Image detection apparatus, console, and radiographic system

Abstract

The present invention provides an image inspection apparatus, a control console, and a radiographic imaging system that reduces the workload of an image inspection process by automatically and accurately overlaying marks indicating the photographic direction and / or laterality of a subject onto a radiographic image. The image inspection apparatus (10) includes: a radiographic image acquisition unit (11) for acquiring a radiographic image (16) of a subject photographed using radiation; a photographic condition recognition unit (13) for recognizing photographic conditions related to the photographic direction and / or laterality of the subject illuminated in the radiographic image (16); and a mark overlay unit (14) for overlaying marks indicating the photographic direction and / or laterality of the subject illuminated in the radiographic image (16) onto the radiographic image (16) using the recognition result of the photographic condition recognition unit (13).

Description

Technical Field

[0001] This invention relates to an image detection device, a control console, and a radiographic system. Background Technology

[0002] In the medical field, the use of radiographic images of subjects obtained through X-rays and other forms of radiation for diagnosis is becoming increasingly common. In radiography, a process called image inspection (QA, or Quality Assurance) is typically performed to confirm whether the obtained radiographic images are suitable for diagnosis. This image inspection process includes, for example, determining imaging errors (i.e., whether a re-enhancing image is necessary), adjusting density and contrast, adjusting the angle of the subject in the radiographic image, trimming parts relevant to diagnosis, and overlapping markings indicating the imaging direction and / or laterality of the subject in the radiographic image.

[0003] In recent years, devices have been known to automatically determine shooting errors using preview images with reduced image quality of radiation images (Patent Document 1). Furthermore, devices have been known to automatically adjust contrast by automatically determining the window level (WL) and window width (WW) (Patent Document 2).

[0004] Previous technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2013-102851

[0007] Patent Document 2: Japanese Patent Application Publication No. 8-96125 Summary of the Invention

[0008] The technical problem to be solved by the invention

[0009] The image inspection process is required for all captured radiographic images, thus placing a significant workload on radiology engineers and physicians performing radiography. Therefore, it is desirable to automate the image inspection process to reduce the workload of radiology engineers and similar professionals.

[0010] In particular, sometimes the markings indicating the photographic direction and / or lateral orientation of the subject cannot be directly determined from the radiographic image that captured its content. In such cases, it is necessary to compare the captured radiographic image with the menus or commands related to the capture, and then input the markings, which results in a heavy workload.

[0011] Therefore, the object of the present invention is to provide an image inspection apparatus, control console, and radiographic system that reduces the workload of the image inspection process by automatically and accurately overlaying marks indicating the photographic direction and / or laterality of the subject onto the radiographic image.

[0012] means for solving technical problems

[0013] The present invention is an image detection device with a processor. The processor acquires a radiation image of a subject taken using radiation, identifies the photographic conditions related to the photographic direction and / or laterality of the subject illuminated in the radiation image, and uses the identification results to overlay a mark indicating the photographic direction and / or laterality of the subject illuminated in the radiation image onto the radiation image.

[0014] The processor preferably retrieves the photography menu related to the capture of the radiation image, and uses the photography menu and the radiation image to identify the photography conditions.

[0015] The processor preferably acquires camera images of the subject using a different method than radiography before obtaining radiographic images, and uses the camera images to identify photographic conditions.

[0016] The processor preferably uses the radiographic image to identify the location in the radiographic image where the marker should be overlapped, and overlaps the marker on top of or moves it to the identified location.

[0017] The processor preferably performs a shooting error determination on radiographic images to determine whether they need to be reshot. For radiographic images that are determined not to need to be reshot in the shooting error determination, the processor uses the recognition results to overlap the markings.

[0018] The processor preferably uses the recognition results to overlay the radiographic images with adjusted concentration and / or contrast when adjusting the concentration and / or contrast of the radiographic images.

[0019] The processor preferably uses the recognition results to overlay the radiographic image with the adjusted angle of the subject when adjusting the angle of the subject in the radiographic image.

[0020] The processor preferably uses the recognition results to overlap the processed radiographic image when performing trimming processing on a portion of the radiographic image.

[0021] The processor preferably displays the history of the image detection process, including overlapping processing of markers.

[0022] The processor preferably receives a correction command for at least a portion of the image detection process including overlapping processing of markers. When a correction command is received, in addition to correcting the image detection process that received the correction command, it automatically re-executes the image detection process performed after at least the image detection process that received the correction command, according to the result of the image detection process that received the correction command.

