JP7130449B2 - Diagnosis support device and diagnosis support system - Google Patents

Description

TECHNICAL FIELD Embodiments of the present invention relate to a diagnosis support device and a diagnosis support system.

2. Description of the Related Art In the medical field, a doctor interprets a medical image displayed on a display unit and observes the state of a lesion such as cancer and changes over time. 2. Description of the Related Art In image diagnosis using medical images, a doctor visually recognizes an abnormal shadow or the like from the medical image and diagnoses what the abnormal shadow or the like is. In addition, for the purpose of providing information that improves the accuracy of image diagnosis by doctors, it is possible to automatically detect lesions by digitizing medical images and analyzing the images, and to perform computer-aided diagnosis (CAD). Devices have been developed to perform the Diagnosis. CAD can automatically detect abnormal shadow candidates as lesions.

JP 2017-192691 A

A problem to be solved by the present invention is to provide appropriate imaging conditions for acquiring a second image to be generated based on the first image.

A diagnosis support apparatus according to an embodiment includes image acquisition means and determination means. The image acquisition means acquires a first image of the subject captured by the first modality device. The determining means determines imaging conditions including an imaging direction of the subject for generating a second image of the subject by the second modality device from the first image acquired by the image acquiring means.

1 is a schematic diagram showing a configuration example of a diagnosis support system according to an embodiment; FIG. FIG. 2 is a diagram showing an operation example of a determination function in the diagnosis support system according to the embodiment; FIG. 3 is a diagram showing an operation example of a detection function in the diagnosis support system according to the embodiment; FIG. 4 is a diagram showing, as a flowchart, the operation of the diagnostic support system according to the embodiment; FIG. 5 is a diagram showing a first modification of a determination function in the diagnosis support system according to the embodiment; FIG. 6 is a diagram showing a second modification of the determination function in the diagnosis support system according to the embodiment; FIG. 7 is a diagram showing a third modification of the determination function in the diagnosis support system according to the embodiment; FIG. 8 is a diagram showing a fourth modification of the determination function in the diagnosis support system according to the embodiment; FIG. 9 is a diagram showing a fifth modification of the determination function in the diagnosis support system according to the embodiment; FIG. 10 is a diagram showing a sixth modification of the determination function in the diagnosis support system according to the embodiment; FIG. 11 is a diagram showing a first modification of the detection function in the diagnosis support system according to the embodiment; FIG. 12 is a diagram showing a second modification of the detection function in the diagnosis support system according to the embodiment; FIG. 13 is a diagram showing a third modification of the detection function in the diagnosis support system according to the embodiment; 14 is a diagram showing a fourth modification of the detection function in the diagnosis support system according to the embodiment; FIG.

Hereinafter, embodiments of a diagnostic support device and a diagnostic support system will be described in detail with reference to the drawings.

FIG. 1 is a schematic diagram showing a configuration example of a diagnosis support system according to an embodiment.

FIG. 1 shows a diagnosis support system 1 according to an embodiment. A diagnosis support system 1 includes a diagnosis support device 10, an image providing device 11, and a second modality device 12 according to the embodiment. The diagnosis support device 10, the image providing device 11, and the second modality device 12 can communicate with each other via a network N such as a LAN (Local Area Network).

Here, the image providing device 11 is a first modality device such as an X-ray CT (Computed Tomography) device or an MRI (Magnetic Resonance Imaging) device that generates a first medical image, or a first modality device It is a medical image management device such as PACS (Picture Archiving and Communication Systems) that acquires and manages first medical images from the device. A case will be described below in which the image providing apparatus 11 is a first modality apparatus, such as an X-ray CT apparatus, and generates a CT image as the first medical image.

Also, the second modality device 12 is a device that can define an imaging direction in imaging and generates a second medical image. A case where the second modality device 12 is, for example, an X-ray diagnostic device and generates an X-ray image as the second medical image will be described below. The X-ray diagnostic apparatus includes plain X-ray apparatus, angiography apparatus (X-ray fluoroscopy apparatus), mammography apparatus, telescope imaging apparatus, and the like.

The diagnosis support apparatus 10 is a medical image management apparatus, workstation, interpretation terminal, etc., and is provided on a system connected via a network N. FIG. Note that the diagnosis support device 10 may be an off-line device. In that case, the diagnosis support apparatus 10 acquires a first medical image, that is, a CT image, from the X-ray CT apparatus 11 via a portable recording medium, From 12 a second medical image, ie an X-ray image, is acquired.

Here, the diagnostic support apparatus 10, the X-ray CT apparatus 11, and the X-ray diagnostic apparatus 12 may constitute a client-server system. Specifically, the diagnosis support apparatus 10 is included in the server apparatus that provides data, the X-ray CT apparatus 11 is included in the first client apparatus that uses the data, and the X-ray diagnostic apparatus 12 uses the data. is included in the second client device.

The diagnosis support device 10 comprises a processing circuit 21 , a memory circuit 22 , an input interface 23 , a display 24 , a network interface 25 , a decision support means 26 and a detection support means 27 . Note that the input interface 23 , the display 24 , the network interface 25 , the determination support means 26 and the detection support means 27 are not essential functions of the diagnosis support device 10 .

The processing circuit 21 means a processing circuit such as a dedicated or general-purpose CPU (Central Processing Unit) or MPU (Micro Processor Unit), an application specific integrated circuit (ASIC), and a programmable logic device. do. Examples of programmable logic devices include simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs). circuit. The processing circuit 21 implements functions described later by reading and executing a program stored in the storage circuit 22 or directly incorporated in the processing circuit 21 .

Moreover, the processing circuit 21 may be configured by a single circuit, or may be configured by a combination of a plurality of independent processing circuit elements. In the latter case, a plurality of memory circuit elements of memory circuit 22 may store programs corresponding to the functions of a plurality of processing circuit elements, respectively, or one memory circuit element of memory circuit 22 may store a plurality of programs. It may store programs corresponding to the functions of the processing circuitry. Note that the processing circuit 21 is an example of a control unit.

The diagnosis support apparatus 10 implements a first image acquisition function 31, a determination function 32, an imaging condition provision function 33, a second image acquisition function 34, and a detection function 35 by the processing circuit 21 executing the programs. . All or part of the functions 31 to 35 may be implemented in the diagnosis support device 10 as a circuit such as an ASIC. Also, the functions 33 to 35 are not essential functions for the diagnosis support device 10 .

A first image acquisition function 31 is a function for acquiring a CT image of the patient from the X-ray CT apparatus 11 via the network interface 25 . Examples of CT images include volume data, slice data of a specific cross section, or one or a plurality of slice data arbitrarily extracted from volume data.

The determination function 32 determines imaging conditions including the imaging direction of the patient for generating the second image of the patient by the X-ray diagnostic apparatus 12 from the first image acquired by the first image acquisition function 31 . It is the function that determines When the second modality apparatus 12 is an X-ray diagnostic apparatus, the imaging condition means the X-ray imaging condition and includes at least the imaging direction, which is the X-ray irradiation angle with respect to the patient. In addition to the imaging direction, the X-ray imaging conditions include at least one of tube current, tube voltage, imaging time, source to image distance (SID), and the like.

