JP2024038298A - Radiography control device, radiography control method, program, and radiography system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a radiography control device capable of suppressing re-taking of a radiation image by detecting foreign matter likely to be reflected in a radiation image and performing an alarm display before taking a radiation image.

SOLUTION: A radiography control device includes: an image acquisition unit for acquiring an image capturing a subject; a foreign matter detection unit for detecting foreign matter likely to be reflected in a radiation image when acquiring the radiation image of the subject from the image acquired by the image acquisition unit; an alarm information generation unit for generating alarm information on the capturing of the radiation image on the basis of the detection of foreign matter in the captured image by the foreign matter detection unit; and a display control unit for executing notification based on the alarm information.

SELECTED DRAWING: Figure 3

COPYRIGHT: (C)2024,JPO&INPIT

Description

The present invention relates to a radiography control device, a radiography control method, a program, and a radiography system that suppress re-imaging of radiographic images.

2. Description of the Related Art Radiography systems that use radiation are known as imaging systems in the medical field. The radiography system irradiates the radiation imaging device with radiation emitted by the radiation generating device via the target patient, and the radiation imaging device generates a radiation image, allowing the user to check the radiation image immediately after radiography. It becomes possible.

Patent Document 1 discloses a radiation diagnostic apparatus that suppresses re-imaging by radiography. An optical camera is attached to the radiation generating device, and images taken by the optical camera are displayed on the screen of the radiation diagnostic device. Then, the positioning image obtained when the user positions the target patient's posture is compared with the captured image taken by the optical camera, and if the target patient's posture does not match between the images, the user is notified. , it is possible to suppress the occurrence of re-imaging of radiographic images due to body movements of the target patient.

JP2011-24721A

However, foreign objects such as clothing buttons or necklaces that appear in the radiographic image may also require retaking the radiographic image. In recent years, from the viewpoint of the patient's privacy, users may be required to take radiographic images while the patient is clothed, or when taking radiographic images of a large number of patients during group medical examinations, etc. Sometimes it is difficult to pay attention to foreign objects reflected in the image. Patent Document 1 does not disclose a configuration for suppressing re-imaging of a radiographic image due to a foreign object appearing in the radiographic image.

The present invention has been developed in view of the above problems, and detects foreign objects that may appear in radiographic images and displays a warning before radiographic images are taken, thereby preventing the need for re-imaging radiographic images. It is an object of the present invention to provide a radiography control device that suppresses radiation exposure.

In order to solve the above problems, the radiography control device according to the present invention includes:
An image acquisition unit that acquires a photographed image of the subject; and a foreign object that may appear in the radiographic image of the subject when acquiring the radiographic image of the subject from the photographed image acquired by the image acquisition unit. a foreign object detection unit; and a warning information generation unit that generates warning information regarding radiographic imaging based on the detection of a foreign object in the captured image by the foreign object detection unit;
and a display control unit that provides notification based on the warning information.

According to the present invention, by detecting a foreign object that may appear in a radiographic image and displaying a warning before radiographic image capturing, it is possible to suppress the occurrence of re-capture of a radiographic image.

System configuration diagram of a radiography system according to Embodiment 1 Hardware configuration diagram of a radiography control device according to Embodiment 1 Software configuration diagram of the radiography control device according to Embodiment 1 Flowchart diagram of the radiography control device according to Embodiment 1 Schematic diagram of a warning screen of the radiography control device according to Embodiment 1 Software configuration diagram of a radiography control device according to Embodiment 2 Flowchart diagram of the radiography control device according to Embodiment 2

Embodiments for carrying out the present invention will be described below with reference to the drawings. Note that the present invention is not limited to the embodiments disclosed below, and various modifications and changes can be made without departing from the scope of the claims.

A foreign object that may appear in a radiographic image used below refers to an object that may appear in a radiographic image obtained by radiography. Objects that may appear in radiographic images include, for example, plastic items such as glasses and buttons, metal items such as belts and zippers, prints on T-shirts, compresses, and body warmers. Note that foreign objects that may appear in radiographic images are not limited to these objects.

[Embodiment 1]
First, the configuration of the radiographic system of this embodiment will be explained using FIGS. 1 to 3.