[0023] Furthermore, this invention is a control console for controlling a radiography system and includes an image detection device. The radiography system includes a radiation generating unit that generates radiation and a radiography unit that uses radiation to photograph a subject. This invention is a radiography system and includes a control console.

[0024] Furthermore, the present invention is a radiography system, comprising: a radiation generating unit that generates radiation; a radiography unit that uses radiation to photograph a subject; and a processor that performs the following processing: identifying photographic conditions related to the photographic direction and / or laterality of the subject illuminated in the radiography image obtained using the radiography unit, and using the identification result, overlaying a mark indicating the photographic direction and / or laterality of the subject illuminated in the radiography image onto the radiography image.

[0025] Invention Effects

[0026] The image inspection apparatus, control console, and radiographic system of the present invention can reduce the workload of the image inspection process by automatically and accurately overlaying markings indicating the photographic direction and / or laterality of the subject onto the radiographic image. Attached Figure Description

[0027] Figure 1 This is an explanatory diagram illustrating the structure of a radiographic imaging system.

[0028] Figure 2 It is a block diagram representing the functionality of the console.

[0029] Figure 3 This is an explanatory diagram illustrating the photography commands and photography menu.

[0030] Figure 4 This is a block diagram representing the function of an image detection device equipped with a photographic condition recognition unit.

[0031] Figure 5 This is an explanatory diagram illustrating the photographic direction (P→A) of the subject.

[0032] Figure 6 This is an explanatory diagram illustrating the photographic direction (A→P) of the subject.

[0033] Figure 7 This is an explanatory diagram illustrating the side profile (left hand) of the subject.

[0034] Figure 8 This is an explanatory diagram illustrating the lateral aspect (right hand) of the subject.

[0035] Figure 9 This is a block diagram representing the function of an image detection device equipped with a photography menu acquisition unit.

[0036] Figure 10 This is a block diagram representing the functions of an image detection device equipped with a camera image acquisition unit.

[0037] Figure 11 This is an explanatory diagram illustrating the overlapping positions of the markings in Example 41.

[0038] Figure 12 This is an explanatory diagram illustrating the overlapping positions of the markings in Example 42.

[0039] Figure 13 This is an explanatory diagram illustrating the overlapping positions of the markings in Example 43.

[0040] Figure 14 This is an explanatory diagram illustrating the overlapping positions of the markings in Example 44.

[0041] Figure 15 This is a block diagram representing the function of the photographic condition recognition unit.

[0042] Figure 16 This is an explanatory diagram illustrating the change in the overlapping position of the markings.

[0043] Figure 17 It is a block diagram representing the functions of an image detection device that has various image detection processes.

[0044] Figure 18 This is an explanatory diagram illustrating the process of adjusting the angle of the subject.

[0045] Figure 19 This is an explanatory diagram illustrating the finishing process.

[0046] Figure 20 This is a block diagram representing the functions of an image detection device, including an image detection history display unit.

[0047] Figure 21 This is an explanatory diagram illustrating the display of image detection history.

[0048] Figure 22 This is a flowchart illustrating the process of image detection based on an image detection device. Detailed Implementation

[0049] The image inspection apparatus of the present invention is used in the image inspection process of a subject after taking a radiographic image using radiation. The image inspection process is performed, for example, by a radiation engineer who has taken the image. The image inspection apparatus can be installed, for example, in the image examination room of a radiology department, or it can be installed in a location outside the radiology department.

[0050] like Figure 1 As shown, the image detection device 10 constitutes the radiography system 20. The radiography system 20 includes a radiation generating unit, i.e., a radiation source 21, a radiography unit 22, a camera 23, a control console 24, and the image detection device 10, etc.

[0051] The console 24 is the main control device (so-called computer) of the radiography system 20, such as a personal computer or workstation with an application program installed to perform the prescribed functions. The image detection device 10 is also, for example, a personal computer or workstation with an application program installed to perform the prescribed functions. In this embodiment, the computer of the console 24 also performs the functions of the image detection device 10. In this case, the console 24 includes the image detection device 10. Furthermore, as in this embodiment, the image detection device 10 can be a computer common to the computer of the console 24, or it can be a computer other than the console 24, and the method is not limited. Therefore, the image detection device 10 can be a device included in other devices, or it can be a single device.