The imaging condition provision function 33 is a function that provides the X-ray imaging conditions determined by the determination function 32 to the X-ray diagnostic apparatus 12 via the network interface 25 . The X-ray diagnostic apparatus 12 executes X-ray imaging of the patient according to the X-ray imaging conditions provided by the imaging condition providing function 33 .

The second image acquisition function 34 acquires an X-ray image of the patient by X-ray imaging by the X-ray diagnostic apparatus 12 according to the X-ray imaging conditions determined by the determination function 32 , via the network interface 25 to the X-ray diagnostic apparatus. It is a function acquired from 12.

The detection function 35 detects the patient based on the first image of the patient acquired by the first image acquisition function 31 and the second image of the patient acquired by the second image acquisition function 34. It is a function to detect the state of the lesion of The detection function 35 can also cause the display 24 to display the state of the detected lesion of the patient.

Note that specific operations of the functions 31 to 35 will be described later with reference to FIGS. 4 to 14. FIG.

The storage circuit 22 is composed of a semiconductor memory device such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, or the like. The storage circuit 22 may be configured by portable media such as USB (Universal Serial Bus) memory and DVD (Digital Video Disk). The storage circuit 22 stores various processing programs (including not only application programs but also an OS (Operating System) and the like) used in the processing circuit 21, data necessary for executing the programs, and the like. In addition, the OS may include a GUI (Graphical User Interface) that uses many graphics to display information on the display 24 for an operator such as a diagnostician, and that allows basic operations to be performed through the input interface 23 . Note that the memory circuit 22 is an example of a memory unit.

The input interface 23 includes an input device that can be operated by an operator, and an input circuit that inputs signals from the input device. Input devices include trackballs, switches, mice, keyboards, touchpads that perform input operations by touching the scanning surface, touch screens that integrate the display screen and touchpad, non-contact input devices using optical sensors, and a voice input device or the like. When the operator operates the input device, the input circuit generates a signal according to the operation and outputs it to the processing circuit 21 . Note that the diagnosis support apparatus 10 may include a touch panel in which the input device is integrated with the display 24 . Note that the input interface 23 is an example of an input unit.

The display 24 is a display device such as a liquid crystal display panel, a plasma display panel, an organic EL (Electro Luminescence) panel, or the like. The display 24 displays medical images generated under the control of the processing circuitry 21 . Note that the display 24 is an example of a display unit.

The network interface 25 is configured by a connector conforming to parallel connection specifications and serial connection specifications. The network interface 25 has a function of performing communication control according to each standard and being able to connect to a network via a telephone line, thereby connecting the diagnostic support apparatus 10 to the network. Note that the network interface 25 is an example of a communication unit.

The decision support means 26 inputs data from the decision function 32 , acquires predetermined data, and outputs the data to the decision function 32 .

For example, the decision support means 26 is a device comprising a GPU (Graphics Processing Unit), a database, and the like. The decision support means 26 constructs a database from a plurality of (a large amount of) CT images obtained by CT imaging with the X-ray CT apparatus 11 . The decision support means 26 uses a plurality of CT images acquired in the past by CT imaging of the X-ray CT apparatus 11 as first learning data, for example, a CNN (convolutional neural network), a convolutional deep belief network (CDBN: Convolutional Deep Belief Network), Reconstructive Topographic Independent Component Analysis (TICA), and other machine learning techniques to construct the first AI (Artificial Intelligence) with mature parameters. Then, the decision support means 26 provides an output to the decision function 32 by inputting the CT image relating to the relevant patient supplied from the decision function 32 to the first AI.

FIG. 2 is a diagram showing an operation example of the determination function 32. As shown in FIG. FIG. 2A is a diagram showing the first training data in constructing the first AI. FIG. 2B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 2C is a diagram showing the flow of data for determining X-ray imaging conditions.

A plurality of, for example, m CT images Fm are input from the X-ray CT apparatus 11 to the decision support means 26 of the diagnosis support apparatus 10 . On the other hand, an appropriate X-ray imaging condition Pm considering the detectability of a lesion such as cancer is input to the decision support means 26 via the input interface 23 so as to correspond to each of the m CT images Fm. be done. As a result, the decision support unit 26 constructs the first AI using the m CT images Fm as the first learning data and the m X-ray imaging conditions Pm as the first teacher data (see FIG. 2 ( A)). In addition, the decision support unit 26 preferably constructs a plurality of AIs for each imaging region as the first AI.

As shown in FIGS. 2B and 2C, the decision function 32 provides the decision support means 26 with the CT image f of the relevant patient from the X-ray CT apparatus 11 . Thereby, the decision support means 26 inputs the CT image f to the first AI, and provides the output of the first AI, that is, the X-ray imaging condition p for the patient to the decision function 32 .

The determination function 32 receives the output of the first AI, that is, the X-ray imaging conditions p from the determination support means 26, and determines appropriate X-ray imaging conditions for the patient based on the X-ray imaging conditions p. , providing it to the X-ray diagnostic apparatus 12 . Here, the determination function 32 determines the X-ray imaging condition p itself from the determination support means 26 as an appropriate X-ray imaging condition for the patient.

Returning to the description of FIG. 1 , the detection support means 27 inputs data from the detection function 35 , acquires predetermined data, and outputs the data to the detection function 35 . For example, the detection support means 27 is a device comprising a GPU, a database, and the like. The detection support means 27 configures a database by combining a plurality of CT images obtained by CT imaging by the X-ray CT apparatus 11 and a plurality of X-ray images obtained by X-ray imaging by the X-ray diagnostic apparatus 12. . The detection support means 27 combines a plurality of CT images obtained in the past by CT imaging by the X-ray CT apparatus 11 and a plurality of X-ray images obtained in the past by X-ray imaging by the X-ray diagnostic apparatus 12. A second AI with matured parameters is constructed by using it for machine learning as second learning data. Then, the detection support means 27 provides an output to the detection function 35 by inputting the combination of the CT image and the X-ray image of the relevant patient to the second AI.

FIG. 3 is a diagram showing an operation example of the detection function 35. As shown in FIG. FIG. 3A is a diagram showing the second training data in building the second AI. FIG. 3B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 3C is a diagram showing the data flow for determining the state of the lesion. The state of a lesion means, for example, the presence or absence of a lesion such as cancer, the amount of change (position and size), and the like.

A plurality of, for example, m CT images Fm that are independent of the patient are input from the X-ray CT apparatus 11 to the detection support means 27 of the diagnosis support apparatus 10 . From the X-ray diagnostic apparatus 12, m X-ray images Gm independent of the patient are input to the detection support means 27 of the diagnosis support apparatus 10. FIG. On the other hand, an appropriate lesion state Qm is input to the detection support means 27 via the input interface 23 so as to correspond to each combination of m CT images and m X-ray images. As a result, the detection support means 27 uses a combination of m CT images Fm and m X-ray images Gm as second learning data, and m lesion states Qm as second teacher data. 2 AI (illustrated in FIG. 3(A)). Note that the detection support unit 27 preferably constructs a plurality of AIs for each imaging region as the second AI.