FIG. 1 shows an example of the overall configuration of the radiation imaging system of this embodiment. The radiography system includes a radiography control device 100 , a radiography device 110 , a radiation generation device 120 , and an image acquisition device 130 via a network 140 . Note that the network 140 may be a wired network or a wireless network.

The radiography control apparatus 100 is an apparatus constructed from an information processing device such as a computer, which communicates with the radiography apparatus 110 via the network 140 and controls radiography. The radiography control device 100 also communicates with the radiation generation device 120 via the network 140 and acquires information when radiation is irradiated from the radiation generation device 120. The radiography control device 100 further communicates with the image acquisition device 130, controls the image acquisition device 130, and acquires images captured by the image acquisition device 130. In addition, a part of each structure of the radiography control apparatus 100 may be comprised by the same apparatus.

The radiographic apparatus 110 transitions to a radiographic image capture enabled state in which it can capture a radiographic image in response to an instruction from the radiographic control apparatus 100. The radiographic device 110 is a device that performs radiographic imaging in synchronization with the radiation generating device 120 and generates a radiographic image based on the radiation emitted from the radiation generating device 120. Note that the number of radiographic apparatuses 110 is not limited to one, and a configuration using a plurality of radiographic apparatuses 110 may be used.

The radiation generating device 120 is a device that receives a radiation irradiation instruction from an exposure switch 121 and generates radiation from a tube 122 based on irradiation conditions set by an input device that accepts user operations such as an operation panel. .

The image acquisition device 130 is a device that photographs a subject according to instructions from the radiography control device 100 and acquires a photographed image. In this embodiment, an optical camera is used as the image acquisition device 130, and images taken by the optical camera are acquired. In this embodiment, the image acquisition device 130 will be described as being attached to the tube 122 and taking images in the radiation generation direction of the tube 122. That is, the imaging range of the photographed image is characterized in that it has at least an overlapping area with the imaging range of the radiation image. However, the position where the image acquisition device 130 is installed is not limited to the tube 122, and there is no restriction on the installation position as long as it is a position where a foreign object in the subject can be photographed. For example, an indoor camera may be used as the image acquisition device 130 and installed on the ceiling of a photography room. FIG. 2 is an example of the hardware configuration of the radiography control device 100 that constitutes the radiography system of this embodiment.

The radiography control apparatus 100 includes a network device 201 that connects to a network 140 and an input device 202 that receives user operations such as a keyboard. Furthermore, the radiography control apparatus 100 includes an operation screen such as a liquid crystal display, a display device 203 that displays radiation images, and a CPU (Central Processing Unit) 204 that controls the entire apparatus. The radiography control device 100 also stores a RAM (Random Access Memory) 205 that provides a work space for the CPU 204, various control programs, radiographic images received from the radiography device 110, captured images received from the image acquisition device 130, and the like. The storage device 206 has a storage device 206 that stores data. The devices constituting the radiography control device 100 are connected via a main bus 207, and can mutually send and receive data.

Note that although the input device 202 and the display device 203 are described as separate devices, these devices may be integrated, for example, a device equipped with a touch panel.

FIG. 3 is an example of the software configuration of the radiography control device 100 that constitutes the radiography system of this embodiment.

Each functional unit shown in FIG. 3 is realized by the CPU 204 on the radiography control apparatus 100 reading out a control program stored in the storage device 206 onto the RAM 205 and executing it.

The radiography control device 100 includes a communication section 301, a system control section 302, an image acquisition section 303, a radiation image processing section 304, a display control section 305, a foreign object detection section 306, and a warning information generation section 309.

The communication unit 301 is software that controls the network device 201 and performs communication.

The system control unit 302 controls the image acquisition device 130, acquires irradiation information of the radiation generating device 120 and imaging information of the radiation imaging device 110, and manages each state via the communication unit 301. Further, the system control unit 302 is a program that realizes the basic functions of the radiography control apparatus 100, and controls the operation of each unit.

The image acquisition unit 303 acquires a radiographic image from the radiation imaging apparatus 110 and a captured image from the image acquisition apparatus 130 via the communication unit 301.

The radiation image processing unit 304 processes the radiation image acquired via the system control unit 302 and generates an image to be used on the radiation imaging control apparatus 100.

The display control unit 305 displays the image generated by the radiation image processing unit 304 via the display device 203.

