT apparatus 1 is an example of a "medical image diagnostic apparatus."

[0011] The gantry device 10 includes, for example, an X-ray tube device 11 incorporating an X-ray tube, a wedge 12, a collimator 13, an X-ray high voltage device 14, an X-ray detector 15, a data acquisition system (hereinafter referred to as DAS: Data Acquisition System) 16, a rotating frame 17, and a control device 18.

[0012] X-ray tube device 11 generates X-rays by causing a built-in X-ray tube to emit thermoelectrons from a cathode (filament) toward an anode (target) in response to a high tube voltage applied by X-ray high voltage device 14. X-ray tube device 11 includes, for example, a vacuum tube as an X-ray tube. In the following description, for ease of explanation, X-ray tube device 11 will be described as an X-ray tube. X-ray tube device 11 is, for example, a rotating anode type X-ray tube that generates X-rays by emitting thermoelectrons from a cathode to a rotating anode. The X-rays generated by X-ray tube device 11 are irradiated onto patient P.

[0013] The wedge 12 is a filter for adjusting the dose (X-ray dose) when the patient P is irradiated with X-rays generated by the X-ray tube device 11. The wedge 12 attenuates the X-rays that pass through it so that the distribution of the X-ray dose irradiated to the patient P becomes a predetermined distribution. The wedge 12 is also called a wedge filter or a bow-tie filter. The wedge 12 is made by processing aluminum so as to have a predetermined target angle and a predetermined thickness, for example.

[0014] The collimator 13 is a mechanism for narrowing the irradiation range of the X-rays that have passed through the wedge 12. The collimator 13 narrows the irradiation range of the X-rays, for example, by combining multiple lead plates to form a slit. The collimator 13 is sometimes called an X-ray aperture. The collimator 13 may be an active collimator whose narrowing range can be mechanically driven.

[0015] 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 and a rectifier, and generates a high voltage to be applied to the X-ray tube device 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 device 11. The high voltage generator may be one that boosts voltage using the above-mentioned transformer, or one that boosts voltage using 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 a fixed frame (not shown) provided on the gantry device 10.

[0016] The X-ray detector 15 detects the intensity of X-rays that are generated by the X-ray tube device 11 and that have passed through the patient P and are incident thereon. The X-ray detector 15 outputs an electrical signal (which may be an optical signal, etc.) corresponding to the intensity of the detected X-rays to the DAS 16. The X-ray detector 15 has, for example, multiple X-ray detection element rows. Each of the multiple X-ray detection element rows has multiple X-ray detection elements arranged in the channel direction along an arc centered on the focal point of the X-ray tube device 11. The multiple X-ray detection element rows are arranged in the slice direction (column direction, row direction).

[0017] The X-ray detector 15 is, for example, an indirect detector having a grid, a scintillator array, and a photosensor array. The scintillator array has multiple scintillators. Each scintillator has scintillator crystals. The scintillator crystals emit light with an amount of light corresponding to the intensity of the incident X-rays. The grid is arranged on the surface of the scintillator array on which X-rays are incident and has an X-ray shielding plate that absorbs scattered X-rays. The grid is sometimes called a collimator (one-dimensional collimator or two-dimensional collimator). The photosensor array has, for example, a photosensor such as a photomultiplier tube (PMT). The photosensor array outputs an electrical signal corresponding to the amount of light emitted by the scintillator. The X-ray detector 15 may also be a direct conversion detector having a semiconductor element that converts incident X-rays into an electrical signal.

[0018] The DAS 16 includes, for example, an amplifier, an integrator, and an A / D converter. The amplifier amplifies the electrical signal output by each X-ray detection element of the X-ray detector 15. The integrator integrates the electrical signal amplified by the amplifier over a view period (described later). The A / D converter converts the electrical signal indicating the integration result by the integrator into a digital signal. The DAS 16 outputs detection data based on the digital signal to the console device 40. The detection data is a digital value of X-ray intensity identified by the channel number and column number of the X-ray detection element that generated the data, and a view number indicating the acquired view. The view number is a number that changes according to the rotation of the rotating frame 17, and is, for example, a number that is incremented according to the rotation of the rotating frame 17. Therefore, the view number is information indicating the rotation angle of the X-ray tube device 11. The view period is the period from the rotation angle corresponding to a certain view number to the rotation angle corresponding to the next view number. In other words, the view period is the period required to obtain detection data for one irradiation time (projection period) of X-rays irradiated from the X-ray tube device 11 to the patient P at the same rotation angle. The DAS 16 may detect the view switch by a timing signal input from the control device 18, by an internal timer, or by a signal acquired from a sensor (not shown). When the X-ray CT device 1 performs a full scan and X-rays are continuously irradiated by the X-ray tube device 11, the DAS 16 collects a group of detection data for the entire circumference (360 degrees). When the X-ray CT device 1 performs a half scan and X-rays are continuously irradiated by the X-ray tube device 11, the DAS 16 collects detection data for half the circumference (180 degrees).

[0019] The rotating frame 17 is an annular member that supports the X-ray tube assembly 11, the wedge 12, the collimator 13, and the X-ray detector 15 in opposing positions. The rotating frame 17 is supported by a fixed frame so as to be rotatable around the patient P placed inside. The rotating frame 17 also supports the DAS 16. Detection data output by the DAS 16 is transmitted by optical communication from a transmitter having a light-emitting diode (LED) provided on the rotating frame 17 to a receiver having a photodiode provided on a non-rotating portion of the gantry device 10 (e.g., a fixed frame not shown), and then 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 optical communication method described above, and any non-contact transmission method may be used. The rotating frame 17 is not limited to an annular member, and may be an arm-like member as long as it can support and rotate the X-ray tube assembly 11 and the like.

[0020] The X-ray CT device 1 is, for example, a rotate / rotate-type X-ray CT device (third generation CT) in which both the X-ray tube device 11 and the X-ray detector 15 are supported by a rotating frame 17 and rotate around the patient P, but is not limited to this and may also be a stationary / rotate-type X-ray CT device (fourth generation CT) in which multiple X-ray detection elements arranged in a circular ring are fixed to a fixed frame and the X-ray tube device 11 rotates around the patient P.

[0021] The control device 18 receives input signals from an input interface (not shown), such as an operation switch, attached to the gantry 10 or an input interface 43 attached to the console device 40, and controls the operation of the gantry 10 and the bed device 30. The control device 18 includes a processing circuit having a processor, such as a CPU (Central Processing Unit), and a drive mechanism including, for example, a motor and an actuator, for moving the gantry 10, rotating the rotating frame 17 of the gantry 10, and moving the bed device 30. Although the embodiment illustrates a case in which the control device 18 is provided in the gantry 10, the control device 18 may also be provided in the console device 40. In this specification, the input interface is not limited to one having physical operation components such as a mouse and a keyboard. For example, an example of an input interface also includes an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs the electrical signal to a control circuit.

