Imaging support device, imaging support method, and imaging support program

Abstract

To provide an imaging support device, an imaging support method, and an imaging support program capable of guiding a subject to move suitably in capturing a radiation image.SOLUTION: The imaging support device, which supports a subject of an imaging target of a radiographic apparatus capable of kymography, comprises a guide information generation unit which generates guide information guiding the movement of the subject based on information on the case of the subject and a control unit which notifies the subject with the guide information.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a photography support device, a photography support method, and a photography support program. [Background technology]

[0002] Conventionally, a radiography system for performing dynamic imaging of a subject is known (see, for example, Patent Document 1). In such a system, a frame image is displayed when the subject is in a specific situation (for example, a situation where the subject feels pain) during dynamic imaging, and a notification is given that the subject is in the specific situation when the frame image is in that state (see, for example, Patent Document 1). This technology associates the subject's specific situation with the captured image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-11927 Summary of the Invention [Problem to be solved by the invention]

[0004] In a radiation imaging system, the subject needs to move appropriately during imaging in order to accurately grasp the extent of the subject's specific condition. Therefore, a configuration is desired that can guide the subject to move appropriately during radiation imaging.

[0005] An object of the present invention is to provide an imaging support device, an imaging support method, and an imaging support program that are capable of guiding a subject to make appropriate movements when capturing a radiographic image. [Means for solving the problem]

[0006] The photography support device according to the present invention comprises: An imaging support device that supports a subject to be imaged by a radiographic imaging device capable of dynamic imaging, a guidance information generating unit that generates guidance information for guiding a movement of the subject based on information about the case of the subject; a control unit that notifies the subject of the guidance information; Equipped with.

[0007] The photography support method according to the present invention comprises: An imaging support method for supporting a subject to be imaged by a radiographic imaging device capable of dynamic imaging, comprising: generating guidance information for guiding a movement of the subject based on information about the case of the subject; notifying the subject of the guidance information; It has.

[0008] The photography support program according to the present invention comprises: An imaging support program for supporting a subject to be imaged by a radiographic imaging device capable of dynamic imaging, On the computer, A process of generating guidance information for guiding the movement of the subject based on information about the case of the subject; a process of notifying the subject of the guidance information; Execute the following. [Effects of the Invention]

[0009] According to the present invention, it is possible to guide a subject to make appropriate movements when capturing a radiographic image. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing a radiography system including an imaging support device according to an embodiment of the present invention; [Figure 2] 1 is a block diagram showing the configuration of a main part of a radiography control device according to an embodiment of the present invention; [Figure 3]FIG. 1 is a block diagram showing the configuration of a main part of a training device that trains a machine learning model. [Figure 4] 10A and 10B are diagrams illustrating an example of display of guidance information on a display device. [Figure 5] 10 is a flowchart illustrating an example of the operation of a process performed by the radiation imaging control device according to the present embodiment. [Figure 6] FIG. 10 is a diagram illustrating an example of a table in which multiple patterns of guidance information are associated with the positions of guidance target portions. DETAILED DESCRIPTION OF THE INVENTION

[0011] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will now be described in detail with reference to the accompanying drawings. Fig. 1 is a block diagram showing a radiography system including an imaging support device according to an embodiment of the present invention.

[0012] A radiation imaging system is configured by a radiation imaging device 20 installed in an imaging room and a radiation imaging control device (console) 10 installed in an operation room. The radiation imaging system is installed in a predetermined facility (for example, a medical facility such as a hospital). The radiation imaging control device 10 corresponds to the "imaging support device" of the present invention.

[0013] Furthermore, external systems are connected via a communication network to the radiation imaging control device 10. The external systems include a picture archiving and communication system (PACS) 31, a hospital information system (HIS) 32, and a radiology information system (RIS) 33.

[0014] The PACS 31 manages image data captured by the radiation imaging device 20. The HIS 32 and the RIS 33 receive order information related to radiation imaging of a subject from, for example, a doctor, and transmit the received order information to the radiation imaging control device 10. The order information includes various information such as the subject's ID, imaging region, imaging direction, and physique.

[0015] An external dynamic analysis device may be connected via a communication network to the radiation imaging control device 10. In the communication network including the radiation imaging system, PACS 31, HIS 32, and RIS 33, information is transmitted and received in accordance with, for example, the DICOM (Digital Image and Communications in Medicine) standard.

[0016] The radiation imaging device 20 includes a generator 21, an exposure switch 22, a radiation source 23, and an imaging unit 24. Based on the operation of the exposure switch 22, the generator 21 applies a voltage according to preset imaging conditions to the radiation source 23, which includes, for example, a tube. When a voltage is applied from the generator 21, the radiation source 23 generates radiation (for example, X-rays) at a dose according to the applied voltage.