Further, the display control unit 305 displays a notification based on the warning information generated by the warning information generation unit 309 via the display device 203. Here, the notification based on the warning information performed by the display control unit 305 may take any form, such as a voice message or an alert sound, as long as it can prompt the user to confirm.

The display control unit 305 also performs processes such as reflecting a process instructed by the system control unit 302 on an image based on an operation from the input device 202 and switching the screen display of the display device 203.

The foreign object detection unit 306 uses the captured image acquired by the image acquisition unit 303 from the image acquisition device 130 to detect a foreign object in the captured image.

The foreign object detection section 306 includes a foreign object identification section 307 and a feature amount storage section 308. When the foreign object identification unit 307 identifies a foreign object that may appear in a radiographic image based on machine learning, the feature amount storage unit 308 contains information in advance so that the foreign object identification unit 307 can identify the foreign object. It is assumed that feature amounts for identifying foreign objects are held from the photographed images that have undergone learning. A foreign object identification unit 307 recognizes a foreign object in the photographed image using the feature quantity for identifying the foreign substance stored in the feature quantity storage unit 308. That is, the foreign object detection section 306 is characterized by having a foreign object identification section 307 that identifies foreign objects using a machine learning algorithm.

Note that there is no limit to the specific method used for machine learning in this embodiment; for example, ResNet may be used as a CNN (Convolutional Neural Network) architecture for identification by the foreign object identification unit 307, or multiple machine learning techniques may be used. Identification may be performed using a combination of the following. Further, the configuration of the foreign object detection unit 306 is not limited as long as it can detect foreign objects from the captured image using the captured image obtained from the image acquisition device 130 by the image acquisition unit 303.

Here, a configuration in which a CNN is used as a discriminator in the foreign object identifying section 307 in the foreign object detecting section 306 will be described. A classifier based on machine learning, such as a CNN, consists of a learning phase in which the classifier is trained, and an inference phase in which inference is performed using the learned classifier.

The configuration of a discriminator using CNN will be described below. In the learning phase, the foreign object identification unit 307 performs learning of a discriminator for identifying foreign objects that may appear in radiographic images. CNN is classified as supervised learning among machine learning technologies. In supervised learning, learning is performed using teacher data that is a pair of a learning image for learning and a correct image showing information on a correct area on the learning image. In this specification, the correct image corresponds to each foreign object that may appear in a radiographic image, and each foreign object is referred to as a class. The output of the classifier for identifying classes is broadly divided into extraction for identifying pixels of the class (foreign object) identified from the image, and classification for identifying the class of foreign objects present on the image. Furthermore, by using a Softmax function in the output layer, the output from the classifier can be obtained as a likelihood. That is, the foreign object identification unit 307 is characterized in that it outputs the identification result as a likelihood.

In the present invention, the model and network structure described here are merely examples, and the classifier used in the foreign object identifying section 307 may be configured using other machine learning techniques or rule-based techniques.

When the learning of the classifier in the learning phase is completed, the foreign object identification unit 307 stores the learned feature amount for foreign object detection in the feature amount storage unit 308. The learning phase of the classifier constituting the foreign object identification unit 307 may be terminated by a predetermined number of times set by the user, or the learning termination condition may be determined by a determination of overlearning, early stopping, or the like. The classifier may be made to learn a plurality of classes corresponding to each of a plurality of foreign objects, or each classifier may be made to learn each foreign object. When a plurality of classifiers are trained, the feature quantity storage unit 308 stores feature quantities for a plurality of foreign object identifications.

In the inference phase, the abnormality identification unit 307 identifies foreign objects using the feature quantities stored in the feature quantity storage unit 308. Note that both phases may be executed by different devices, and the device configuration does not matter as long as the feature amount for foreign object identification generated in the learning phase can be acquired when inference is made in the inference phase.

When the output result from the classifier in the foreign object identification unit 307 is the likelihood, the foreign object detection unit 306 determines whether a foreign object is present on the captured image acquired by the image acquisition unit 303, for example, by threshold processing. . In other words, the likelihood can be rephrased as the confidence in the classification result of the machine learning-based classifier in the foreign object identification unit 307. Therefore, the foreign object identification unit 306 sets a threshold value in advance, and detects a foreign object corresponding to a class equal to or higher than the predetermined threshold value as a foreign object existing on the captured image. That is, the foreign object detection section 306 is characterized in that it detects foreign objects having a likelihood equal to or higher than a threshold value from among the output results from the foreign object identification section 307 .