[0022] The control device 18, for example, rotates the rotating frame 17, tilts the gantry 10, moves the housing of the gantry 10 (hereinafter simply referred to as the "gantry 10") horizontally toward the top board 33 of the bed 30 (in the Z-axis direction), and moves the top board 33 of the bed 30 up and down in the Y-axis direction (which may include lateral movement in the X-axis direction and rotational movement around the Z-axis). When tilting the gantry 10, the control device 18 tilts the rotating frame 17 about an axis parallel to the Z-axis direction based on a tilt angle (tilt angle) input to an input interface (not shown) or an input interface 43. The control device 18 grasps the tilt angle of the rotating frame 17 based on the output of a sensor (not shown), etc. The control device 18 provides the tilt angle of the rotating frame 17 to the processing circuit 50 as needed.

[0023] The bed device 30 is a device that places and moves the patient P to be scanned and introduces the patient P into the rotating frame 17 of the gantry device 10. The bed device 30 includes, for example, a base 31, a bed drive device 32, a tabletop 33, and a support frame 34. The base 31 includes a housing that supports the support frame 34 so that the support frame 34 can move in the vertical direction (up and down direction). The bed drive device 32 includes a motor and an actuator. The bed drive device 32 moves the tabletop 33 on which the patient P is placed in the vertical direction (Y-axis direction). The bed drive device 32 may move the tabletop 33 on which the patient P is placed laterally in the horizontal direction (X-axis direction) or rotate around the Z-axis. The bed drive device 32 may move the tabletop 33 on which the patient P is placed along the support frame 34 in the longitudinal direction of the tabletop 33 (Z-axis direction). If the X-ray CT apparatus 1 is a movable gantry-type X-ray CT apparatus, the amount of longitudinal movement of the tabletop 33 by the bed driving device 32 may be an amount sufficient to introduce into the rotating frame 17 a portion of the patient P that is not already inside the rotating frame 17 even when the control device 18 moves the gantry 10 to the maximum horizontal position, i.e., an amount sufficient to compensate for the horizontal movement of the gantry 10. If the gantry 10 is movable in the Z-axis direction, the bed driving device 32 may move the gantry 10 so that the rotating frame 17 is positioned around the patient P. The bed driving device 32 may be configured to move both the gantry 10 and the tabletop 33. The tabletop 33 is a plate-like member on which the patient P rests. The X-ray CT apparatus 1 may be an apparatus in which the patient P is scanned in a standing or sitting position. In this case, the X-ray CT apparatus 1 has a subject support mechanism instead of the bed device 30, and the gantry device 10 rotates the rotating frame 17 about an axial direction perpendicular to the floor surface.

[0024] The console device 40 includes, for example, a memory 41, a display 42, an input interface 43, a network connection circuit 44, and a processing circuit 50. In this embodiment, the console device 40 is described as being separate from the gantry device 10, but the gantry device 10 may include some or all of the components of the console device 40.

[0025] The memory 41 is realized by, for example, a semiconductor memory element such as a read-only memory (ROM), a random access memory (RAM), or a flash memory, a hard disk drive (HDD), or an optical disk. The memory 41 stores, for example, detection data output by the DAS 16, projection data generated based on the detection data, reconstructed images, CT images, and other data. These data may be stored in an external memory with which the X-ray CT apparatus 1 can communicate, instead of (or in addition to) the memory 41. The external memory is controlled by, for example, a cloud server that manages the external memory, by the cloud server accepting a read / write request. The external memory may be realized by, for example, a system called a PACS (Picture Archiving and Communication Systems). A PACS is a medical image management system that systematically stores images captured by various imaging diagnostic apparatuses.

[0026] The display 42 displays various types of information. For example, the display 42 displays images such as reconstructed images and CT images generated by the processing circuitry 50, and GUI (Graphical User Interface) images that accept various operations by the user. The display 42 is, for example, a liquid crystal display (LCD), a CRT (Cathode Ray Tube) display, or an organic EL (Electroluminescence) display. The display 42 may be provided on the gantry device 10. The display 42 may be a desktop type, or may be a display device (for example, a tablet terminal) that can wirelessly communicate with the main body of the console device 40.

[0027] The input interface 43 accepts various input operations by the user and outputs electrical signals indicating the contents of the accepted input operations to the processing circuit 50. For example, the input interface 43 accepts input operations such as collection conditions for collecting detection data, generation conditions for generating projection data, reconstruction conditions for reconstructing a reconstructed image, and image processing conditions for generating a post-processed image from the reconstructed image. The input interface 43 may be implemented, for example, by a mouse, keyboard, touch panel, trackball, switch, button, joystick, camera, infrared sensor, microphone, etc. The input interface 43 may be implemented by a display device (e.g., a tablet terminal) capable of wireless communication with the main body of the console device 40. In this specification, the input interface 43 is not limited to one having physical operation components such as a mouse or keyboard. For example, an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs the electrical signal to a control circuit is also included as an example of an input interface.

[0028] The network connection circuit 44 includes, for example, a network card having a printed circuit board or a wireless communication module. The network connection circuit 44 implements an information communication protocol according to the type of network to be connected. Examples of networks include a local area network (LAN), a wide area network (WAN), the Internet, a cellular network, and a dedicated line. The network connection circuit 44 realizes a connection between the console device 40 and an external memory realized by, for example, the above-mentioned PACS.

[0029] The processing circuitry 50 controls the overall operation of the X-ray CT apparatus 1. The processing circuitry 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 determination function 56, a display control function 57, etc. The processing circuitry 50 realizes these functions by, for example, a hardware processor executing a program (software) stored in the memory 41.

[0030] The hardware processor refers to a circuit such as a CPU, a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD) or a complex programmable logic device (CPLD), or a field programmable gate array (FPGA)). Instead of storing a program in memory 41, the program may be directly embedded in the circuit of the hardware processor. In this case, the hardware processor realizes its functions by reading and executing the program embedded in the circuit. The hardware processor is not limited to being configured as a single circuit, but may be configured as a single hardware processor by combining multiple independent circuits to realize each function. Multiple components may be integrated into a single hardware processor to realize each function. Multiple components may be incorporated into a single dedicated LSI to realize each function. Here, the program (software) may be stored in advance in a storage device (storage device having a non-transitory storage medium) constituting a storage device such as a ROM, RAM, a semiconductor memory element such as a flash memory, or a hard disk drive, or may be stored in a removable storage medium (non-transitory storage medium) such as a DVD or CD-ROM, and installed in the storage device of the console device 40 by inserting the storage medium into a drive device of the console device 40. The program (software) may be downloaded in advance from another computer device via a network connected by the network connection circuit 44, and then installed in the storage device of the console device 40. The program (software) installed in the storage device of the console device 40 may be transferred to a processing circuit of the control device 18 and executed.