[0052] Radiation source 21 generates radiation Ra used in radiography. In this embodiment, radiation source 21 is an X-ray source that generates X-rays. Therefore, radiography system 20 is an X-ray radiography system that acquires an X-ray image of a subject Obj by using X-ray imaging. The subject Obj is, for example, a human being.

[0053] The radiography unit 22 uses radiation Ra generated by the radiation source 21 to image the subject Obj. The radiography unit 22 includes a radiation detector, such as an FPD (Flat Panel Detector). The FPD detects the radiation Ra transmitted through the subject Obj and converts it into an electrical signal, thereby outputting a radiographic image of the subject Obj. In imaging using the radiography unit 22, a grid (not shown) can be used as needed. The grid is a device for removing the scattered radiation component, such as a fixed Lysholm grating or a movable planar aperture. In this embodiment, the radiography unit 22 includes a radiation detector that outputs one radiographic image per exposure to radiation Ra.

[0054] The radiation detector included in the radiography unit 22 can be either an indirect conversion type radiation detector or a direct conversion type radiation detector. An indirect conversion type radiation detector refers to a detector that uses a scintillator made of CsI (cesium iodide) or similar material to convert radiation Ra into visible light, and indirectly obtains an electrical signal by photoelectric conversion of that visible light. A direct conversion type radiation detector refers to a detector that uses a scintillator made of amorphous selenium or similar material to directly convert radiation Ra into an electrical signal. Furthermore, the radiation detector included in the radiography unit 22 can be either a PSS (Penetration Side Sampling) type radiation detector or an ISS (Irradiation Side Sampling) type radiation detector. The PSS type refers to a method in which the scintillator is positioned on the subject's obj side relative to the TFT (Thin Film Transistor) that performs the electrical signal readout. The ISS method refers to the opposite of the PSS method, in which the scintillator and TFT are arranged from the subject (Obj) side in the order of TFT and scintillator.

[0055] Camera 23 uses visible light or infrared light (light with a wavelength or energy distribution different from radiation Ra) to photograph the subject Obj positioned on the radiography unit 22. More specifically, camera 23 is, for example, a digital camera or a digital video camera. Furthermore, the photographic range SR of camera 23 at least includes the irradiation range of radiation Ra. In the radiography system 20, the images captured by camera 23 (a collection of still images, i.e., moving images; hereinafter referred to as camera images) are used for the identification of the orientation and / or laterality of the subject Obj in radiography. The camera images, etc., will be explained later.

[0056] The control console 24 is the main control unit (so-called computer) of the radiography system 20, for example, controlling the radiography system 20, or communicating with the RIS (Radiology Information System) 31, HIS (Hospital Information System) 32, or other external systems. The control console 24 receives imaging commands from the RIS 31 or HIS 32, and acquires the radiographic images output from the radiography unit 22 and sends them to the respective units.

[0057] like Figure 2As shown, the control console 24 includes a photography menu setting unit 25, an operation unit 26, and an image detection device 10. The photography menu setting unit 25 allows users to manually input or obtain photography commands from RIS31, HIS32, or other external systems. Furthermore, it sets the photography menu corresponding to the obtained photography commands. Figure 3 As shown, the radiographic command 33 is a request for radiographic imaging, which includes information such as "command ID" to identify the command, "subject ID" to identify the subject Obj (subject Obj is the identification number of the examinee, etc.), and "radiography menu" to identify the radiographic location and direction of the subject Obj.

[0058] The photography menu represents a specific photography item and is set according to the photography command. For example, when the photography command is "commission to photograph one frontal view of the chest (P→A) and one frontal view of the chest (A→P) of a specific subject Obj", the photography menu setting unit 25 sets "Frontal View of Chest (P→A)" and "Frontal View of Chest (A→P)" as the photography menu for that specific subject Obj. "Frontal View of Chest (P→A)" is a menu that photographs the chest of the subject Obj from the front by irradiating radiation Ra from the rear surface (Posterior) towards the front surface (Anterior). Similarly, "Frontal View of Chest (A→P)" is a menu that photographs the chest of the subject Obj from the front by irradiating radiation Ra from the front surface towards the rear surface. Figure 3 In the "Photography Commands", the "Command ID" is "OD0001", the "Subject ID" is "H500", and the "Photography Menu" is "Chest Front / Standing Pose / P→A".

[0059] The operation unit 26 includes, for example, a keyboard and / or a pointing device used for inputting settings such as photographic conditions, and for operating the radiation source 21 and the radiographic unit 22. The operation unit 26 can be configured as a touch panel. Furthermore, by operating the operation unit 26, it is possible to set or change the photographic menu.