When the foreign object detection section 306 detects a foreign object, the warning information generation section 309 generates warning information indicating that a foreign object has been detected. Thereafter, the warning information generation unit 309 instructs the display control unit 305 to display the generated warning information on the screen. Details are described in the configuration diagram of the warning screen of the warning information generation unit 309.

That is, the radiography control apparatus 100 includes an image acquisition unit 303 that acquires a photographed image of a target patient as a subject, and a foreign object that may appear in the radiographic image from the photographed image acquired by the image acquisition unit 303. It has a foreign object detection section 306 that detects. The radiography control device 100 also includes a warning information generation unit 309 that generates warning information regarding radiographic imaging based on the detected foreign body when a foreign body is detected by the foreign body detection unit 306; The present invention is characterized by having a display control unit 305 that displays a notification based on the warning information generated by 309.

In this embodiment, a display processing method of the radiography control apparatus 100 during imaging of a target patient will be described with reference to FIG. 4, and a configuration diagram regarding the display of a warning image will be described with reference to FIG.

FIG. 4 is a flowchart of the notification display process of the radiography control apparatus 100.

In step S401, the system control unit 302 puts the radiography control apparatus 100 into an examination start state based on the user's operation. Specifically, the system control unit 302 transmits an instruction to prepare for imaging to the radiation imaging apparatus 110 via the communication unit 301 based on the imaging conditions of the target patient instructed by the user's operation. When the radiography apparatus 110 completes its own radiography preparation, it sends a return preparation completion notification to the radiography control apparatus 100. After the radiography control apparatus 100 receives the preparation completion notification, the system control unit 302 puts the radiography control apparatus 100 into an examination start state and accepts step S406, which will be described later. The system control unit 302 also transmits an instruction to the image acquisition device 130 to start imaging via the communication unit 301. After receiving the instruction to start imaging, the image acquisition device 130 sequentially transmits the captured images acquired by itself to the radiation imaging control device 100.

The processing between steps S403 and S405 is executed by the system control unit 302 until step S406 is executed or it is decided to cancel the test by a user operation.

In step S402, the image acquisition unit 303 displays the captured image acquired from the image acquisition device 130 via the communication unit 301 on the display device 203 via the display control unit 305.

In step S403, the foreign object detection unit 306 detects a foreign object in the photographed image based on the photographed image acquired via the image acquisition unit 303.

If the foreign object detection unit 306 detects a foreign object in the captured image in step S403, the warning information generation unit 309 generates warning information in step S404.

In step S405, the display control unit 305 notifies the display device 203 based on the warning information.

If no foreign object is detected by the foreign object detection unit 306 in step S403, the radiography control apparatus 100 does not generate warning information. That is, the radiography control apparatus 100 does not execute the processes of step S404 and step S405.

In step S406, the user presses the exposure switch 121 of the radiation generating device 120 to start imaging. When radiographic image capturing is started, the radiation generating device 120 generates radiation from the tube 122, the radiation that has passed through the target patient is notified to the radiographic device 110, and the radiographic device generates a radiographic image. Thereafter, the radiography apparatus 110 transmits the radiography image to the radiography control apparatus 100. Further, in parallel with the processing, the radiation generating device 120 transmits radiation irradiation information regarding the radiography to the radiography control device 100.

FIG. 5 is a schematic diagram of notifications displayed by the radiography control apparatus 100 of this embodiment.

The display screen 500 displayed by the display device 203 is a screen that includes the photographed image acquired from the image acquisition device 130 and displayed on the display device 203 in step S402. Note that objects within the imaging range of the image acquisition device 130, such as the presence of the radiation imaging device 110 behind the target patient, are reflected in the actual captured image, but for simplicity, the display screen 500 shows the target patient. Use a diagram that expresses only physical information. Similarly, for the photographed images and images related to the photographed images displayed on the display screen 500 thereafter, unless otherwise specified, diagrams expressing only the physical information of the target patient are used.