[0031] Each component of the console device 40 or the processing circuitry 50 may be distributed and realized by multiple pieces of hardware. The processing circuitry 50 may not be a component of the console device 40, but may be realized 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 (e.g., a cloud server) connected to multiple X-ray CT devices and collectively executing processing equivalent to that of the processing circuitry 50 described below. In other words, the configuration of this embodiment can also be realized as an X-ray CT diagnostic system (medical diagnostic system) in which an X-ray CT device and other processing devices are connected via a network.

[0032] The system control function 51 controls various functions of the processing circuit 50 based on, for example, an input operation received by the input interface 43. For example, the system control function 51 controls the X-ray high voltage device 14, the DAS 16, the control device 18, and the bed driving device 32 to perform processing such as collection of detection data in the gantry device 10.

[0033] The pre-processing function 52 performs pre-processing such as logarithmic conversion processing, offset correction processing, inter-channel sensitivity correction processing, and beam hardening correction on the detection data output by the DAS 16 to generate projection data, and stores the generated projection data in the memory 41.

[0034] The reconstruction processing function 53 performs a predetermined reconstruction process using a filtered back projection method, an iterative reconstruction method, or the like on the projection data generated by the preprocessing function 52 to generate a reconstructed image, and stores the generated reconstructed image in the memory 41.

[0035] The image processing function 54 performs post-processing of the reconstructed image using a known method based on the input operation received by the input interface 43, and converts it into a CT image such as a three-dimensional image or a cross-sectional image of an arbitrary cross section. The conversion into a three-dimensional image or a cross-sectional image by post-processing may be performed by the pre-processing function 52. The image processing function 54 stores the converted CT image in the memory 41.

[0036] The scan control function 55 controls the collection process of detection data in the gantry device 10 by issuing instructions to the X-ray high voltage device 14, the DAS 16, the control device 18, and the bed driving device 32. The scan control function 55 controls the operation of each part when capturing an alignment image, an actual captured image, and an image used for an examination or diagnosis.

[0037] The determination function 56 determines whether or not the patient P has developed a disease other than the disease currently being diagnosed, based on the projection data (which may include detection data and reconstructed images). The determination of the development of the other disease in the determination function 56 is performed, for example, by using a deep learning function, which is one of the machine learning functions using AI (artificial intelligence), to estimate whether or not a disease other than the disease being diagnosed is included in the image surrounding the part of the subject imaged in the projection data. However, since making an estimation for all diseases other than the disease being diagnosed would impose a heavy processing load and extend the processing time, the estimation is performed for diseases specified based on predetermined conditions, thereby reducing the processing load and shortening the processing time required to obtain a determination result of the development of the other disease. Details of this function of the determination function 56 (such as the function of identifying a disease) will be described later.

[0038] The display control function 57 controls the display mode of the display 42. For example, the display control function 57 controls the display 42 to display a reconstructed image generated by the processing circuitry 50, the results of determining whether or not different diseases have occurred, a GUI image that accepts various operations by the user, and the like. The display control function 57 is an example of a "display control unit." The display 42 is an example of a "display device."

[0039] With this configuration, the X-ray CT device 1 irradiates the patient P with X-rays generated by the X-ray tube device 11 to capture (scan) the patient P. The X-ray CT device 1 can scan the patient P in various ways, including helical scanning, conventional scanning, and step-and-shoot. Helical scanning is a method of scanning the patient P in a spiral manner by rotating the rotating frame 17 while moving the top plate 33. Conventional scanning is a method of scanning the patient P in a circular orbit by rotating the rotating frame 17 while keeping the top plate 33 stationary. Step-and-shoot is a method of performing conventional scanning in multiple scan areas by moving the position of the top plate 33 at regular intervals.

[0040] A CT diagnosis of a patient P using the X-ray CT apparatus 1 is performed mainly according to the following procedure.

[0041] (Procedure 1): The diagnostician requests a CT scan to the radiology department. At this time, the diagnostician notifies the radiology department of the request for the CT scan. The request includes, for example, information about the patient P requesting the CT scan (hereinafter referred to as "patient information") and information about the scan of the patient P (hereinafter referred to as "scanning information"). Here, the patient information is, for example, information about the age, gender, medical history, etc. of the patient P. The scan information is, for example, information about the part of the patient P to be scanned and the disease to be diagnosed. The scan information may include conditions such as whether or not a contrast agent is used during scanning.

[0042] (Procedure 2): The user sets an imaging protocol in the X-ray CT device 1 according to the request notified by the diagnostician, and performs imaging of the patient P. The imaging protocol includes, for example, imaging conditions for CT imaging and reconstruction conditions for generating a reconstructed image. Here, imaging conditions include, for example, the tube voltage and tube current of the X-ray tube device 11, HP (helical pitch), FOV (field of view), imaging range, and whether or not a contrast agent is used. Reconstruction conditions include, for example, the noise reduction level, slice thickness, matrix size, reconstruction range, and other conditions for generating a reconstructed image.

[0043] (Procedure 3): After the user confirms that there are no problems with the reconstructed image after the scan, they perform post-processing to generate a CT image with corrected image quality. The user then forwards the generated CT image to the diagnostician who requested the CT scan.

[0044] (Procedure 4): The diagnostician makes a diagnosis of patient P based on the patient information and the CT images transferred by the user.

[0045] In the X-ray CT device 1, while a CT diagnosis is being performed on the patient P according to the above procedure, the determination function 56 included in the processing circuitry 50 determines whether or not the patient P has developed a different disease, and presents the determination result to the user or the diagnostician. More specifically, in the X-ray CT device 1, the determination result regarding the different disease in the patient P is presented to the user, for example, while the user is checking the reconstructed image in procedure 3 or while a post-processed CT image is being generated, until the user transfers the CT image to the diagnostician. In the X-ray CT device 1, the determination result regarding the different disease in the patient P is presented to the diagnostician, for example, until the diagnostician completes the diagnosis of the patient P in procedure 4.

[0046] [Functional configuration of the judgment function] Next, a description will be given of the configuration for realizing the function of determining the presence or absence of different diseases in the determination function 56. Fig. 2 is a diagram showing an example of the functional configuration of the determination function 56 provided in the medical image diagnostic apparatus (X-ray CT apparatus 1) according to the embodiment. The determination function 56 includes, for example, an acquisition function 562, a detection probability calculation function 564, a disease selection function 566, and a disease determination function 568.

[0047] The acquisition function 562 acquires information used by the determination function 56 to determine the presence or absence of different diseases (whether or not different diseases have developed in the patient P). The acquisition function 562 acquires, for example, patient information, imaging information, imaging conditions, and reconstruction conditions as information used to determine the presence or absence of different diseases. The patient information and imaging information may be, for example, information input by the user through the input interface 43 when setting an imaging protocol in the X-ray CT apparatus 1 in accordance with the request notified by the diagnostician, or may be information included in the request from the diagnostician received by the network connection circuit 44. In the following description, the patient information and imaging information are referred to as "patient information" without distinction. The imaging conditions and reconstruction conditions may be, for example, information input by the user through the input interface 43, or may be information included in the imaging protocol set by the user in the X-ray CT apparatus 1. In the following description, the imaging conditions and reconstruction conditions are referred to as "imaging conditions" without distinction.