[0017] The generator 21 and the radiation source 23 generate radiation in a manner corresponding to the type of radiographic image (for example, a still image or a moving image). Specifically, in the case of a still image, the generator 21 and the radiation source 23 irradiate radiation once per pressing of the exposure switch 22. In the case of a moving image, for example, the generator 21 and the radiation source 23 irradiate pulsed radiation multiple times per predetermined time per pressing of the exposure switch 22.

[0018] The imaging unit 24 generates digital image data showing the imaging region of the subject. The imaging unit 24 may be, for example, a portable FPD.

[0019] In addition to this configuration, the radiation imaging apparatus 20 has a display device 25. The display device 25 serves as an output device for outputting guidance for guiding the movement of the subject when capturing a radiation image.

[0020] The radiation imaging device 20 is connected to the radiation imaging control device 10 via a communication cable. The radiation imaging control device 10 controls the imaging operation of the radiation imaging device 20 by controlling the generator 21. The radiation imaging control device 10 also performs image analysis of the radiation image obtained by the imaging unit 24. The radiation imaging device 20 may be installed in an imaging room, or may be configured to be mobile by being incorporated into a medical cart or the like.

[0021] FIG. 2 is a block diagram showing the configuration of the main parts of the radiation imaging control device 10 according to this embodiment.

[0022] As shown in FIG. 2, the radiography control device 10 is also called a console and is configured, for example, by a personal computer. The radiography control device 10 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. In the radiography control device 10, the CPU reads a program corresponding to the processing content from the ROM and loads it into the RAM. The radiography control device 10 then cooperates with the loaded program to centrally control the operation of each unit. The radiography control device 10 includes a communication unit 11, a display unit 12, an operation unit 13, a control unit 14, a memory unit 15, a guidance information generation unit 16, etc.

[0023] The communication unit 11 acquires order information input to an external system such as the HIS 32 or the RIS 33, and inputs the order information to the display unit 12. The display unit 12 is, for example, a display device of the radiation imaging control device 10, and displays the order information on a screen.

[0024] The operation unit 13 is a user interface in the radiation imaging control device 10. A user (e.g., a radiologist) controls and operates the radiation imaging device 20 based on the displayed order information via the operation unit 13. The radiation imaging control device 10 may also have a function to automatically control the radiation imaging device 20 based on the order information.

[0025] The control unit 14 controls each block of the radiation imaging control device 10, sets imaging conditions for the radiation imaging device 20, etc., and controls the reading operation of the captured radiation images. The imaging conditions include, for example, subject conditions related to the subject, irradiation conditions related to the irradiation of radiation, and image reading conditions related to the image reading by the imaging unit 24.

[0026] The subject conditions include, for example, the imaging region, imaging direction, physique, etc. The irradiation conditions include, for example, the tube voltage (kV), the tube current (mA), the irradiation time (ms), the current-time product (mAs value), etc. The image reading conditions include, for example, the pixel size, the image size, and the frame rate.

[0027] The storage unit 15 stores a machine learning model 15A used to generate guidance information. The guidance information is information for guiding the movement of a subject in the radiation imaging device 20 when dynamic imaging is performed on the subject, and is generated by the guidance information generation unit 16 using the machine learning model 15A. Details of the guidance information generation unit 16 and the guidance information will be described later.

[0028] The machine learning model 15A is trained based on video data in which a predetermined subject performs a predetermined action, and is a model that receives input case information and outputs a specific frame image from the video data in which the predetermined subject is in a specific situation. The predetermined subject is a subject (second subject) with a case that is the same as or similar to the case of a subject (first subject) who is the subject of imaging by the radiation imaging device 20. The second subject may be the same subject as the first subject, or may be a subject different from the first subject.

[0029] A similar case may be, for example, a case whose type is different from that of the first subject and whose site is the same as that of the first subject. For example, if the first subject's case is osteoarthritis of the elbow, similar cases may include elbow-related diseases other than osteoarthritis of the elbow (e.g., fractures, dislocations, etc.).

[0030] The predetermined motion is a motion for checking a specific condition of the subject by moving a part related to the case, and the motion differs for each case. For example, if the part related to the case is the elbow, the predetermined motion may be a motion of bending and straightening the elbow. For example, if the part related to the case is the shoulder, the predetermined motion may be a motion of raising the arm from a lowered position while maintaining the arm in an extended position.