A notification 501 in FIG. 5 is a notification based on warning information indicating the presence of a foreign object generated by the warning information generation unit 309 in step S404. In FIG. 5A, when the foreign object detection unit 306 detects a foreign object, a warning text is displayed as a notification 501 on the captured image 500 to prompt the user to confirm. However, there is no limit to the configuration as long as the presence of a foreign object can be presented to the user. For example, as in the configuration shown in FIG. 5B, a notification 501 and foreign object detection information 502 indicating an area on the photographed image determined to be a foreign object may be displayed. The user can immediately understand that a foreign object has been detected and the position of the foreign object on the display screen 500. Further, on the display screen 500, as in the configuration shown in FIG. 5C, the display control unit 304 may display warning information 501 and foreign object content information 503 representing foreign object information on the photographed image determined to be a foreign object. Furthermore, by combining the foreign object detection information 502 and the foreign object content information 503, the foreign object information 503 allows the user to grasp the foreign object information that the user should check, and the foreign object detection information 502 allows the user to understand the foreign object information regarding the position of the foreign object. It becomes possible to check for foreign objects.

That is, the warning information generating section 309 is characterized in that it generates warning information indicating the foreign object detected by the foreign object detecting section 306 and the position of the foreign object.

Note that the display methods described above may be combined, or only a part of the configuration may be displayed on the display screen 500.

As described above, in the radiography control apparatus 100 according to the first embodiment, the foreign object detection unit 305 detects foreign objects that may be reflected in the radiography image, with respect to the captured image acquired before acquiring the radiographic image. The display control unit 305 issues a notification on the display device 203 based on the warning. This allows the user to notice the foreign object, which leads to suppressing the occurrence of re-imaging of radiographic images, and as a result, it becomes possible to suppress ineffective exposure of the target patient.

(Modification 1)
Here, processing when there is a plurality of foreign objects detected by the foreign object detection unit 306 will be described using FIG. 4. In step S403, the foreign object detection unit 306 detects a plurality of foreign objects. The foreign object identification section 307 in the foreign object detection section 306 may identify foreign objects as a plurality of different classifier classes, or may identify a plurality of foreign objects as each of the same classifier class.

When the discriminator in the foreign object identification unit 307 identifies foreign objects in different image areas on the photographed image, the foreign object detection unit 306 performs threshold processing on the likelihood corresponding to each foreign object by the foreign object identification unit 307. is executed, and a foreign object corresponding to a class having a likelihood greater than or equal to a threshold value is detected as a foreign object. In addition, when the foreign object identification unit 307 detects multiple foreign objects in an image area that includes the same image area on the photographed image, the likelihood of the classes corresponding to each of the multiple foreign objects is compared, and the likelihood is calculated. A foreign object corresponding to the highest class is detected as a foreign object. Alternatively, if the area of the same image area on the captured image is below a certain level, or if the ratio of the image area showing a foreign object to the detected foreign object area is below a certain level, the discriminator in the foreign object identification unit 307 identifies multiple foreign objects. The identification results may be detected as foreign objects.

In step S404, the warning information generation unit 309 generates the notification 501 so that each foreign object corresponding to the plurality of pieces of foreign object information acquired from the foreign object detection unit 306 can be identified. For example, if the number of foreign objects detected by the foreign object detection unit 306 makes it difficult to display notifications based on the respective warning information on the same display screen 500, the warning information generation unit 309 In response, a notification 501 is generated and a switching display is performed. Alternatively, if the detected foreign object can be displayed on the display screen 500, the warning information generation unit 309 generates warning information so as to display a plurality of pieces of foreign object information as warning information on the same display screen. When displaying a plurality of foreign object information on the same display screen by the display control unit 305, processing such as changing the display density or color is performed so that each foreign object can be identified from the other. You may do so.

In step S405, the display control unit 305 displays the warning information acquired from the warning information generation unit 309 on the display screen 500. The display control unit 305 displays each of the plurality of foreign objects in a distinguishable manner. Further, the display by the display control unit 305 allows each of a plurality of foreign objects to be displayed in a switchable manner, and a plurality of pieces of foreign object information are displayed on the same display screen. The display by the display control unit 305 is not limited to this, and for example, foreign objects to be confirmed may be displayed in a list format. Further, if there are multiple foreign objects, the display control unit 305 may display a checklist (item) for inputting whether or not the user has confirmed the check.