[0060] The image detection device 10 can have communication capabilities and can communicate with the camera menu setting unit 25 or operation unit 26 of the control console 24 or with external devices. Therefore, the camera menu setting unit 25 or operation unit 26 of the control console 24 or external devices and the image detection device 10 can send and receive data.

[0061] like Figure 4As shown, the image detection apparatus 10 includes a radiation image acquisition unit 11 and an image detection processing unit 12. In the image detection apparatus 10, programs related to the radiation image acquisition unit 11 and the image detection processing unit 12 are programmed into a memory (not shown). The program is executed by a control unit (not shown) composed of a processor, thereby realizing the functions of the radiation image acquisition unit 11 and the image detection processing unit 12. The image detection processing unit 12 includes a photographic condition recognition unit 13 and a mark overlay unit 14. The radiation image acquisition unit 11 acquires a radiation image 16 via a photographic menu setting unit 25 included in the control console 24. The photographic condition recognition unit 13 identifies photographic conditions related to the photographic direction and / or laterality of the subject illuminated in the radiation image 16. The mark overlay unit 14 uses the recognition results of the photographic condition recognition unit 13, etc., to overlay marks indicating the photographic direction and / or laterality of the subject illuminated in the radiation image 16 onto the radiation image 16. The radiation image 16 with the overlaid marks is sent to an image server 15, etc.

[0062] The following is a detailed description of each part of the image detection apparatus 10. The radiographic image acquisition unit 11 acquires, for example, the radiographic image 16 output by the radiographic imaging unit 22 via the imaging menu setting unit 25. Here, the acquired radiographic image 16 may sometimes be a medical image suitable for diagnosis, or, for various reasons, not suitable for diagnosis. Furthermore, there are cases where it is difficult to determine the imaging direction and / or lateral orientation of the subject. Therefore, in order to make it a medical image suitable for diagnosis, an image detection process is performed on the radiographic image 16. The image detection process may include multiple steps.

[0063] In this embodiment, the image detection process includes five steps: determining a shooting error that indicates a failed shot, adjusting density and contrast, adjusting the angle of the subject in the radiographic image, trimming parts related to cropping and diagnosis, and overlapping markers indicating the photographic direction and / or laterality of the subject in the radiographic image. A radiographic image 16, having completed the image detection process, is provided for diagnostic purposes.

[0064] The radiation image 16 acquired by the radiation image acquisition unit 11 is sent to the imaging condition recognition unit 13 of the image detection processing unit 12. To automate the process of overlapping markers indicating the photographic direction and / or laterality of the subject illuminated in the radiation image 16 during the image detection process, the imaging condition recognition unit 13 identifies the imaging conditions (hereinafter referred to as directional imaging conditions) related to the photographic direction and / or laterality of the subject illuminated in the radiation image 16. The directional imaging conditions are the conditions for overlapping markers indicating the photographic direction and / or laterality of the subject illuminated in the radiation image 16 in the radiation image 16. Based on the identification result of the imaging condition recognition unit 13, which identifies which directional imaging condition of the subject illuminated in the radiation image 16 is present, the marker overlap unit 14 overlaps the markers in the radiation image 16.

[0065] The imaging direction of the subject in radiographic image 16 refers to the orientation of the subject when it is positioned relative to the radiographic unit 22. When the subject is a person, this refers to the patient's orientation or patient position. The lateral orientation of the subject in radiographic image 16 refers to whether the right or left side of the subject is being photographed. Furthermore, patient orientation is described in the DICOM (Digital Imaging and Communications in Medicine) standard ("Explanation of Patient Orientation in Annex A").

[0066] Regarding directional photography conditions, for example, Figure 5 and Figure 6 As shown, the subject is a person. Figure 5 and Figure 6 A radiation source 21 is present on the front side of the paper, and a radiographic unit 22 is present in the depth direction of the paper. When the front of the chest of the subject Obj is photographed from a standing position, in Figure 5 In the case of example 41 shown, the anterior surface of the subject Obj is positioned facing the radiographic unit 22 in a "frontal chest (P→A)" orientation. As the patient's position, the lower side of the paper is designated "F" (Foot), and the right side is designated "R" (Right). On the other hand, in Figure 6 In the case of example 42 shown, the subject Obj is positioned with his back facing the radiographic unit 22 in a "frontal chest (A→P)" orientation. As for the patient's position, the direction below the paper is "F" (Foot), and the direction to the right of the paper is "L" (Left). As described above, in the directional imaging conditions of the subject in the radiographic image 16 of the frontal chest image, Figure 5 In the case of example 41 shown, the shooting direction is "P→A". Figure 6 In the case of example 42 shown, the shooting direction is "A→P".