As shown in FIGS. 3B and 3C, the detection function 35 receives a CT image f of the patient from the X-ray CT apparatus 11 and an X-ray image g of the patient from the X-ray diagnostic apparatus 12. and are provided to the detection support means 27 . As a result, the detection support means 27 inputs the combination of the CT image f and the X-ray image g to the second AI, and the output of the second AI, that is, the state q of the lesion of the patient is detected by the detection function 35. provide to

The detection function 35 receives the output of the second AI, that is, the state q of the lesion from the detection support means 27, and detects an appropriate state of the lesion of the patient based on the state q of the lesion. . Here, the detection function 35 detects the state q itself of the lesion from the detection support means 27 as an appropriate state of the lesion of the patient. Further, the detection function 35 causes the display 24 to display the appropriate state of the lesion of the patient.

Returning to the description of FIG. 1, the X-ray CT apparatus 11 performs CT imaging on a patient to generate a CT image and provides it to the diagnosis support apparatus 10 via the network N. FIG. Here, the X-ray CT apparatus (first client apparatus) 11 is provided with transmission means for transmitting a CT image of the relevant patient to the diagnosis support apparatus (server apparatus) 10 .

The X-ray diagnostic apparatus 12 performs X-ray imaging according to the X-ray imaging conditions provided from the diagnostic support apparatus 10 to generate an X-ray image, and provides the X-ray image to the diagnostic support apparatus 10 via the network N. Here, the X-ray diagnostic apparatus 12 may request the X-ray imaging conditions for the patient from the diagnosis support apparatus 10 and acquire the X-ray imaging conditions in response to the request. In this case, the X-ray diagnostic apparatus (second client apparatus) 12 includes request means for requesting the X-ray imaging conditions for the patient from the diagnosis support apparatus (server apparatus) 10, and diagnosis support in response to the request by the request means. and acquisition means for acquiring the X-ray imaging conditions from the apparatus 10 .

FIG. 4 is a diagram showing the operation of the diagnostic support system 1 as a flowchart. In FIG. 4, numerals attached to "ST" indicate respective steps of the flow chart.

There is a workflow in which an operator such as a diagnostician observes the progress of a lesion using medical images such as CT images for treatment of a lesion such as cancer and follow-up observation of prognosis. The diagnosis support system 1 is used, for example, for treatment of lesions and follow-up of patients in follow-up of prognosis.

The first image acquisition function 31 acquires a CT image of a relevant patient having a lesion from the X-ray CT apparatus 11 via the network interface 25 (step ST1). Here, in order to perform follow-up observation of the lesion from the acquisition of the CT images acquired in step ST1, it is common to generate CT images using an X-ray CT apparatus, which is the same modality apparatus. However, it takes too much time and effort for the patient to go to a large hospital equipped with an X-ray CT apparatus to perform CT imaging, and the patient is exposed to a large amount of radiation during CT imaging.

On the other hand, in order to perform follow-up observation of the lesion from the acquisition of the CT image acquired in step ST1, a modality device other than the same modality device, such as an X-ray that can be possessed by a medical institution close to the patient, is used. It is also possible that the diagnostic equipment is used to generate the X-ray images. In this case, comparative images can be easily obtained at a medical institution close to the patient, and the patient's exposure to radiation is relatively low.

Therefore, in order to observe the progress of the lesion from the acquisition of the CT image acquired in step ST1, an X-ray diagnostic apparatus 12 that is not of the same modality is used to generate an X-ray image with high detection capability of the lesion. think of doing The determination function 32 determines appropriate X-ray imaging conditions for acquiring an X-ray image of the relevant patient by the X-ray diagnostic apparatus 12 from the CT image obtained by CT imaging by the X-ray CT apparatus 11 (step ST2 ). Note that the X-ray imaging conditions include at least the imaging direction.

For example, the decision function 32 controls the decision support means 26 to decide appropriate X-ray imaging conditions for acquiring X-ray images of the relevant patient. As described above, the decision support unit 26 constructs the first AI by using a plurality of CT images acquired in the past by CT imaging of the X-ray CT apparatus 11 as learning data for machine learning. Then, the decision support means 26 inputs the CT image of the relevant patient supplied from the decision function 32 to the first AI, thereby setting the X-ray imaging conditions corresponding to the CT image of the relevant patient to the first AI. It is obtained as the output of AI and output to the decision function 32 . The determination function 32 determines appropriate X-ray imaging conditions for the patient based on the output of the determination support means 26, that is, the X-ray imaging conditions.

The imaging condition providing function 33 provides the X-ray imaging conditions determined in step ST2 to the X-ray diagnostic apparatus 12 (step ST3). The X-ray diagnostic apparatus 12 performs X-ray imaging of the patient according to the X-ray imaging conditions provided in step ST3 (step ST4). The CT image of the patient and the determined X-ray imaging conditions are input to the first AI as training data.

In this way, by providing appropriate X-ray imaging conditions, the X-ray diagnostic apparatus 12 generates an X-ray image with a high detectability of lesions comparable to the CT image of the patient in step ST4. can do.

The second image acquisition function 34 acquires, from the X-ray diagnostic apparatus 12 via the network interface 25, the X-ray image of the patient captured by the X-ray diagnostic apparatus 12 under the X-ray imaging conditions determined in step ST2. (step ST5). The detection function 35 detects the relevant patient based on the CT image of the relevant patient obtained by the CT imaging of the X-ray CT apparatus 11 and the X-ray image obtained by the X-ray imaging of the X-ray diagnostic apparatus 12. The state of the lesion is detected (step ST6).

For example, the detection function 35 controls the detection support means 27 to detect an appropriate change in the lesion of the patient. As described above, the detection support means 27 uses a plurality of CT images acquired in the past by CT imaging by the X-ray CT apparatus 11 and a plurality of X-rays acquired in the past by X-ray imaging by the X-ray diagnostic apparatus 12. A second AI is constructed by using combinations with images as learning data for machine learning. Then, the detection support means 27 inputs the combination of the CT image and the X-ray image of the relevant patient, supplied from the detection function 35, to the second AI, so that the state of the lesion of the relevant patient is displayed as the second AI. is acquired as the output of the AI and output to the detection function 35 . The detection function 35 detects an appropriate state of the lesion of the patient based on the output of the detection support means 27, that is, the state of the lesion of the patient. A change in a lesion means, for example, the presence or absence of a lesion such as cancer, the amount of change (position and size), and the like.

The detection function 35 causes the display 24 to display the change in the lesion of the patient detected in step ST6 (step ST7). In step ST7, the detection function 35 may display the change in the lesion by overlaying it on the X-ray image of the patient, or display it as character information. A combination of the CT image and the X-ray image of the patient and the state of the detected lesion are input to the second AI as training data.

In this way, in step ST7, the diagnosis support apparatus 10 determines an appropriate diagnosis for the patient based on the CT image of the patient and the X-ray image of the patient, which has an appropriate ability to detect lesions. The automatic diagnosis result (state of the lesion) can be presented to an operator such as a diagnostician. Based on the change in the lesion presented via the display 24, the operator can visually confirm the highly accurate automatic diagnosis result for the patient.