According to this modification, even when multiple foreign objects are detected by the foreign object detection unit 306, the user can grasp the presence or absence of the foreign object, the position of the foreign object, and information about the foreign object, and eliminate invalid exposure due to re-imaging of radiation images. can be suppressed.

(Modification 2)
In the first embodiment, a configuration was described in which the image acquisition device 130 includes an optical camera, and the foreign object detection unit 306 detects foreign objects using an optical image as the acquired image. . However, the foreign object detection unit 306 may detect foreign objects by using a near-infrared camera in the image acquisition device 130 and a near-infrared image as the acquired photographic image, or may detect foreign objects using a photographic image photographed by an optical camera and a near-infrared image. Foreign objects may be detected by combining the image with an image taken by an infrared camera. That is, the photographed image acquired by the image acquisition unit 303 is a photographed image photographed by at least one of the optical camera and the near-infrared camera.

Below, as a method for detecting a foreign object using a plurality of captured images, a configuration will be described in which the foreign object identification unit 307 in the foreign object detection unit 306 identifies a foreign object using a CNN-based classifier.

When the foreign object identification unit 307 in the foreign object detection unit 306 performs identification based on machine learning such as CNN, training data is required as described above. Up to the above, the foreign object identification unit 307 uses, as training data, a pair of an image captured by an optical camera and a correct image showing information on a correct region corresponding to a class indicating the foreign object as a correct image. The discriminator was trained in , and the feature quantities for identifying foreign objects generated by the learning were stored in the feature quantity storage unit 308 . When a captured image is composed of multiple images captured by an optical camera or a near-infrared camera, for example, like the training data for the discriminator explained above, the image captured by the optical camera and the correct area are combined. A discriminator is trained using training data that is a pair of correct images showing information, and further trained using training data that is a pair of an image taken by a near-infrared camera and a correct image showing information on the correct area. There is a method of learning the classifier separately. The classifier can learn different features from training data that includes images captured by an optical camera and training data that includes images captured by a near-infrared camera, and based on the results of foreign object identification by each classifier, By comparing the results, it is possible to prevent failure to detect foreign objects and ensure consistency in foreign object identification.

Another method is to use a photographed image taken by an optical camera and a photographed image taken by a near-infrared camera as learning images, and have the discriminator learn them as correct images corresponding to the correct answer area. By using an image taken by an optical camera and an image taken by a near-infrared camera as learning images at the same time, it becomes possible to generate feature amounts based on the relationship between the images.

Although a near-infrared camera has been described here as an example of an infrared camera, the wavelength band acquired by the infrared camera is not limited to near-infrared only.

Note that the information given to the foreign object detection unit 306 is not limited to only the above-mentioned information, and may include information other than the captured image, such as an image captured by a metal detector, etc., that can obtain image information that cannot be obtained by an optical camera. It may also be configured to input.

(Modification 3)
In Modification 2, a configuration has been described in which a photographed image taken by an optical camera or a near-infrared camera is input, and a foreign object is detected by the foreign object detection unit 306. In this modification, a configuration will be described in which at least one of the captured image to be detected by the foreign object detection unit 306 or the feature amount of the foreign object discriminator is adjusted according to the condition of the target patient, which is the subject. do.

For example, when radiographic images are taken, the state of the imaging range may differ depending on the target patient. For example, it is conceivable that the difficulty level of detecting a foreign object from the photographed image and the feature amount for detecting the foreign object are different between photographed images taken with clothes on and with clothes not worn. Therefore, in this modification, by inputting or determining the condition of the target patient by the user, the input to the foreign object detection section 306 that detects foreign objects and the feature amount by the foreign object identification section 307 in the foreign object detection section 306 are determined. , adjust at least one side. When a user is wearing clothes, it is more difficult to detect a foreign object from an image taken by an optical camera than when the user is not wearing clothes. If the user is wearing clothes, at least an image taken by a near-infrared camera is input to the foreign object detection unit 306. Furthermore, the feature amount that the foreign object discriminator 307 in the foreign object detection section 306 acquires from the feature amount storage section 308 also selects a feature amount generated by learning training data that includes at least an image taken by a near-infrared camera.