[0048] Furthermore, the acquisition function 562 acquires the projection data of the patient P currently imaged. The acquisition function 562 may acquire, for example, the projection data stored in the memory 41, or may directly acquire the projection data generated by the pre-processing function 52. The acquisition function 562 may acquire the detection data or the reconstructed image. In this case, the acquisition function 562 may acquire all of the detection data, the projection data, and the reconstructed image, or may acquire any one of them.

[0049] The acquisition function 562 outputs the acquired patient information and imaging conditions to the detection probability calculation function 564. The acquisition function 562 outputs the acquired projection data to the disease determination function 568. The acquisition function 562 is an example of an "acquisition unit." The patient information and imaging information are an example of "first subject information." The imaging conditions and reconstruction conditions are an example of "first imaging conditions." The projection data (which may include detection data and reconstructed images) is an example of "first imaging data."

[0050] The detection probability calculation function 564 calculates the probability (hereinafter referred to as the "detection probability") that a disease that has developed in patient P and is different from the disease currently being diagnosed will be detected from the projection data, based on the patient information and imaging conditions output by the acquisition function 562. The detection probability calculation function 564 calculates the detection probability by inputting patient information and imaging conditions into a trained model for each disease (hereinafter referred to as the "probability calculation model") trained using, for example, an AI-based machine learning function. The probability calculation model is a trained model that has been trained in advance using, for example, a convolutional neural network (CNN) or a deep neural network (DNN) to output the detection probability of a corresponding disease being detected from the projection data captured when patient information and imaging conditions are input. A CNN is a neural network in which several layers, such as a convolution layer and a pooling layer, are connected. A DNN is a neural network in which layers of any type are connected in a multi-layered manner. The probability calculation model is a trained model generated in advance by, for example, performing machine learning using a machine learning model for each disease by a computing device (not shown). The machine learning model is a model in the form of, for example, CNN or DNN, with parameters provisionally set. When performing machine learning of the probability calculation model, the estimation result (estimation result) of a trained model that estimates a corresponding disease (hereinafter referred to as a "diagnostic estimation model") is trained together with patient information and imaging conditions. The diagnostic estimation model is also a trained model that has been trained using, for example, technology such as CNN or DNN, to output, as an estimation result, an index related to a disease corresponding to image data when the image data is input. The disease index is information indicating whether a corresponding disease is captured in the image data (information indicating the presence or absence of the disease). The diagnostic estimation model is, for example, an existing trained model that has been trained and generated in advance in the field of disease estimation to estimate the presence or absence of each disease.Like the probability calculation model, the diagnostic estimation model is also learned using an AI machine learning function in, for example, a computing device (not shown).

[0051] Here, a method for generating a probability calculation model used by the detection probability calculation function 564 to calculate the detection probability will be described. In the following description, the probability calculation model will be described as being generated in advance by a calculation device (not shown). Fig. 3 is a diagram schematically showing an example of a method for generating a probability calculation model used in the detection probability calculation function 564 provided in the medical image diagnostic apparatus (X-ray CT apparatus 1) according to the embodiment. Fig. 3 schematically shows an example of a method for generating a probability calculation model that outputs the detection probability of one disease-A.

[0052] When a calculation device (not shown) makes a probability calculation model learn the probability of detection of disease-A, it inputs, as input data, patient information of a plurality of subjects (including subjects different from patient P) collected in previous diagnoses and imaging conditions corresponding to each patient information to the input side of the probability calculation model. Fig. 3 shows how a plurality of patient information and a plurality of imaging conditions corresponding to each patient information are input to the probability calculation model TMP-A of disease-A.

[0053] Furthermore, a calculation device (not shown) generates image data by applying image construction conditions suitable for diagnosing disease-A to each projection data corresponding to the patient information and imaging conditions input to the probability calculation model, i.e., each projection data captured under a combination of patient information and imaging conditions. For example, the calculation device (not shown) generates, as image data, a reconstructed image by applying the reconstruction conditions used to generate a reconstructed image suitable for diagnosing disease-A from the projection data. The calculation device (not shown) then inputs the generated image data to a diagnostic estimation model, and inputs the estimation result (presence or absence of disease) output from the diagnostic estimation model to the probability calculation model as training data. Figure 3 shows how image data (reconstructed image) obtained by applying image construction conditions suitable for diagnosing disease-A to the projection data corresponding to the patient information and imaging conditions input to the probability calculation model TMP-A is input to the diagnostic estimation model TMR-A for disease-A, and how the estimation result output by the diagnostic estimation model TMR-A is input to the probability calculation model TMP-A.

[0054] Then, a computing device (not shown) trains the probability calculation model to determine the possibility that the captured projection data contains disease-A based on the patient information, imaging conditions, and estimation results input to the probability calculation model. At this time, the computing device (not shown) trains the model so that the detection probability output by the probability calculation model is consistent with the estimation result (teaching data) output by the diagnostic estimation model (so that the detection probability approaches the estimation result output by the diagnostic estimation model). For example, the computing device (not shown) trains the model so that the difference between the detection probability and the estimation result becomes zero. More specifically, the computing device (not shown) trains the model so that the detection probability is the ratio of the number of estimation results indicating the presence of a disease to the number of image data sets input to the diagnostic estimation model. In other words, the computing device (not shown) trains the model so that when patient information and imaging conditions are input to the probability calculation model, the estimation result output by the diagnostic estimation model is obtained. At this time, the computing device (not shown) adjusts temporary parameters set in the machine learning model of the probability calculation model. Methods for adjusting the parameters include, for example, backpropagation (backpropagation). The machine learning model, whose parameters have been adjusted by a computing device (not shown), becomes a probability calculation model. As a result, the probability calculation model becomes a trained model that outputs the probability (detection probability) of detecting Disease-A captured in the captured projection data or image data (which may include detection data or reconstructed images) from patient information and imaging conditions, without inputting the projection data or image data. Figure 3 shows how the probability calculation model TMP-A is trained so that the detection probability output by the probability calculation model TMP-A is consistent with the estimation result output by the diagnostic estimation model TMR-A.

[0055] In this way, the calculation device (not shown) generates in advance a probability calculation model for each disease, that is, a plurality of probability calculation models corresponding to one disease (for example, the number of types of diseases for which the detection probability is to be calculated).

[0056] The probability computation model (probability computation model TMP-A in Figure 3) is an example of a "first model." The disease (disease-A in Figure 3) for which the probability computation model calculates the detection probability is an example of a "first disease." The diagnostic estimation model (diagnostic estimation model TMR-A in Figure 3) that outputs the estimation results obtained by machine learning of the probability computation model is an example of a "third model." The estimation results that the diagnostic estimation model (diagnostic estimation model TMR-A in Figure 3) that outputs the estimation results obtained by machine learning of the probability computation model estimates the presence or absence of a disease are an example of a "third indicator."