[0067] And, as Figure 7 and Figure 8 As shown, for example, the subject is a person, in Figure 7 and Figure 8 There is a radiation source 21 on the front side of the paper, and a radiation imaging unit 22 is located in the depth direction of the paper. When a human hand is photographed as the subject (Obj), Figure 7 In the case of example 43 shown, the left hand is positioned with the palm facing the radiographic unit 22 in an inward rotation position. Figure 8 In Example 44, the right hand is positioned with the palm facing the radiographic unit 22 in an inward rotation position. Therefore, in Examples 43 and 44, the subject Obj is photographed from the front by irradiating it with radiation Ra from the rear surface to the front surface, and thus the imaging direction is "P→A". Furthermore, in Figure 7 In the case of example 43 shown, the patient's orientation is "H" (Head) on the bottom of the paper and "L" (Left) on the right side of the paper. On the other hand, in Figure 8 In the case of example 44 shown, the patient's orientation is "H" (Head) on the bottom of the paper and "L" on the right side of the paper. As described above, regarding the directional imaging conditions of the subject in the radiographic image 16, in example 43, the imaging direction is "P→A" and the lateral orientation is left or left hand, while in example 44, the imaging direction is "P→A" and the lateral orientation is right or right hand.

[0068] As for the method by which the photographic condition recognition unit 13 identifies the directional photographic conditions of the subject illuminated in the radiation image 16, any method capable of identifying the directional photographic conditions of the subject illuminated in the radiation image 16 can be used. For example, a method that uses correspondence information to pre-establish a correspondence between the radiation image 16 and the directional photographic conditions of the subject illuminated in the radiation image 16 can be used. That is, the directional photographic conditions of the radiation image 16 acquired by the radiation image acquisition unit 11 are inferred using the radiation image 16 and the corresponding information, and the inferred directional photographic conditions are set as the recognition result of the directional photographic conditions of the radiation image 16. In the inference, known image analysis techniques, image recognition techniques, or image processing techniques can be used, specifically, for example, methods for extracting feature points through image processing of the radiation image 16 or machine learning-based methods can be used.

[0069] Furthermore, the camera menu related to the acquisition of the radiographic image 16 can be obtained, and the camera menu and the radiographic image 16 can be used to identify directional camera conditions. For example... Figure 9As shown, the image detection device 10 may include a photography menu acquisition unit 17. The photography menu acquisition unit 17 acquires a photography menu 17a related to the capture of the radiographic image 16 via a photography menu setting unit 25 included in the control console 24 of the radiographic system 20 connected to the image detection device 10. Then, it acquires the photographic direction or lateral aspect of the subject included in the photography menu 17a. At this time, directional photographic conditions are identified from the subject captured in the radiographic image 16, and based on a comparison with the photographic direction or lateral aspect of the subject included in the photography menu 17a, directional photographic conditions for the radiographic image 16 are set. For example, when the photography menu 17a is "Chest Front / Standing Posture / P→A", and the directional photographic conditions for the subject captured in the radiographic image 16 are identified as "P→A" (see reference...). Figure 3 The photography condition recognition unit 13 identifies the photography direction as "P→A" in the direction photography conditions.

[0070] Furthermore, before obtaining radiographic images 16, camera images of the subject Obj can be acquired using a different method than radiographic photography, and these camera images can be used to identify directional photographic conditions. For example... Figure 10 As shown, the image detection device 10 may include a camera image acquisition unit 18 for acquiring camera image 18a. The radiography system 20 includes a camera 23 for acquiring camera image 18a. The camera 23 is controlled by a control console 24. The camera image acquisition unit 18 of the image detection device 10 acquires the camera image 18a acquired by the camera 23. The photography condition recognition unit 13 uses the acquired camera image 18a to identify the directional photography conditions of the subject Obj.