According to this modification, the foreign object detection unit 306 can perform foreign object detection adapted to the condition of the target patient, and can suppress foreign objects from appearing in the image.

[Embodiment 2]
Next, a second embodiment of the present invention will be described.

In the configuration of the first embodiment, the foreign object detection unit 306 detects foreign objects that may appear in the radiographic image from the captured image acquired by the image acquisition unit 303. However, even if the user is wearing an object that is detected as a foreign object in the captured image, it may be desirable not to detect it as a foreign object and not to notify the user, as long as it is not visible in the radiation image. .

Generally, the radiation emitted from the radiation generator 120 is irradiated onto a limited area on the subject using a collimator or the like, and this area is called an irradiation field. If the irradiation field is the target patient's chest, for example, even if a compress applied to the arm is visible in the captured image, it will not be visible in the radiation image, so the compress applied to the arm will be detected as a foreign object. It is preferable not to do so and not to notify. This is because if a foreign object is detected in an area that is not visible in the radiation image and a warning is displayed to the user, the user must check for the foreign object even though the object does not affect the radiographic image capture. This is because there is a possibility that the effort of the work will increase.

Therefore, in the configuration of the second embodiment, the generation of irradiation field information by the radiography control apparatus 100 and the processing of foreign object detection by the foreign object detection unit 306 are added. Hereinafter, only the differences from Embodiment 1 will be explained using FIGS. 6 and 7.
FIG. 6 is an example of the software configuration of the radiography control apparatus 100. The radiography control apparatus 100 according to the present embodiment includes an irradiation field information generation section 600 in addition to the configuration described in the first embodiment. Here, the irradiation field information generation unit 600 generates irradiation field information representing positional information of the irradiation field area within the photographed image. In this embodiment, the irradiation field information generation unit 600 generates irradiation field information by analyzing changes in brightness in a photographed image obtained from the image acquisition device 130 when the irradiation field lamp of the tube 122 is turned on. do. Specifically, after the irradiation field lamp of the tube 122 is turned off, the radiography control device continues to hold the irradiation field information immediately before the lamp was turned off in the storage device 206.

That is, the radiation imaging apparatus 100 includes an irradiation field information generation unit 600 that detects the position of the irradiation field by the radiation generating device from the captured image and generates irradiation field information, and the foreign object detection unit 306 detects the position of the irradiation field by the radiation generating device from the captured image and generates irradiation field information. The method is characterized in that a foreign object is detected from an image area corresponding to the irradiation field information generated by the irradiation field information generation unit 600 among the acquired photographic images.

Note that there is no restriction on the specific method as long as the positional information of the irradiation field area can be acquired by the irradiation field information generation unit 600. For example, the aperture information of the tube 122 can be acquired from the radiation generator 120 in advance and the irradiation field area can be calculated. or may be calculated based on machine learning such as CNN.

FIG. 7 is a flowchart showing the process of displaying warning information of the radiography control apparatus 100 of this embodiment.

In step S701, the irradiation field information generation unit 600 generates irradiation field information indicating an irradiation field region from the photographed image through the image analysis described above.

In step S702, the foreign object detection unit 306 performs foreign object detection processing on the image area on the captured image corresponding to the irradiation field area based on the irradiation field information from the captured image acquired by the image acquisition unit 303. Unlike step S403 described in the first embodiment, when a foreign object is detected in the image area corresponding to the irradiation field area on the captured image based on the irradiation field information by the foreign object detection in step S702, the subsequent step The processes of S404 and S405 are executed.

As described above, in the second embodiment, in the radiography control apparatus 100, the irradiation field information generation section 600 generates irradiation field information from the captured image acquired by the image acquisition section 303 at the start of the examination, and the foreign object detection section 306 generates irradiation field information. A warning is displayed on the display device 203 as to whether or not there is a foreign object that may appear in the radiographic image in the image area corresponding to the irradiation field area in the field information. This eliminates the need to display a warning even if a foreign object that may appear in a radiographic image is detected from an area that does not affect the imaging, making it possible to prevent the increase in user effort due to confirmation work. .

(Other examples)
Further, the present invention can also be realized by executing the following processing. That is, the software (program) that realizes the functions of the embodiments described above is supplied to a system or device via a network or various storage media, and the computer (or CPU, MPU, etc.) of the system or device reads the program. This is the process to be executed.