As described above, according to the decision function 32 of the diagnosis support system 1, it is possible to provide the X-ray diagnostic apparatus 12 with appropriate X-ray imaging conditions according to the lesion in the CT image of the patient. Further, according to the detection function 35 of the diagnosis support system 1, it is possible to detect an appropriate state of the lesion according to the combination of the CT image and the X-ray image of the patient and present it to the operator.

1. Modified Example of Operation of Determining Function 32 A case where the determining function 32 determines appropriate X-ray imaging conditions for a patient based on a CT image of the patient has been described with reference to FIGS. However, it is not limited to that case. Modifications of the operation of the determination function 32 will be described below with reference to FIGS. 5 to 10. FIG.

1-1. First Modified Example of Operation of Determination Function 32 The determination function 32 uses the determination support means 26A to determine appropriate X-ray imaging conditions using the CT image and additional information of the CT image.

FIG. 5 is a diagram showing an operation example of the decision function 32. As shown in FIG. FIG. 5A is a diagram showing the first training data in constructing the first AI. FIG. 5B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 5C is a diagram showing the flow of data for determining X-ray imaging conditions.

A plurality of patient-independent CT images Fm, for example, m CT images Fm and accompanying information Rm are input from the X-ray CT apparatus 11 to the decision support means 26A of the diagnosis support apparatus 10 . The additional information means findings, positions of lesions, overlay information, and the like included in DICOM (Digital Imaging and Communications in Medicine) data attached to medical images. On the other hand, m X-ray imaging conditions Pm are input to the determination support means 26A via the input interface 23 so as to correspond to each of the m CT images. As a result, the decision support unit 26A uses the combination of the m CT images Fm and the m incidental information Rm as the first learning data, and the m X-ray imaging conditions Pm as the first teacher data. (illustrated in FIG. 5(A)).

As shown in FIGS. 5B and 5C, the decision function 32 provides the CT image f of the relevant patient from the X-ray CT apparatus 11 and its incidental information r to the decision support means 26A. Thereby, the decision support means 26A inputs the CT image f and the additional information r to the first AI, and provides the output of the first AI, that is, the X-ray imaging condition p for the patient to the decision function 32. .

The determination function 32 receives the output of the first AI, that is, the X-ray imaging conditions p from the determination support means 26A, and determines appropriate X-ray imaging conditions for the patient based on the X-ray imaging conditions p. , providing it to the X-ray diagnostic apparatus 12 . Here, the determination function 32 determines the X-ray imaging condition p itself from the determination support means 26A as an appropriate X-ray imaging condition for the patient.

As described above, according to the operation of the determination function 32 shown in FIG. 5, the output of the X-ray imaging conditions for the patient is more efficient than the case where the X-ray imaging conditions are determined based only on the CT image of the patient. Improves accuracy.

1-2. Second Modified Example of Operation of Decision Function 32 The decision function 32 utilizes the decision support means 26B to determine appropriate X-ray images for re-imaging using X-ray images generated by X-ray imaging in addition to CT images. Determine line imaging conditions.

FIG. 6 is a diagram showing a second modification of the operation of the decision function 32. As shown in FIG. FIG. 6A is a diagram showing the first training data in constructing the first AI. FIG. 6B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 6C is a diagram showing the flow of data for determining X-ray imaging conditions.

A plurality of patient-independent CT images Fm, for example, m CT images Fm, are input from the X-ray CT apparatus 11 to the decision support means 26B of the diagnosis support apparatus 10 . On the other hand, m X-ray imaging conditions Pm and m X-ray images Gm independent of the patient are input from the X-ray diagnostic apparatus 12 to the decision support means 26B of the diagnosis support apparatus 10 . A combination of the m-th CT image Fm and the m-th X-ray image Gm is associated with the m-th X-ray imaging condition Pm. The decision support means 26 uses a combination of m CT images Fm and m X-ray images Gm as first learning data, and uses m X-ray imaging conditions Pm as first teacher data to generate a first AI (illustrated in FIG. 6(A)).

As shown in FIGS. 6B and 6C, the determination function 32 receives a CT image f of the patient from the X-ray CT apparatus 11 and an X-ray image g of the patient from the X-ray diagnostic apparatus 12. and are provided to the decision support means 26B of the diagnosis support device 10 . The X-ray image g is generated by, for example, X-ray imaging of the relevant patient under the X-ray imaging conditions determined by the determination functions 32 and 32A described above. Thereby, the decision support means 26B inputs the CT image f and the X-ray image g to the first AI, and provides the output of the first AI, that is, the X-ray image p of the relevant patient to the decision function 32. do.

The determination function 32 receives the output of the first AI, that is, the X-ray imaging conditions p from the determination support means 26B, and determines appropriate X-ray imaging conditions for the patient based on the X-ray imaging conditions p. , providing it to the X-ray diagnostic apparatus 12 . Here, the determination function 32 determines the X-ray imaging condition p itself from the determination support means 26B as an appropriate X-ray imaging condition for the patient.

As described above, according to the operation of the determination support means 26B shown in FIG. 6, the optimum X-ray imaging conditions for the relevant patient are obtained, and the appropriate X-ray imaging conditions for re-imaging of the relevant patient can be determined by X-ray diagnosis. device 12 can be provided.

1 to 6, the decision support means 26 to 26B directly acquire the X-ray imaging conditions and output them to the decision functions 32 to 32B, and the decision functions 32 to 32B output the decision support means 26 to 26B. A case has been described in which the X-ray imaging condition itself is determined as an appropriate X-ray imaging condition. However, it is not limited to that case. For example, the decision support means generates a plurality of parameters for determining appropriate X-ray imaging conditions and outputs them to the decision function 32, and the decision function 32 determines appropriate conditions based on the parameters output from the decision support means. It may also be the case of determining the appropriate X-ray imaging conditions. This case will be described with reference to FIGS. 7 to 10. FIG.

1-3. Third Modified Example of Operation of Determination Function 32 The determination function 32 utilizes the determination support means 26C to generate a plurality of virtual X-ray images as a plurality of parameters for determining appropriate X-ray imaging conditions. , determine appropriate X-ray imaging conditions based on a plurality of virtual X-ray images.

FIG. 7 is a diagram showing a third modification of the operation of the decision function 32. As shown in FIG. FIG. 7A is a diagram showing the first training data in constructing the first AI. FIG. 7B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 7C is a diagram showing the flow of data for determining X-ray imaging conditions.

A plurality of, for example, m CT images Fm that do not depend on the patient are input from the X-ray CT apparatus 11 to the decision support means 26C of the diagnosis support apparatus 10 . On the other hand, m X-ray imaging conditions Pm and m X-ray images Gm independent of the patient are input from the X-ray diagnostic apparatus 12 to the decision support means 26C of the diagnosis support apparatus 10 . The m CT images Fm and the m X-ray imaging conditions Pm are respectively associated with the m X-ray images Gm. The decision support means 26C uses a combination of m CT images Fm and m X-ray imaging conditions Pm as first learning data, and uses m X-ray images Gm as first teacher data to generate a first AI (illustrated in FIG. 7(A)).