[0057] The probability computation model is not limited to one generated by learning using techniques such as CNN, DNN, etc. For example, the probability computation model may be one generated using any machine learning technique such as a gradient method such as SGD (Stochastic Gradient Descent), Momentum SGD, AdaGrad, RMSprop, AdaDelta, or Adam (Adaptive moment estimation), a logistic regression analysis, or a technique based on a support vector machine.

[0058] The probability calculation model is not limited to a trained model that outputs the detection probability of a single corresponding disease. The probability calculation model may be a trained model that outputs the detection probability of multiple diseases captured in the captured projection data or image data (which may include detection data and reconstructed images) based on input patient information and imaging conditions. In other words, a trained model that outputs the detection probability of multiple types of diseases for each disease by inputting patient information and imaging conditions into a single probability calculation model may be used. A method for generating a single probability calculation model that outputs the detection probability of multiple types of diseases may be equivalent to the method for generating the probability calculation model TMP described using FIG. 3, and can be easily conceived based on the description using FIG. 3. Therefore, a detailed description of the method for generating a single probability calculation model that outputs the detection probability of multiple diseases will be omitted. A single probability calculation model that outputs the detection probability of multiple types of diseases for each disease is an example of a "fourth model."

[0059] Returning to Figure 2, the detection probability calculation function 564 calculates the detection probability of each disease by inputting the patient information and imaging conditions output by the acquisition function 562 into each probability calculation model that outputs the detection probability of the corresponding disease. Then, the detection probability calculation function 564 outputs information representing the calculated detection probability of each disease to the disease selection function 566. The detection probability calculation function 564 is an example of a "first derivation unit."

[0060] For example, after the acquisition function 562 acquires and outputs the patient information and the imaging conditions, the detection probability calculation function 564 may calculate the detection probability of each disease based on the patient information and the imaging conditions before starting imaging of the patient P. In this case, the detection probability calculation function 564 may make suggestions to the user based on the calculated detection probability, such as expanding the imaging range for imaging the patient P so that diseases with a non-zero detection probability are captured in the projection data, or optimizing the imaging conditions to an extent that does not significantly affect the X-ray exposure dose or the request content notified by the diagnostician. In this case, the detection probability calculation function 564 outputs information such as the imaging range to be expanded and the imaging conditions to be optimized to the display control function 57, and causes the display control function 57 to display the output information and a message indicating the suggested content on the display 42 to present to the user. This allows the user to set an imaging protocol in the X-ray CT device 1 that takes into account the imaging range and imaging conditions suggested by the detection probability calculation function 564 and perform imaging of the patient P. Information such as the expanded shooting range or the optimized shooting conditions, or a message indicating the content of a proposal to expand the shooting range or to optimize the shooting conditions, is an example of "information indicating the first shooting conditions."

[0061] The disease selection function 566 selects a specific target disease for which the presence or absence of the onset of a disease different from the disease currently being diagnosed is to be determined (estimated) based on the detection probability of each disease output by the detection probability calculation function 564. The disease selection function 566 selects a specific target disease (hereinafter referred to as a "target disease") to be determined, for example, according to preset selection conditions. The selection conditions are, for example, conditions for selecting a predetermined number of diseases as target diseases in descending order of the detection probability of each disease output by the detection probability calculation function 564. Alternatively or in addition to this, the selection conditions may be conditions for selecting target diseases based on, for example, the processing time required to determine the presence or absence of a disease for which the detection probability calculation function 564 has output a non-zero detection probability, so that as many diseases as possible can be determined within the estimated time (required time) from the start of imaging of the patient P to the transfer of the CT image to the diagnostician. The processing time required to determine the presence or absence of a disease is the time required for the disease determination function 568 to determine the presence or absence of the disease, calculated in advance for each disease. The estimated time (required time) from the start of imaging of patient P to the transfer of CT images to the diagnostician is, for example, the time that patient P can remain on the tabletop 33, taking into account the possibility of re-imaging. Alternatively or additionally, the selection condition may be a condition for selecting a target disease, for example, a condition for prioritizing a disease with a high diagnostic priority among diseases for which a non-zero detection probability is output by the detection probability calculation function 564. The diagnostic priority is set in advance for each disease, for example, so that the higher the disease progresses, the greater the risk upon onset, i.e., the earlier the disease is detected, the more effective the treatment for patient P. The selection condition may be set, for example, when the X-ray CT apparatus 1 is installed in the radiology department's imaging room, or may be set by the user after being notified by the diagnostician as part of a CT imaging request, or may be set by the user when setting up the imaging protocol for the X-ray CT apparatus 1. The disease selection function 566 outputs information indicating the selected target disease to the disease determination function 568. The disease selection function 566 is an example of a "selection unit."

[0062] The disease determination function 568 determines whether the patient P has the target disease output by the disease selection function 566 (a disease different from the disease currently being diagnosed) based on the projection data output by the acquisition function 562. The disease determination function 568 determines whether the target disease has occurred (presence or absence of the target disease) by inputting image data into a diagnostic estimation model corresponding to the target disease that has been trained using an AI machine learning function, for example. The diagnostic estimation model is the same as the diagnostic estimation model (see FIG. 3) used to generate the probability calculation model. That is, it is an existing trained model that has been trained to perform estimation on image data generated for each disease and output an index related to the corresponding disease (information indicating whether the disease corresponding to the image data is captured in the image data) as an estimation result. The diagnostic estimation model may be any trained model that, when inputted with projection data, determines whether a corresponding disease is captured in the projection data. The image data that the disease determination function 568 inputs to the diagnostic estimation model may be projection data output by the acquisition function 562, or may be image data converted into image data suitable for estimation of the projection data by the diagnostic estimation model. Image data suitable for estimating projection data using a diagnostic estimation model may be, for example, a reconstructed image generated by applying image construction conditions (reconstruction conditions) suitable for determining the presence or absence of a target disease to projection data. When the diagnostic estimation model outputs an estimation result indicating that the corresponding target disease has developed (the target disease is present), the disease determination function 568 outputs this estimation result and information representing the target disease (e.g., the name of the target disease) as a determination result to the display control function 57. When the disease determination function 568 generates image data by applying image construction conditions suitable for determining the target disease to the projection data, it may output this image data together with the determination result to the display control function 57.

[0063] The diagnostic estimation model is an example of a "second model." The estimation result output by the diagnostic estimation model after making an estimation (determining whether or not the target disease is present) is an example of a "second index." The disease determination function 568 is an example of a "second derivation unit." The determination result output by the disease determination function 568 is an example of an "index related to a specific disease."