100 Radiography control device 110 Radiography device 120 Radiation generation device 130 Image acquisition device 301 Communication unit 302 System control unit 303 Image acquisition unit 304 Radiographic image processing unit 305 Display control unit 306 Foreign object detection unit 309 Warning information generation unit

In order to solve the above problems, the radiography control device according to the present invention includes:
An image acquisition unit that acquires a photographed image of the subject taken by at least one of an optical camera and an infrared camera ; and an image acquisition unit that acquires a radiographic image of the subject from the photographed image acquired by the image acquisition unit; a foreign object detection unit that detects a foreign object that may be reflected in the image; a warning information generation unit that generates warning information regarding radiographic imaging based on the detection of a foreign object in the captured image by the foreign object detection unit;
a display control unit that makes a notification based on the warning information ,
Acquisition of the photographed image by the image acquisition unit is performed before acquisition of the radiation image .

Claims (14)

an image acquisition unit that acquires a captured image of the subject;
a foreign object detection unit that detects a foreign object that may appear in the radiation image when acquiring a radiation image of the subject from the captured image acquired by the image acquisition unit;
a warning information generation unit that generates warning information regarding radiographic imaging based on the detection of a foreign object in the captured image by the foreign object detection unit;
A radiography control device comprising: a display control unit that provides notification based on the warning information. The radiography control device according to claim 1, wherein the photographed image acquired by the image acquisition unit is a photographed image photographed by at least one of an optical camera and an infrared camera. 3. The radiography control apparatus according to claim 1, wherein the foreign object detection section includes a foreign object identification section that identifies foreign objects from the captured image using a machine learning algorithm. The radiography control apparatus according to claim 3, wherein the foreign object identification section outputs the identification result as a likelihood. The radiography control apparatus according to claim 4, wherein the foreign object detection section detects a foreign object having a likelihood equal to or higher than a threshold value from among the output results from the foreign object identification section. The radiography control apparatus according to any one of claims 1 to 5, wherein the warning information generation section generates warning information indicating the foreign object detected by the foreign object detection section and the position of the foreign object. . The radiography control device includes an irradiation field information generation unit that detects the position of an irradiation field by the radiation generating device from the captured image and generates irradiation field information,
The foreign object detection section detects a foreign object from an image area corresponding to the irradiation field information generated by the irradiation field information generation section, of the photographed image acquired by the image acquisition section. 7. The radiography control device according to any one of 1 to 6. 8. If there are a plurality of foreign objects detected by the foreign object detection section, the display control section displays each of the plurality of detected foreign objects in a distinguishable manner. The radiography control device described in . The radiography control device according to claim 8, wherein when there are a plurality of foreign objects detected by the foreign object detection section, the display control section displays each of the plurality of detected foreign objects in a switchable manner. . The radiography according to any one of claims 1 to 9, wherein the display control unit displays an item for inputting whether or not the user has confirmed the foreign object detected by the foreign object detection unit. Control device. an image acquisition step of acquiring a captured image of the subject;
a foreign object detection step of detecting a foreign object that may appear in the radiation image of the subject when acquiring the radiation image from the photographed image acquired in the image acquisition step;
a warning information generation step of generating warning information regarding radiographic imaging based on the detection of a foreign object in the photographed image in the foreign object detection step;
A radiography control method comprising: a display control step of making a notification based on the warning information. A program for causing a computer to execute the radiography control method according to claim 11. A radiography system includes an image acquisition device that photographs a subject and acquires a photographed image, a radiography device that radiographs the subject and acquires a radiographic image, and a radiography control device that controls radiography. hand,
The radiography control device includes:
a foreign object detection unit that detects a foreign object that may appear in a radiation image of the subject when acquiring a radiation image of the subject from the photographed image acquired by the image acquisition device;
a warning information generation unit that generates warning information regarding radiographic imaging based on the fact that the foreign object detection unit detects a foreign object in the captured image;
A radiation imaging system comprising: a display control unit that provides notification based on the warning information. 14. The radiation imaging system according to claim 13, wherein the imaging range of the captured image by the image acquisition device is an imaging range that has at least an overlapping area with the imaging range of the radiation image by the radiation imaging device.

Images