[0031] The specific situation is, for example, a situation in which the subject feels a problem such as pain or discomfort. The specific situation includes, for example, a situation in which the subject vocalizes due to the problem, a situation in which the subject operates an operating device such as a push button when feeling the problem, and a situation in which a significant change is observed in the subject's facial expression or posture.

[0032] The specific frame image is a frame image when the predetermined subject is in a specific situation, from among the plurality of frame images constituting the above-mentioned moving image data.

[0033] The machine learning model 15A may be trained by, for example, a training device 40 shown in Fig. 3. The training device 40 has a training data storage unit 41, an extraction unit 42, and a training unit 43. The training data storage unit 41 stores video data obtained by dynamic imaging of a predetermined subject performing a predetermined movement, linked to each case type.

[0034] When the predetermined subject is a first subject, the moving image data is data of the first subject performing a predetermined movement in the past, and is stored in the training data storage unit 41 for the number of times the predetermined movement has been performed. When the predetermined subject is a second subject, the moving image data is stored in the training data storage unit 41 for the number of second subjects who performed the predetermined movement.

[0035] The moving image data may be moving image data captured by a predetermined imaging device. The predetermined imaging device may be provided in the training device 40 or may be provided separately from the training device 40. The predetermined imaging device may be the above-mentioned radiation imaging device 20 or may be an imaging device different from the radiation imaging device 20.

[0036] The extraction unit 42 extracts specific frame images from the video data (plurality of frame images) in the training data storage unit 41. Specifically, the extraction unit 42 may extract specific frame images based on predetermined input information.

[0037] The predetermined input information is information input to the predetermined imaging device and is information indicating that the predetermined subject has entered a specific situation. For example, if the predetermined imaging device is equipped with a sound collection device such as a microphone, the input information may be a voice uttered by the predetermined subject due to a malfunction and input to the imaging device via the sound collection device. Furthermore, for example, if the predetermined imaging device is equipped with an operating device such as a push button, the input information may be operation information input to the imaging device by operating the operating device when the predetermined subject experiences a malfunction.

[0038] The extraction unit 42 extracts the frame image when the above-mentioned predetermined input information is input to the imaging device as the specific frame image. For example, if the control unit of the imaging device is configured to send a pulse signal each time a frame image is generated when dynamic imaging is performed, the frame image corresponding to the pulse signal when the input information is input may be set as the specific frame image.

[0039] The extraction unit 42 may also extract a specific frame image based on changes over time in a plurality of frame images. For example, if the facial expression of a specific subject changes relatively significantly between frames, the extraction unit 42 may extract the frame image after the change as the specific frame image.

[0040] The training unit 43 trains the machine learning model 15A based on the video data (plurality of frame images) stored in the training data storage unit 32. Specifically, the training unit 43 inputs case information linked to the video data to the machine learning model 15A, and acquires information on the frame images related to the output results from the machine learning model 15A.

[0041] The training unit 43 adjusts the parameters of the machine learning model 15A according to the error between the frame image related to the output result and the specific frame image extracted by the extraction unit 42. For example, if the machine learning model 15A is realized by a neural network, the training unit 43 may adjust the parameters of the machine learning model 15A according to the backpropagation method according to the error between the frame image related to the output result and the specific frame image.

[0042] The training unit 43 ends the training process when the above-mentioned processing is completed for all video data stored in the training data storage unit 41. The training unit 43 provides the acquired machine learning model 15A to the radiation imaging control device 10 as a trained model.

[0043] The extraction unit 42 may be provided separately from the training device 40. In this case, the specific frame images extracted by the extraction unit 42 may be stored in the training data storage unit 41 in association with case information.

[0044] Next, the guide information generating unit 16 and the guide information will be described in detail.

[0045] As described above, the guide information generating unit 16 generates guide information based on the machine learning model 15A (information related to the case) stored in the storage unit 15.

[0046] Specifically, the guidance information generation unit 16 acquires information about the subject who is the subject of imaging by the radiation imaging device 20. The guidance information generation unit 16 refers to the electronic medical record or the like included in the subject information to extract case information about the subject and inputs the extracted information to the machine learning model 15A. Then, the guidance information generation unit 16 generates guidance information for guiding the movement of the subject so that the subject's condition becomes a specific condition based on the output information (information about a specific frame image) output from the machine learning model 15A.

[0047] The guidance information may be, for example, guidance information displayed on the display device 25 provided in the radiation imaging apparatus 20. The guidance information is information that guides the movement of the subject to a position in a specific situation. The guidance information may be written in a table or the like as information associated with each case.