[0064] As a result, the display control function 57 displays on the display 42 information corresponding to the determination result output by the disease determination function 568 together with the reconstructed image generated by the reconstruction processing function 53, and presents this to the user. For example, the display control function 57 displays on the display 42 a message recommending that the patient P also undergo a diagnosis of a disease (target disease) different from the disease to be diagnosed, and presents this to the user. When the disease determination function 568 outputs image data to which the suitable image composition conditions used for the determination of the target disease have been applied (image data input to the diagnostic estimation model) along with the determination result, the display control function 57 may display on the display 42 a message recommending that the image data be forwarded to a diagnostician together with the CT image, and present this to the user. The message recommending that the patient P also undergo a diagnosis of the target disease (which may include image data to which the suitable image composition conditions used for the determination of the target disease from the disease determination function 568 have been applied) is an example of "information representing an indicator related to a specific disease."

[0065] The disease determination function 568 determines whether or not the patient P has developed any of the remaining diseases during times when imaging by the X-ray CT device 1 is not being performed, such as at night. In other words, the disease determination function 568 retroactively determines whether or not the patient P has developed multiple types of diseases other than the target disease output by the disease selection function 566. As a result, the disease determination function 568 makes determinations for all diseases other than the disease to be diagnosed (for example, many diseases that are candidates for diagnosis, such as hundreds of types). This determination may be made regardless of the detection probability calculated by the detection probability calculation function 564 (that is, including diseases with a detection probability of zero). The disease determination results retroactively determined by the disease determination function 568 are intended to be referenced by a diagnostician, for example, at the time of the next diagnosis of the patient P. For this reason, the disease determination function 568 may store the follow-up disease determination results in, for example, the memory 41, or in a storage device provided in a terminal device (for example, a computer device such as a personal computer (PC) installed in an examination room, or a tablet terminal) through which a diagnostician checks CT images for CT diagnosis. When the disease determination function 568 generates image data by applying image configuration conditions suitable for disease estimation to the projection data when making a diagnosis for each disease, the disease determination function 568 may associate this image data with the determination results and store it in the memory 41 or a storage device provided in the terminal device. Furthermore, the disease determination function 568 may associate the follow-up disease determination results and the name of the disease with patient information, imaging conditions, and projection data (which may include image data to which suitable image configuration conditions have been applied), and output them to, for example, a computing device (not shown) for use in additional learning in the probability calculation model (updating the probability calculation model).

[0066] [Operation of the judgment function] Next, a description will be given of an example of an operation of determining whether or not different diseases have developed in the patient P in the determination function 56. Fig. 4 is a flowchart showing an example of a processing flow for determining the presence or absence of different diseases in the determination function 56 included in the medical image diagnostic apparatus (X-ray CT apparatus 1) according to the embodiment.

[0067] When the X-ray CT device 1 completes imaging of the patient P, the processing circuit 50 generates projection data using a pre-processing function 52 and generates a reconstructed image using a reconstruction processing function 53. When the pre-processing function 52 completes generating the projection data, the determination function 56 starts processing to determine the presence or absence of different diseases.

[0068] When the process of determining the presence or absence of a different disease is started, the acquisition function 562 acquires patient information (patient information and imaging information) of the patient P currently imaged (step S100). Furthermore, the acquisition function 562 acquires imaging conditions (imaging conditions and reconstruction conditions) of the patient P currently imaged (step S102). The acquisition function 562 outputs the acquired patient information and imaging conditions to the detection probability calculation function 564. Thereafter, the acquisition function 562 acquires the projection data of the patient P generated currently (step S104). The acquisition function 562 outputs the acquired projection data to the disease determination function 568.

[0069] The detection probability calculation function 564 calculates the detection probability of detecting different diseases that may have developed in the patient P from the projection data based on the patient information output by the acquisition function 562 and the imaging conditions (step S200). Here, an example of a method for calculating the detection probability of different diseases by the detection probability calculation function 564 will be described.

[0070] FIG. 5 is a diagram schematically illustrating an example of a method for calculating the detection probability of different diseases in the detection probability calculation function 564 included in the medical image diagnostic apparatus (X-ray CT apparatus 1) according to the embodiment. FIG. 5 schematically illustrates an example of calculating the detection probability of detecting multiple different diseases (here, Disease-A, Disease-B, . . . , Disease-X) from projection data. In this case, the detection probability calculation function 564 inputs the same patient information and imaging conditions output by the acquisition function 562 to each of a probability calculation model TMP-A corresponding to Disease-A, a probability calculation model TMP-B corresponding to Disease-B, and a probability calculation model TMP-X corresponding to Disease-X. As a result, each probability calculation model TMP outputs the detection probability of the corresponding disease as a result. The detection probability calculation function 564 then outputs information indicating the disease corresponding to each probability calculation model TMP (e.g., the name of the disease) and the detection probability result output by the probability calculation model TMP to the disease selection function 566. Figure 5 shows a schematic example of a case in which the probability calculation model TMP-A outputs a result that the probability of detecting disease-A is 20%, the probability calculation model TMP-B outputs a result that the probability of detecting disease-B is 70%, and the probability calculation model TMP-X outputs a result that the probability of detecting disease-X is 1%, and this information is output to the disease selection function 566.

[0071] In FIG. 5, the probability computation model TMP-A is an example of one "first model" among the "plurality of first models," and Disease-A is an example of the "corresponding disease" to which this first model corresponds. In this case, the probability computation model TMP-B is an example of another "first model" among the "plurality of first models," and Disease-B is an example of the "corresponding disease" to which this first model corresponds. Similarly, the probability computation model TMP-X is an example of yet another "first model" among the "plurality of first models," and Disease-X is an example of the "corresponding disease" to which this first model corresponds.

[0072] Incidentally, it is conceivable that a probability calculation model corresponding to one of the diseases may output a result indicating that the detection probability of the corresponding disease is 0%. In this case, the detection probability calculation function 564 may also output information about the disease for which a detection probability of 0% has been output to the disease selection function 566, or may omit outputting information about the disease for which the detection probability is 0% to the disease selection function 566. Furthermore, it is conceivable that the probability calculation models may include those corresponding to diseases that are clearly not related to the disease to be diagnosed at this time, or that clearly develop in areas that do not exist within or around the imaging range for diagnosing the disease to be diagnosed at this time. In other words, it is conceivable that the probability calculation models may include those that clearly output a result indicating that the detection probability is 0%. In this case, the detection probability calculation function 564 may omit inputting patient information and imaging conditions for probability calculation models that clearly output a result indicating that the detection probability is 0%.