As shown in FIGS. 7B and 7C, the determination function 32 extracts the CT image f of the patient from the X-ray CT apparatus 11 and the X-ray imaging conditions arbitrarily set via the input interface 23. and p are input to the decision support means 26C. Thereby, the decision support unit 26C inputs the CT image f and the X-ray imaging condition p to the first AI, and outputs the first AI, that is, the third medical image (virtual X-ray image) of the patient. ) h to decision function 32 . Further, the determination support means 26C uses the first AI in a loop according to a plurality of different X-ray imaging conditions, so that a plurality of virtual X-ray images are used as a plurality of parameters for determining appropriate X-ray imaging conditions. h can be generated.

The decision function 32 receives the output of the first AI, that is, the plurality of virtual X-ray images h corresponding to the plurality of X-ray imaging conditions from the decision support means 26C, and based on the plurality of virtual X-ray images h, Appropriate X-ray imaging conditions for the patient are determined and provided to the X-ray diagnostic apparatus 12 . For example, while the operator refers to a plurality of virtual X-ray images h, the operator operates the input interface 23 to select a desired virtual X-ray image from among the plurality of virtual X-ray images h. 26C determines the conditions corresponding to the selected virtual X-ray image as appropriate X-ray imaging conditions. Further, for example, the determination support unit 26C determines, as an appropriate X-ray imaging condition, a condition corresponding to a virtual X-ray image having the highest image quality information among the plurality of virtual X-ray images h. Quality information means an index (for example, 1 to 10 points) based on the S/N (Signal to Noise) and CNR (Contrast to Noise Ratio) ratios of an image.

As described above, according to the operation of the determination function 32 shown in FIG. 7, a virtual X-ray image is generated as a parameter for determining appropriate X-ray imaging conditions, and an appropriate X-ray image is generated based on the virtual X-ray image. Imaging conditions can be set.

1-4. Fourth Modified Example of Operation of Determination Function 32 The determination function 32 utilizes the determination support means 26D to generate a plurality of pieces of quality information as a plurality of parameters for determining appropriate X-ray imaging conditions. Appropriate X-ray imaging conditions are determined based on the information. Quality information means an index (for example, 1 to 10 points) based on the S/N (Signal to Noise) and CNR (Contrast to Noise Ratio) ratios of an image.

FIG. 8 is a diagram showing a fourth modification of the operation of the decision function 32. As shown in FIG. FIG. 8A is a diagram showing the first training data in constructing the first AI. FIG. 8B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 8C is a diagram showing the flow of data for determining X-ray imaging conditions.

A plurality of patient-independent CT images Fm, for example, m CT images Fm are input from the X-ray CT apparatus 11 to the decision support means 26D of the diagnosis support apparatus 10 . From the X-ray diagnostic apparatus 12, m X-ray images Gm independent of the patient are input to the decision support means 26D of the diagnosis support apparatus 10. FIG. In addition, m pieces of quality information Sm independent of the patient generated based on the CT image and the X-ray image are input to the decision support means 26D. A combination of the m-th CT image Fm and the m-th X-ray image Gm is associated with the m-th quality information Sm. The decision support means 26D constructs a first AI using a combination of m CT images Fm and m X-ray images Gm as first learning data and m quality information Sm as first teacher data. (illustrated in FIG. 8(A)).

As shown in FIGS. 8B and 8C, the determination function 32 receives a CT image f of the patient from the X-ray CT apparatus 11 and an X-ray image g of the patient from the X-ray diagnostic apparatus 12. are input to the decision support means 26D of the diagnosis support device 10. Thereby, the decision support means 26D inputs the CT image f and the X-ray image g to the first AI and provides the output of the first AI, ie the quality information s about the patient, to the decision function 32 . In addition, the determination support means 26D uses the first AI in a loop with a plurality of different CT images and X-ray images, and uses a plurality of quality information as a plurality of parameters for determining appropriate X-ray imaging conditions. s can be generated.

The decision function 32 receives the output of the first AI, that is, the plurality of quality information s corresponding to the plurality of X-ray imaging conditions from the decision support means 26D, and based on the plurality of quality information s, determines Appropriate X-ray imaging conditions are determined and provided to the X-ray diagnostic apparatus 12 . For example, while referring to a plurality of quality information s, the operator operates the input interface 23 to select the desired quality information from among the plurality of quality information s. A condition corresponding to the quality information is determined as an appropriate X-ray imaging condition. Also, for example, the determination support unit 26D determines the conditions corresponding to the highest quality information as appropriate X-ray imaging conditions.

As described above, according to the operation of the determination function 32 shown in FIG. 8, image quality information is generated as a parameter for determining appropriate X-ray imaging conditions, and appropriate X-ray imaging is performed based on the image quality information. Imaging conditions can be set.

1-5. Fifth Modified Example of Operation of Determination Function 32 The determination function 32 utilizes the determination support means 26E to generate a plurality of quality information as a plurality of parameters for determining appropriate X-ray imaging conditions, Appropriate X-ray imaging conditions are determined based on the information.

FIG. 9 is a diagram showing a fifth modification of the operation of the decision function 32. As shown in FIG. FIG. 9A is a diagram showing the first training data in constructing the first AI. FIG. 9B is a block diagram showing a configuration for determining X-ray imaging conditions. FIG. 9C is a diagram showing the flow of data for determining X-ray imaging conditions.

A plurality of patient-independent CT images Fm, for example, m CT images Fm, are input from the X-ray CT apparatus 11 to the decision support means 26E of the diagnosis support apparatus 10 . From the X-ray diagnostic apparatus 12, m X-ray imaging conditions Pm independent of the patient are input to the decision support means 26E of the diagnosis support apparatus 10. FIG. In addition, m pieces of quality information Sm independent of the patient generated based on the CT image and the X-ray image are input to the decision support means 26E. Each combination of the m-th CT image Fm and the m-th X-ray imaging condition Pm is associated with the m-th quality information Sm. The decision support means 26E uses a combination of m CT images Fm and m X-ray imaging conditions Pm as first learning data and m quality information Sm as first teacher data to generate a first AI. construct (illustrated in FIG. 9(A)).

As shown in FIGS. 9B and 9C, the determination function 32 uses the CT image f of the patient from the X-ray CT apparatus 11 and the X-ray imaging conditions arbitrarily set via the input interface 23. and p are input to the decision support means 26E of the diagnosis support device 10 . Thereby, the decision support means 26E inputs the CT image f and the X-ray imaging condition p to the first AI, and outputs the output of the first AI, that is, the appropriate quality information s related to the patient to the detection function 32. offer. In addition, the determination support means 26E uses the first AI in a loop according to a plurality of different X-ray imaging conditions, and obtains a plurality of pieces of quality information s as a plurality of parameters for determining appropriate X-ray imaging conditions. can be generated.