[0048] For example, suppose the specific situation is one in which the subject is in a state where the elbow is bent, as shown in Fig. 4. Since the subject's arms are initially stretched out to the sides, when dynamic imaging by the radiation imaging device 20 starts, the display device 25 displays the subject in a state where the arms are stretched out to the sides (see the upper diagram in Fig. 4).

[0049] In this case, the guidance information may be text information encouraging the patient to bend the elbow. The guidance information may also be information indicating the position of the arm (the area related to the case) in a specific condition superimposed on the image of the patient (see the dashed line). The guidance information may also be information indicating the direction in which the patient should move the arm superimposed on the image of the patient.

[0050] In this way, it is possible to guide the subject's movements so that the subject's situation becomes a specific situation. In other words, it is possible to encourage the subject to make appropriate movements when capturing a radiographic image. As a result, it is possible for the subject to more easily align their arms with the position for the specific situation in accordance with the guidance information (see the lower diagram in Figure 4). In other words, it is possible to more easily capture an image of a situation in which the subject feels discomfort, and therefore it is possible to accurately grasp the extent of the subject's discomfort.

[0051] Furthermore, the guidance information may be generated, for example, by comparing the current image of the subject with a specific frame image, if the difference between the images is equal to or greater than a predetermined threshold value (which can be set arbitrarily).

[0052] Next, a description will be given of the flow of processing performed by the radiation imaging control device 10 according to this embodiment. Fig. 5 is a flowchart showing an example of the operation of processing performed by the radiation imaging control device 10 according to this embodiment.

[0053] 5, the radiation imaging control device 10 acquires information about the subject (step S101), extracts case information based on the subject information, and inputs the case information to the machine learning model 15A (step S102). Next, the radiation imaging control device 10 acquires the output result of the machine learning model 15A (step S103).

[0054] After acquiring the output result of the machine learning model 15A, the radiation imaging control device 10 generates guidance information (step S104). After generating the guidance information, the radiation imaging control device 10 displays the guidance information on the display device 25 (step S105). Then, this control ends.

[0055] According to the present embodiment configured as described above, guidance information for guiding the movement of the subject is generated based on information relating to the case of the subject, and the guidance information is displayed (notified) to the subject.

[0056] This makes it possible to guide the subject to make appropriate movements when capturing a radiation image.

[0057] Furthermore, the guidance information generating unit 16 generates guidance information based on the machine learning model 15A that is trained based on video data of a predetermined subject performing a predetermined movement. Specifically, the guidance information generating unit 16 generates guidance information such that the subject's situation becomes a specific situation based on the machine learning model 15A.

[0058] This makes it easier for the machine learning model 15A to determine positions that are likely to result in specific situations for each case, making it easier to guide the subject's movement to those positions.

[0059] Furthermore, if the predetermined subject is a subject different from the subject being imaged, the position of the specific situation can be determined based on statistical data according to the case, which makes it easier to determine the position of the specific situation.

[0060] Furthermore, if the predetermined subject is the same subject as the subject being imaged, a position specific to that subject can be set as the position of the specific situation.

[0061] Although one pattern of guidance information is exemplified in the above embodiment, the present invention is not limited to this. The guidance information generating unit 16 may generate multiple patterns of guidance information. In other words, the guidance information generating unit 16 may change the way of moving the guidance target part of the subject depending on the position of the guidance target part of the subject.

[0062] The plurality of patterns of guidance information may be associated with the positions of the guidance target portions and written in a table, as shown in FIG.

[0063] For example, when the part of the subject to be guided reaches a position in a specific situation, the guidance information generating unit 16 may generate guidance information for making the part to be guided stand still.

[0064] For example, if the subject bends his / her elbow and the arm has not yet reached the specific situation (dashed line), the position of the guidance target part is lower than the specific situation position. Therefore, the guidance information is information that prompts the subject to raise the arm, and the information is continuously displayed on the display device 25.

[0065] When the subject reaches the position where the elbow is bent (the specific situation position), the guidance information is switched to information that encourages the subject to maintain that posture. When the subject passes the specific situation position, the position of the guidance target part is to the right of the specific situation position, so the guidance information is switched to information that encourages the subject to bend the arm to the left.

[0066] This allows the subject's movements to be guided accurately.

[0067] Furthermore, for example, the guidance information generating unit 16 may generate guidance information that changes the speed at which the guidance target part is moved depending on the difference between the position of the guidance target part of the subject and the target position of the guidance target part. The target position of the guidance target part may be, for example, the position of the guidance target part related to a specific situation.