[0073] Returning to Fig. 4, the disease selection function 566 selects a target disease based on information representing the disease and information relating the disease's detection probability, output by the detection probability calculation function 564 (step S300). For example, when the detection probability calculation function 564 outputs information indicating that the detection probability of disease-A is 20%, the detection probability of disease-B is 70%, and the detection probability of disease-X is 1%, as in the example shown in Fig. 5, if the selection condition is a detection probability of 50% or more, the disease selection function 566 selects disease-B as the target disease. In the example shown in Fig. 5, if the selection condition is to select the two diseases with the highest detection probabilities as target diseases, the disease selection function 566 selects disease-A and disease-B as the target diseases. In the example shown in Figure 5, if the processing time required for the disease determination function 568 to determine the presence or absence of a target disease is 40 seconds for disease-A, 40 seconds for disease-B, and 20 seconds for disease-X, and the selection conditions are such that the time required from the start of imaging of patient P to the transfer of the CT image to the diagnostician is 60 seconds, the disease selection function 566 selects disease-B and disease-X as the target diseases. In the example shown in Figure 5, if disease-A has the highest priority, disease-B has the next highest priority, and disease-X has the lowest priority, and the selection conditions are such that the target disease with the highest priority is selected, the disease selection function 566 selects disease-A as the target disease. The disease selection function 566 outputs information indicating the selected target disease to the disease determination function 568.

[0074] The disease determination function 568 determines the presence or absence of each target disease output by the disease selection function 566 for the projection data output by the acquisition function 562 (step S400). That is, the disease determination function 568 determines for each target disease whether or not an image of a symptom indicating the onset of the target disease is included in the projection data. Here, an example of a method for determining the presence or absence of a target disease by the disease determination function 568 will be described.

[0075] FIG. 6 is a diagram schematically illustrating an example of a method for determining different diseases (target diseases) in the disease determination function 568 included in the medical image diagnostic apparatus (X-ray CT apparatus 1) according to the embodiment. FIG. 6 schematically illustrates an example of determining whether or not the projection data contains an image showing symptoms of disease-B when the disease selection function 566 selects disease-B as the target disease. In this case, the disease selection function 566 generates image data by applying image construction conditions suitable for estimating disease-B to the projection data. FIG. 6 also illustrates an example of generating a reconstructed image by acquiring reconstruction conditions suitable for estimating disease-B from, for example, a storage device (which may be the memory 41) that stores reconstruction conditions for each disease, and applying the acquired reconstruction conditions to the projection data as suitable image construction conditions. The disease determination function 568 then inputs the image data generated by applying the suitable image construction conditions to the diagnostic estimation model TMR-B corresponding to disease-B. As a result, the diagnostic estimation model TMR-B outputs an estimation result (presence or absence of disease-B) of the corresponding disease-B. When the diagnostic estimation model TMR-B outputs an estimation result indicating that disease-B has developed (the presence of disease-B), the disease determination function 568 outputs this estimation result, information indicating disease-B (e.g., the name of disease-B), and image data (reconstructed image) generated by applying reconstruction conditions suitable for estimating disease-B to the projection data, as a determination result to the display control function 57. When the disease selection function 566 outputs information indicating multiple target diseases, the disease determination function 568 replaces the diagnostic estimation model TMR-B shown in FIG. 6 with a diagnostic estimation model corresponding to another target disease, performs estimation for each target disease in the same way, and outputs the determination result to the display control function 57.

[0076] In FIG. 6, the diagnostic estimation model TMR-B is an example of a "second model," and Disease-B is an example of a "specific disease." The estimation result (presence or absence of Disease-B) obtained by the diagnostic estimation model TMR-B is an example of a "second index." In FIG. 6, the determination result output by the disease determination function 568 is an example of an "index related to a specific disease."

[0077] 4, the display control function 57 displays the information on the determination result output by the disease determination function 568 (for example, a message recommending that a diagnosis for disease B also be made, or a reconstructed image suitable for estimating disease B) together with the reconstructed image generated by the reconstruction processing function 53 on the display 42, and presents it to the user (step S500). As a result, the user transfers the requested CT image and the information on the determination result to the diagnostician.

[0078] With this configuration and operation, in the determination function 56, the detection probability calculation function 564 calculates the detection probability of a disease captured in the projection data or image data based on the patient information and imaging conditions, and the disease selection function 566 selects a target disease other than the disease to be diagnosed. Then, in the determination function 56, the disease determination function 568 performs an estimation to determine the presence or absence of the selected target disease. This allows the determination function 56 to estimate a specific target disease with a reduced processing load and shorter processing time than when estimating all diseases other than the disease to be diagnosed during imaging of the patient P. This allows the X-ray CT device 1 to notify the user and even the diagnostician of target diseases that may have developed along with the disease to be diagnosed. In other words, the X-ray CT device 1 provides the user and the diagnostician with an opportunity to notice other diseases that are included in the projection data but are difficult to detect from the reconstructed images or CT images, thereby assisting (assisting) the diagnostician in diagnosing the patient P.

[0079] Furthermore, as described above, in the determination function 56, the detection probability calculation function 564 calculates the detection probability of each disease that may be detected in the captured projection data or image data before starting to capture the patient P, and can make suggestions to the user to ensure that the diseases whose detection probabilities have been calculated are captured in the projection data (such as suggestions to expand the imaging range or suggestions to optimize the imaging conditions). This allows the X-ray CT device 1 to assist (support) the user in capturing the image of the patient P.

[0080] As described above, the medical image diagnostic apparatus (X-ray CT apparatus) of the embodiment calculates the probability of detecting a disease different from the current diagnosis target disease from the projection data or image data based on patient information and imaging conditions, selects a disease (target disease) for which onset is to be determined based on this detection probability, and performs a diagnosis on the selected target disease. In other words, the medical image diagnostic apparatus (X-ray CT apparatus) of the embodiment calculates the probability of detecting a different disease from the projection data or image data without using the projection data or image data, and performs onset determination limited to target diseases with a high detection probability. This allows the medical image diagnostic apparatus (X-ray CT apparatus) of the embodiment to perform a diagnosis on a specific target disease with reduced processing load and shorter processing time than if all diseases different from the current diagnosis target disease that may be detected based on the projection data or image data were diagnosed. This allows the medical image diagnostic apparatus (X-ray CT apparatus) of the embodiment to notify the user or a diagnostician of other diseases that may have developed along with the current diagnosis target disease. As a result, the medical image diagnostic apparatus (X-ray CT apparatus) according to the embodiment can reduce the risk of overlooking the onset of a disease different from the disease currently being diagnosed.

[0081] In the above-described embodiment, the medical image diagnostic apparatus is described as an X-ray CT apparatus. However, this is merely an example, and the medical image diagnostic apparatus is not limited to an X-ray CT apparatus as long as it captures medical images and performs diagnosis. The medical image diagnostic apparatus may be, for example, an MRI (Magnetic Resonance Imaging) apparatus, a PET (Positron Emission Tomography) apparatus, a PET-CT apparatus, or a SPECT (Single Photon Emission Computed Tomography) apparatus. The configuration, operation, and processing of the medical image diagnostic apparatus in these cases may be equivalent to the configuration, operation, and processing of the medical image diagnostic apparatus (X-ray CT apparatus) in the above-described embodiment. Therefore, detailed descriptions of the configuration, operation, and processing of medical image diagnostic apparatuses other than an X-ray CT apparatus will be omitted.