The decision function 32 receives the output of the first AI, that is, the plurality of quality information s corresponding to the plurality of X-ray imaging conditions from the decision support means 26E, and based on the plurality of quality information s, determines Appropriate X-ray imaging conditions are determined and provided to the X-ray diagnostic apparatus 12 . For example, while referring to a plurality of quality information s, the operator operates the input interface 23 to select the required quality information from among the plurality of quality information s. A condition corresponding to the quality information is determined as an appropriate X-ray imaging condition. Also, for example, the determination support unit 26E determines the conditions corresponding to the highest quality information as appropriate X-ray imaging conditions.

As described above, according to the operation of the determination function 32 shown in FIG. 9, image quality information is generated as a parameter for determining appropriate X-ray imaging conditions, and appropriate X-ray imaging conditions are generated based on the image quality information. Imaging conditions can be set.

In the above, the case where the decision function 32 decides the X-ray imaging conditions using any one of the five decision support functions 26 to 26E, that is, the five AIs has been described, but the present invention is limited to that case. is not. For example, decision function 32 may use three decision support functions 26C-26E, ie, three AIs, to determine x-ray imaging conditions.

FIG. 10 is a diagram showing a sixth modification of the operation of the decision function 32. As shown in FIG. FIG. 10 is a diagram showing the flow of data when using multiple AIs.

As shown in FIG. 10, decision function 32 receives a plurality of virtual X-ray images h (shown in FIG. 7) corresponding to a plurality of X-ray imaging conditions from decision support means 26C, and a plurality of X-ray images h from decision support means 26D. A plurality of quality information s (shown in FIG. 8) corresponding to the radiographic conditions are received, and a plurality of quality information s (shown in FIG. 8) corresponding to the plurality of X-ray imaging conditions are received from the decision support means 26E. The operator operates the input interface 23 while referring to quality information corresponding to a plurality of virtual X-ray images h and a plurality of quality information s to select desired quality information from a plurality of quality information. , determination function 32 determines the conditions corresponding to the selected quality information as appropriate X-ray imaging conditions. Also, for example, the determination function 32 determines, as an appropriate X-ray imaging condition, a condition corresponding to the highest quality information among a plurality of pieces of quality information including quality information corresponding to a plurality of virtual X-ray images h.

As described above, according to the operation of the determination function 32 shown in FIG. 10, the virtual X-ray image and image quality information are generated as parameters for determining appropriate X-ray imaging conditions, and the virtual X-ray image and image quality information are generated. appropriate X-ray imaging conditions can be determined based on this quality information.

2. Modified Example of Operation of Detecting Function 35 Using FIGS. 1 to 4, the case where the detecting function 35 detects an appropriate state of a lesion of a relevant patient based on a CT image and an X-ray image of the relevant patient. explained. However, it is not limited to that case. Modified examples of the operation of the detection function 35 will be described below with reference to FIGS. 11 to 14. FIG.

2-1. First Modified Example of Operation of Detection Function 35 The detection function 35 uses the detection support means 27A to detect the appropriate state of the lesion based on the X-ray imaging conditions in addition to the CT image and the X-ray image. A plurality of pieces of position information are generated as a plurality of parameters of , and an appropriate state of the lesion is detected based on the plurality of pieces of position information.

11A and 11B are diagrams showing a first modification of the operation of the detection function 35. FIG. FIG. 11A is a diagram showing the second training data in constructing the second AI. FIG. 11B is a block diagram showing a configuration for determining the state of lesions. FIG. 11C is a diagram showing the data flow for determining the state of the lesion.

A plurality of, for example, m CT images Fm that do not depend on the patient are input from the X-ray CT apparatus 11 to the detection support means 27A of the diagnosis support apparatus 10 . Combinations of m X-ray imaging conditions Pm independent of patients and m X-ray images Gm are input from the X-ray diagnostic apparatus 12 to the detection support means 27A of the diagnosis support apparatus 10 . In addition, m lesion positions Tm are input to the detection support means 27A via the input interface 23 . Alternatively, the position Tm of the lesion is specified by CAD, which automatically detects the lesion by image analysis of the CT image, and is input to the detection support means 27A. A combination of the m-th CT image Fm, the m-th X-ray image Gm, and the m-th X-ray imaging condition Pm is associated with the m-th lesion state Qm. The detection support means 27A uses combinations of m CT images Fm, m X-ray images Gm, and m X-ray imaging conditions Pm as second learning data, and m lesion states Qm as A second AI is constructed as second teacher data (illustrated in FIG. 11(A)).

As shown in FIGS. 11B and 11C, the detection function 35 receives a CT image f of the patient from the X-ray CT apparatus 11 and an X-ray imaging condition p and an X-ray image from the X-ray diagnostic apparatus 12. A combination of the images g is input to the detection support means 27A of the diagnosis support device 10 . As a result, the detection support means 27A inputs the combination of the CT image f, the X-ray image g, and the X-ray imaging condition p to the second AI, and outputs the output of the second AI, that is, the lesion area of the patient. to the detection function 35 . The detection support means 27A detects the lesion area of the patient based on the CT image f having the position information of the lesion area of the patient and the latest X-ray image (projection image) g of the patient. Current position information (candidate area of lesion) can be obtained. In addition, the detection support unit 27A uses the second AI in a loop according to a plurality of different X-ray imaging conditions to obtain a plurality of parameters related to the lesion as a plurality of parameters for detecting an appropriate state of the lesion. You can also generate a position.

The detection function 35 receives the output of the second AI, that is, the position t of the lesion from the detection support means 27A, and based on the position t of the lesion, determines the appropriate state of the lesion (lesion The presence or absence of and the amount of change (position and size)) are detected and displayed on the display 24 .

As described above, according to the operation of the detection function 35 shown in FIG. state output accuracy is improved.

2-2. Second Modified Example of Operation of Detection Function 35 The detection function 35 utilizes the detection support means 27B to determine appropriateness of the lesion based on the virtual X-ray image in addition to the CT image, the X-ray image, and the X-ray imaging conditions. A plurality of pieces of positional information are generated as a plurality of parameters for detecting an appropriate state, and an appropriate state of the lesion is detected based on the plurality of pieces of positional information.

12A and 12B are diagrams showing a second modification of the operation of the detection function 35. FIG. FIG. 12(A) is a diagram showing the second training data in constructing the second AI. FIG. 12B is a block diagram showing the configuration for determining the state of the lesion. FIG. 12C is a diagram showing the data flow for determining the state of the lesion.

A plurality of, for example, m CT images Fm that do not depend on the patient are input from the X-ray CT apparatus 11 to the detection support means 27B of the diagnosis support apparatus 10 . Combinations of m X-ray imaging conditions Pm and X-ray images Gm independent of patients are input from the X-ray diagnostic apparatus 12 to the detection support means 27B of the diagnosis support apparatus 10 . In addition, m lesion positions Tm are input to the detection support means 27B via the input interface 23 . Alternatively, the position Tm of the lesion is specified by CAD, which automatically detects the lesion by image analysis of the CT image, and is input to the detection support means 27B. Further, the virtual X-ray image Hm is input from the determination support means 26C (illustrated in FIG. 7) to the detection support means 27B of the diagnosis support apparatus 10. FIG.