[0068] For example, when the subject's guidance target portion reaches the vicinity of the position of the guidance target portion related to the specific situation, the guidance information generating unit 16 generates guidance information to prompt the subject to slow down the speed of moving the guidance target portion, and changes the display mode of the guidance information from the mode before the guidance target portion reached the vicinity of the position of the specific situation.

[0069] This makes it easier to acquire frame images before and after the subject feels the discomfort, making it easier to confirm the circumstances under which the discomfort occurs, and as a result, it becomes easier to accurately grasp the condition of the subject's discomfort.

[0070] Furthermore, in the above embodiment, there is no particular limitation as to whether the predetermined subject is a subject (first subject) to be imaged by the radiation imaging device 20 or a second subject different from the first subject. For example, if the predetermined subject is the first subject, the machine learning model 15A is trained using the past predetermined movements of the first subject. This makes it possible to compare the current movement of the first subject with the past movement of the first subject, thereby enabling the effectiveness of treatment to be confirmed.

[0071] That is, the predetermined subject to be learned in the machine learning model 15A is the subject to be imaged by the radiation imaging apparatus 20, and the case of the predetermined subject may be the same as the case of the subject to be imaged.

[0072] This makes it possible to take time-dependent differences from past imaging results, allowing for accurate assessment of treatment effects, and as a result, accurate future treatment plans can be made.

[0073] In the above embodiment, the guidance information is notified to the subject by displaying it on the display device 25. However, the present invention is not limited to this. For example, the guidance information may be notified to the subject by voice from the voice output unit.

[0074] Furthermore, in the above embodiment, the machine learning model 15A takes case information as input, but the present invention is not limited to this, and may take, for example, information on the subject's target area to be guided (area related to the case) as input.

[0075] In the above embodiment, the guidance information is generated using the machine learning model 15A, but the present invention is not limited to this. For example, the guidance information generating unit 16 may generate guidance information by referring to a table that associates cases with predetermined guidance information (guidance information not based on a machine learning model). The guidance information not based on a machine learning model may be, for example, information that prompts the subject to move a guidance target part to a position in a specific situation that is predetermined by an experiment or the like.

[0076] Furthermore, the above-described embodiments are merely examples of specific embodiments for carrying out the present invention, and the technical scope of the present invention should not be construed as being limited by these embodiments. In other words, the present invention can be carried out in various forms without departing from the gist or main features thereof. [Explanation of symbols]

[0077] 10 Radiography control device 11 Communications Department 12 Display section 13 Control section 14 Control Unit 15 Storage section 15A Machine Learning Models 16 Guidance information generation section 20 Radiography equipment 21 Generator 22 Exposure switch 23 Radiation Source 24 Filming Department 25 Display device 31 PACS 32 HIS 33RIS 40 Training equipment 41 Training data storage unit 42 Extraction part 43 Training Department

Claims

1. An imaging support device that supports a subject to be imaged by a radiographic imaging device capable of dynamic imaging, a guidance information generating unit that generates guidance information for guiding a movement of the subject based on information about the case of the subject; a control unit that notifies the subject of the guidance information; A photography support device comprising:

2. the guidance information generation unit generates the guidance information based on a machine learning model trained on video image data of a predetermined subject performing a predetermined movement. The photography support device according to claim 1 .

3. the machine learning model is a model that receives input of information about the case and outputs a specific frame image of the predetermined subject in a specific situation from the moving image data, the guidance information generation unit generates the guidance information for guiding the movement of the subject so that the situation of the subject becomes the specific situation. The photography support device according to claim 2 .

4. The predetermined subject is a subject different from the subject to be imaged. The photography support device according to claim 2 .

5. the predetermined subject is the same subject as the subject to be imaged, The case of the predetermined subject is the same as the case of the subject to be imaged. The photography support device according to claim 2 .

6. the guidance information generating unit changes a manner of moving the guidance target portion according to a position of the guidance target portion of the subject. The photography support device according to claim 1 .

7. the guidance information generating unit changes a speed of moving the guidance target part of the subject in accordance with a difference between a position of the guidance target part of the subject and a target position of the guidance target part. The photography support device according to claim 6.

8. An imaging support method for supporting a subject to be imaged by a radiographic imaging device capable of dynamic imaging, comprising: generating guidance information for guiding a movement of the subject based on information about the case of the subject; notifying the subject of the guidance information; A photography support method comprising:

9. An imaging support program for supporting a subject to be imaged by a radiographic imaging device capable of dynamic imaging, On the computer, A process of generating guidance information for guiding the movement of the subject based on information about the case of the subject; a process of notifying the subject of the guidance information; Execute Photography support program.

Images