[0082] The above-described embodiment can be expressed as follows. processing circuitry; The processing circuitry acquiring first object information relating to an object, first imaging conditions relating to imaging of the object, and first imaging data obtained by imaging the object; deriving detection probabilities of a plurality of diseases by inputting the first object information and the first imaging conditions into each of a plurality of first models that output detection probabilities of corresponding diseases in response to input of the object information and the imaging conditions; selecting a specific disease from the plurality of diseases based on the detection probability of the plurality of diseases; deriving an index for the specific disease by inputting the first imaging data into a second model, which is a model for the selected specific disease and outputs a second index for the corresponding disease in response to input of imaging data; Medical imaging diagnostic equipment.

[0083] According to at least one of the embodiments described above, there is provided an acquisition unit (562) that acquires first object information (patient information) regarding an object (P), first imaging conditions (imaging information) regarding imaging of the object, and first imaging data (projection data) obtained by imaging the object; a first derivation unit (564) that derives detection probabilities of a plurality of diseases by inputting the first object information and the first imaging conditions into each of a plurality of first models (e.g., TMP-A and TMP-B) that output detection probabilities of corresponding diseases (e.g., Disease-A and Disease-B) in response to input of the object information and the imaging conditions; and By having a selection unit (566) that selects a specific disease (target disease) from the multiple diseases, and a second derivation unit (568) that is a model for the specific disease selected by the selection unit and that derives an index for the specific disease (e.g., a judgment result indicating the presence or absence of a specific disease) by inputting the first imaging data into a second model (e.g., TMR-A) that outputs a second index for the corresponding disease (e.g., Disease-A) (e.g., an estimation result indicating the presence or absence of Disease-A) by inputting imaging data, it is possible to more efficiently diagnose different diseases based on images captured by a medical image diagnostic device.

[0084] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0085] 1···X-ray CT device, 10···Gantt device, 11···X-ray tube device, 12···Wedge, 13···Collimator, 14···X-ray high voltage device, 15···X-ray detector, 16···Data acquisition system (DAS), 17···Rotating frame, 18···Control device, 30···Couch device, 31···Base, 32···Couch drive device, 33···Top plate, 34···Support frame, 40···Console device, 41···Memory, 42...Display, 43...Input interface, 44...Network connection circuit, 50...Processing circuit, 51...System control function, 52...Preprocessing function, 53...Reconstruction processing function, 54...Image processing function, 55...Scan control function, 56...Determination function, 562...Acquisition function, 564...Detection probability calculation function, 566...Disease selection function, 568...Disease determination function, 57...Display control function

Claims

1. an acquisition unit that acquires first subject information related to a subject, first imaging conditions related to imaging of the subject, and first imaging data obtained by imaging the subject; a first derivation unit that derives detection probabilities of a plurality of diseases by inputting first object information and first imaging conditions into each of a plurality of first models that output detection probabilities of corresponding diseases in response to input of the object information and imaging conditions; a selection unit that selects a specific disease from the plurality of diseases based on the detection probabilities of the plurality of diseases; a second derivation unit that inputs the first imaging data into a second model that is a model for the specific disease selected by the selection unit and outputs a second index related to the corresponding disease in response to input of imaging data, thereby deriving an index related to the specific disease; A medical image diagnostic device comprising:

2. The first model is a trained model that has been trained so that each detection probability output by inputting a plurality of pieces of subject information collected in a previous diagnosis and the imaging conditions matches each third index related to the corresponding disease output by inputting imaging data acquired under a combination of the subject information and the imaging conditions into a third model that outputs a third index related to the corresponding disease by inputting imaging data; The medical image diagnostic apparatus according to claim 1 .

3. the second derivation unit applies, to the first imaging data, image data generated by applying an image construction condition suitable for outputting the second index related to the disease corresponding to the second model, and inputs the image data to the second model as the first imaging data; The medical image diagnostic apparatus according to claim 2 .

4. The second model is the same trained model as the third model. The medical image diagnostic apparatus according to claim 3 .

5. a display control unit that controls the display of information on the display device; the second derivation unit outputs information representing the derived index related to the specific disease to the display control unit; the display control unit causes the display device to display information representing the index related to the specific disease. The medical image diagnostic apparatus according to claim 4.

6. the first derivation unit outputs, to the display control unit, information representing the first imaging condition under which a disease whose detection probability is not zero is imaged in the first imaging data based on the derived detection probability; the display control unit causes the display device to display information representing the first shooting condition. The medical image diagnostic apparatus according to claim 5 .

7. the selection unit selects a predetermined number of diseases as the specific diseases; The medical image diagnostic apparatus according to any one of claims 1 to 6.

8. the selection unit selects a predetermined number of diseases in descending order of the detection probability as the specific diseases. The medical image diagnostic apparatus according to claim 7.

9. the selection unit selects the specific disease based on a processing time required by the second derivation unit to derive an index related to the disease whose detection probability is not zero. The medical image diagnostic apparatus according to claim 7.

10. the selection unit selects the specific disease based on a priority of diagnosis for a disease whose detection probability is not zero. The medical image diagnostic apparatus according to claim 7.

11. an acquisition unit that acquires first subject information related to a subject, first imaging conditions related to imaging of the subject, and first imaging data obtained by imaging the subject; a first derivation unit that derives detection probabilities of a plurality of diseases by inputting the first object information and the first imaging conditions into a fourth model that outputs detection probabilities of each of a plurality of diseases in response to input of the object information and the imaging conditions; a selection unit that selects a specific disease from the plurality of diseases based on the detection probabilities of the plurality of diseases; a second derivation unit that inputs the first imaging data into a second model that is a model for the specific disease selected by the selection unit and outputs a second index related to the corresponding disease in response to input of imaging data, thereby deriving an index related to the specific disease; A medical image diagnostic device comprising:

12. The computer in the medical imaging diagnostic equipment acquiring first subject information relating to a subject, first imaging conditions relating to imaging of the subject, and first imaging data obtained by imaging the subject; deriving detection probabilities of a plurality of diseases by inputting the first object information and the first imaging conditions into each of a plurality of first models that output detection probabilities of corresponding diseases in response to input of the object information and the imaging conditions; selecting a specific disease from the plurality of diseases based on the detection probability of the plurality of diseases; deriving an index related to the specific disease by inputting the first imaging data into a second model, the second model being a model for the selected specific disease and outputting a second index related to the corresponding disease in response to input of imaging data; Medical imaging diagnostic methods.

13. The computer of the medical imaging diagnostic equipment acquiring first subject information about a subject, first imaging conditions for imaging the subject, and first imaging data obtained by imaging the subject; deriving detection probabilities of a plurality of diseases by inputting the first object information and the first imaging conditions into each of a plurality of first models that output detection probabilities of corresponding diseases in response to input of the object information and the imaging conditions; selecting a specific disease from the plurality of diseases based on the detection probability of the plurality of diseases; inputting the first imaging data into a second model, which is a model for the selected specific disease and outputs a second index related to the corresponding disease in response to input of imaging data, thereby deriving an index related to the specific disease; program.

Images