The combination of the m-th CT image Fm, the m-th X-ray image Gm, the m-th X-ray imaging condition Pm, and the m-th X-ray image Hm is the state of the m-th lesion area. Qm. The detection support means 27B uses a combination of m CT images Fm, m X-ray images Gm, m X-ray imaging conditions Pm, and m virtual X-ray images Hm as second learning data, A second AI is constructed using the m lesion positions Tm as second teacher data (illustrated in FIG. 12A).

As shown in FIGS. 12B and 12C, the detection function 35 receives a CT image f of the patient from the X-ray CT apparatus 11 and an X-ray imaging condition p and an X-ray image from the X-ray diagnosis apparatus 12. Input a combination of images g and a virtual X-ray image h from the decision support means 26C. Thereby, the detection support means 27B inputs the combination of the CT image f, the X-ray image g, the X-ray imaging condition p, and the virtual X-ray image h to the second AI, and the output of the second AI, that is, It provides the detection function 35 with the appropriate lesion location t for that patient. The detection support means 27B detects the lesion area of the patient based on the CT image f having the position information of the lesion area of the patient and the latest X-ray image (projection image) g of the patient. Current position information (candidate area of lesion) can be obtained. In addition, the detection support unit 27B uses the second AI in a loop according to a plurality of different X-ray imaging conditions to obtain a plurality of parameters related to the lesion as parameters for detecting an appropriate state of the lesion. You can also generate a position.

The detection function 35 receives the output of the second AI, that is, the position t of the lesion from the detection support means 27B, and detects the appropriate state of the lesion of the patient based on the position t of the lesion. , to display it on the display 24 .

As described above, according to the operation of the detection function 35 shown in FIG. 12, compared to the case where the state of the lesion is detected based only on the CT image, the X-ray image, and the X-ray imaging conditions of the patient, The output accuracy of the state of the lesion related to the patient is improved.

2-3. Third Modification of Operation of Detection Function The detection function 35 uses the detection support means 27C to detect an appropriate state of the lesion based on virtual X-ray images in addition to CT images and X-ray images.

13A and 13B are diagrams showing a third modification of the operation of the detection function 35. FIG. FIG. 13A is a diagram showing the second training data in constructing the second AI. FIG. 13B is a block diagram showing the configuration for determining the state of the lesion. FIG. 13(C) is a diagram showing the flow of data for determining the state of the lesion.

A plurality of, for example, m CT images Fm are input from the X-ray CT apparatus 11 to the detection support means 27C of the diagnosis support apparatus 10 . From the X-ray diagnostic apparatus 12, m X-ray images Gm independent of the patient are input to the detection support means 27C of the diagnosis support apparatus 10. FIG. In addition, m lesion states Sm are input to the detection support means 27C via the input interface 23 . Furthermore, the virtual X-ray image Hm is input to the detection support means 27C from the decision support means 26C (shown in FIG. 7).

A combination of the m-th CT image Fm and the m-th X-ray image Gm is associated with the m-th lesion state Sm. A combination of the m-th CT image Fm and the m-th virtual X-ray image Hm is associated with the m-th lesion state Sm. The detection support means 27C constructs a second AI using m CT images Fm and m X-ray images Gm as second learning data and m lesion states Sm as second teacher data. (illustrated in FIG. 13(A)).

As shown in FIGS. 13B and 13C, the detection function 35 receives a CT image f of the patient from the X-ray CT apparatus 11 and an X-ray image g from the X-ray diagnosis apparatus 12 for diagnosis. It is input to the detection support means 27C of the support device 10 . As a result, the detection support means 27C inputs the combination of the CT image f and the X-ray image g to the second AI, and the output of the second AI, that is, the state s of the lesion of the patient is detected by the detection function 35. can be supplied to On the other hand, the detection function 35 inputs the CT image f of the relevant patient from the X-ray CT apparatus 11 and the virtual X-ray image h from the decision support means 26C to the detection support means 27C. As a result, the detection support means 27C inputs the combination of the CT image f and the virtual X-ray image h to the second AI, and detects the output of the second AI, that is, the state s of the lesion of the patient. 35 can be supplied.

The detection function 35 receives the output of the second AI, that is, the plurality of states s related to the lesion from the detection support means 27C, and based on the plurality of states s related to the lesion, determines the lesion of the patient. A suitable condition is detected and displayed on the display 24 . Here, the detection function 35 may detect the representative value itself of the plurality of states s related to the lesion from the detection support means 27C as the appropriate state of the lesion related to the patient.

As described above, according to the operation of the detection function 35 shown in FIG. 13, the change in the position t of the lesion can be presented to the operator.

In the above, the case where the detection function 35 detects the state of the lesion using any one of the four detection support functions 27 to 27C, that is, the four AIs has been described, but the present invention is limited to that case. is not. For example, the detection function 35 can detect the state of the lesion using two detection support functions 27B-27C, that is, two AIs.

2-4. Fourth Modification of Operation of Detection Function FIG. 14 is a diagram showing a fourth modification of the operation of the detection function 35 . FIG. 14 is a diagram showing a data flow when using a plurality of AIs.

As shown in FIG. 14, the detection function 35 receives multiple lesion positions t from the detection support means 27A and 27B. The operator operates the input interface 23 to select a desired lesion position from among the plurality of lesion positions t while referring to the positions t of the plurality of lesions. The position of the selected lesion is detected as an appropriate state of the lesion.

As described above, according to the operation of the detection function 35 shown in FIG. 14, the position of the lesion is generated as a parameter for detecting the appropriate state of the lesion, and the appropriate state of the lesion is generated based on the position of the lesion. state can be detected.

3. Modification of Configuration of Diagnosis Support System 1 3-1. Modification of First Modality Apparatus 11 In FIG. 1, the case where the first modality apparatus 11 of the diagnosis support system 1 is one type of modality apparatus (X-ray CT apparatus) has been described, but it is not limited to that case. not something. The first modality apparatus 11 may be a plurality of types, for example, two types of X-ray CT apparatus and MRI apparatus. In that case, the decision support means 26 sets a combination of a plurality of CT images and MRI images acquired in the past as the first learning data, sets the X-ray imaging conditions corresponding to each combination as the first teaching data, and uses the machine Build the first AI by using it for learning. Then, the decision function 32 inputs the CT image and the MRI image of the patient into the first AI, thereby deciding the appropriate X-ray imaging conditions corresponding to the CT image and the MRI image of the patient. Provided to the line diagnosis device 12 .

According to the configuration of the first modality apparatus 11 as described above, the output accuracy of the X-ray imaging conditions for the patient is improved compared to the case where only the CT image for the patient is input to the first AI. .

3-2. Modification of Second Modality Apparatus 12 In FIG. 1, the case where the second modality apparatus 12 of the diagnosis support system 1 is one type of modality apparatus (X-ray diagnostic apparatus) has been described, but it is not limited to that case. not something. The second modality device 12 may be a plurality of types, for example, two types of X-ray diagnostic device and ultrasonic diagnostic device. In that case, the decision support means 26 treats a plurality of CT images acquired in the past as first learning data, and treats the X-ray imaging conditions and ultrasound scanning conditions corresponding to each CT image as first teaching data. , to build the first AI by using it for machine learning. Then, the determination function 32 inputs the CT image of the patient to the first AI, thereby determining appropriate X-ray imaging conditions and ultrasonic scan conditions corresponding to the CT image of the patient, and X They are provided for line diagnostic equipment and ultrasonic diagnostic equipment, respectively. The ultrasound scan conditions mean the ultrasound transmission frequency, pulse repetition frequency (PRF), frame rate, and the like.

According to the configuration of the second modality apparatus 12 as described above, the output accuracy of the X-ray imaging conditions for the patient is improved compared to the case where the X-ray imaging conditions are determined based only on the CT image of the patient. improves.

Further, when the second modality device 12 is two types of X-ray diagnostic device and ultrasonic diagnostic device, the decision support means 26 uses a plurality of CT images acquired in the past as first learning data, and each A first AI is constructed by using the type of modality device corresponding to a CT image as first teacher data and using it for machine learning. Then, the determination function 32 inputs the CT image of the relevant patient to the first AI, determines the type of appropriate modality device corresponding to the CT image of the patient, and determines the appropriate X-ray diagnostic device or It can also be provided for ultrasound diagnostic equipment.

According to at least one embodiment described above, appropriate imaging conditions for acquiring the second image to be generated can be provided based on the first image.

The first image acquisition function 31 is an example of image acquisition means and first image acquisition means. The decision function 32 is an example of decision means. The imaging condition providing function 33 is an example of imaging condition providing means. The second image acquisition function 34 is an example of second image acquisition means. The detection function 35 is an example of detection means.

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

1 diagnosis support system 10 diagnosis support device 11 image providing device (first modality device)
12 second modality device 21 processing circuit 22 storage circuits 26 to 26E determination support means 27 to 27C detection support means 31 first image acquisition function 32 determination function 33 imaging condition provision function 34 second image acquisition function 35 detection function

Claims (17)

an image acquisition means for acquiring a first image of a subject imaged by an X-ray CT apparatus;
determination means for determining imaging conditions including an imaging direction of the subject for generating a second image of the subject by an X-ray diagnostic apparatus from the first image obtained by the image obtaining means;
A diagnostic support device comprising: an imaging condition providing means for providing the imaging condition determined by the determining means to the X-ray diagnostic apparatus and causing the X-ray diagnostic apparatus to perform imaging of the subject according to the imaging condition;
The diagnosis support device according to claim 1. The determination means obtains information specifying the position of the lesion from the first image, and determines imaging conditions including the imaging direction of the subject based on the information.
3. The diagnostic support device according to claim 1 or 2. a first image acquisition means as the image acquisition means;
a second image acquiring means for acquiring a second image of the subject imaged by the X-ray diagnostic apparatus under the imaging condition;
detecting the state of the lesion of the subject based on the first image acquired by the first image acquisition means and the second image acquired by the second image acquisition means; further comprising a detection means;
3. The diagnostic support device according to claim 1 or 2. The image acquisition means further acquires incidental information of the first image,
The determination means determines the imaging conditions from a combination of the first image and the incidental information acquired by the image acquisition means.
5. The diagnosis support device according to any one of claims 1 to 4. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device ;
The determining means determines the imaging conditions from a combination of the first image and the second image.
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device ;
The determination means acquires a plurality of third images from a combination of the first image and a plurality of imaging conditions, and determines the imaging conditions from the plurality of third images.
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device ;
The determining means acquires a plurality of quality information from a combination of a plurality of first images and a plurality of second images, and determines the imaging condition from the plurality of quality information.
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device ;
The determining means obtains a plurality of quality information from a combination of the first image and a plurality of imaging conditions, and determines the imaging condition from the plurality of quality information.
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device ;
The determining means is
Acquiring a plurality of third images from a combination of the first image and a plurality of imaging conditions;
obtaining a plurality of quality information from a combination of the plurality of first images and the plurality of second images;
Obtaining a plurality of pieces of quality information from a combination of the first image and a plurality of imaging conditions;
determining the imaging condition from a plurality of quality information including a plurality of quality information based on the plurality of third images;
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device under the imaging condition ;
detecting the state of the lesion of the subject based on the first image acquired by the first image acquisition means and the second image acquired by the second image acquisition means; a detection means,
The detection means acquires a plurality of positions of the lesion from a combination of the first image, the second image, and a plurality of imaging conditions, and obtains a plurality of positions of the lesion from the plurality of positions of the lesion. detect the state,
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device under the imaging condition ;
detecting the state of the lesion of the subject based on the first image acquired by the first image acquisition means and the second image acquired by the second image acquisition means; a detection means,
The determining means acquires a third image from a combination of the first image and imaging conditions,
The detection means acquires a plurality of positions of the lesion from a combination of the first image, the second image, a plurality of imaging conditions, and the third image, and obtains a plurality of positions of the lesion. Detecting the state of the lesion from
Diagnostic support device. a first image acquiring means for acquiring a first image of the subject imaged by the first modality device ;
determining an imaging condition including an imaging direction of the subject for generating a second image of the subject by a second modality device from the first image obtained by the first image obtaining means; a determining means;
a second image acquiring means for acquiring a second image of the subject imaged by the second modality device under the imaging condition ;
detecting the state of the lesion of the subject based on the first image acquired by the first image acquisition means and the second image acquired by the second image acquisition means; a detection means,
The determining means acquires a third image from a combination of the first image and imaging conditions,
The detection means is
obtaining the state of the lesion from the combination of the first image and the second image;
obtaining the state of the lesion from the combination of the first image and the third image;
detecting the state of the lesion from the acquired states of the plurality of lesions;
Diagnostic support device. a first image acquiring means for acquiring a first image of a subject imaged by an X-ray CT apparatus;
determination means for determining imaging conditions of the subject from the first image acquired by the first image acquisition means;
a second image acquiring means for acquiring a second image of the subject imaged by the X-ray diagnostic apparatus under the imaging condition;
detecting the state of the lesion of the subject based on the first image acquired by the first image acquisition means and the second image acquired by the second image acquisition means; detecting means;
Diagnostic support device. an image acquisition means for acquiring a first image of a subject captured by an X-ray CT apparatus included in the first client apparatus;
a determining means for determining an imaging condition including an imaging direction of the subject from the first image acquired by the image acquiring means;
a server device comprising: an imaging condition transmitting means for transmitting the imaging conditions determined by the determining means;
a second client device including an X-ray diagnostic apparatus, comprising an acquisition unit that acquires the imaging conditions transmitted from the server device;
diagnostic support system. The second client device further comprises request means for requesting the imaging conditions from the server device,
The acquisition means acquires the imaging conditions from the server device in response to a request from the request means.
The diagnosis support system according to claim 15 . The first client device
transmitting means for transmitting the first image of the subject to the server device;
The diagnostic support system according to claim 15 or 16 .

Images