[0071] In this embodiment, camera 23 is a digital video camera, and it uses visible light to photograph the subject Obj. The directional photographic conditions of the subject are identified using camera image 18a, so camera image 18a includes part or all of the subject Obj to the extent that this identification process is feasible. The configuration of camera 23 can be arbitrary as long as it is within the range where the directional photographic conditions of the subject can be identified using camera image 18a, but in this embodiment, it is set substantially integrally with radiation source 21. This is because the subject Obj is positioned within the irradiation range of radiation Ra, and the subject Obj is reliably photographed to the extent that the aforementioned identification process is performed, without over- or under-photographing.

[0072] As a method for the photography condition recognition unit 13 to recognize the directional photography conditions of the subject Obj using camera image 18a, any method capable of recognizing the directional photography conditions of the subject illuminated in camera image 18a can be used, and any known method can be employed. For example, a method using correspondence information that pre-establishes a correspondence between camera image 18a and the directional photography conditions of the subject illuminated in camera image 18a can be used. That is, the directional photography conditions of the acquired camera image 18a can be inferred using the correspondence information that pre-establishes a correspondence between camera image 18a and the directional photography conditions of the subject illuminated in the radiation image 16, and the inferred directional photography conditions can be set as the recognition result of the directional photography conditions of camera image 18a. In comparisons of camera image 18a, methods such as known image analysis techniques, image recognition techniques, or image processing techniques can be used; more specifically, methods for extracting feature points through image processing of camera image 18a or machine learning-based methods can be employed.

[0073] Furthermore, regarding the acquisition of camera image 18a, it is sufficient to obtain the directional photography conditions. Therefore, it can be done not only before radiography, but also during or after radiography. Preferably, the acquisition of radiographic image 16 and camera image 18a should not be staggered in time, so as to avoid the directional photography conditions of the subject in radiographic image 16 being different from those of the subject in camera image 18a. However, as long as the directional photography conditions can be obtained, the strictness of the timing is not important.

[0074] The identification of the directional photographic conditions of the subject illuminated in the radiation image 16 can be achieved by combining multiple results obtained through the methods described above to set the final identification result. For example, the final identification result can be obtained by comparing the results of image analysis based on radiation image 16, the results of image analysis based on camera image 18a, and the results based on the photography menu 17a. By combining multiple identification mechanisms, the identification result of the directional photographic conditions of the subject illuminated in radiation image 16 can be obtained more accurately.

[0075] The overlapping marker 14 uses the imaging condition recognition unit 13 to identify the recognition result of the directional imaging conditions, and overlays the marker representing the directional imaging conditions of the subject Obj photographed in the radiographic image 16 onto the radiographic image 16. The overlapping marker uses conventionally used markers representing the directional imaging conditions of the subject in the radiographic image 16. For example, it may be "A→P" or "AP", "P→A" or "PA", "standing", "supine" or "lateral", "R" or "L" or "right hand" or "left hand", etc.

[0076] The position of the overlapping markers on the radiographic image 16 (hereinafter referred to as the marker overlap position) is preferred if it is a position that makes it easy for a doctor to identify the markers when making a diagnosis using the radiographic image 16 and will not cause problems during the diagnosis. Therefore, the marker overlap position can be preset at any of the four corners of the radiographic image 16, or it can be determined for each radiographic image 16. Furthermore, there can be one or more overlapping markers. When there are multiple markers, the marker overlap position can be the same position or multiple positions.

[0077] For example, such as Figure 11 As shown, in Example 41, the overlapping mark 14, based on the recognition result of the photographic condition recognition unit 13, overlaps marks indicating the photographic direction and / or lateral orientation of the subject photographed in the radiographic image 51, namely "R, P→A" and "standing posture," at the upper left end 55 and upper right end 56, respectively, towards the radiographic image 51. Furthermore, for example, as... Figure 12 As shown in Example 42, in the radiographic image 52 obtained in Example 42, similarly, markings indicating the photographic direction and / or laterality of the subject photographed in the radiographic image 52, namely "R, A→P" and "standing posture", are superimposed on the upper left end 55 and the upper right end 56, respectively.

[0078] And, for example, such as Figure 13 As shown, in Example 43, in the radiographic image 53 acquired in Example 43, similarly, markings indicating the photographic direction and / or laterality of the subject in the radiographic image 53, namely "P→A," and "L" indicating the left hand, are superimposed on the upper left end 55 and lower left end 57, respectively. Furthermore, for example, as... Figure 14 As shown in Example 44, in the radiographic image 54 obtained in Example 44, similarly, the markings “P→A” indicating the photographic direction and / or laterality of the subject photographed in the radiographic image 54 and “R” indicating the right hand are superimposed on the upper left end 55 and the lower left end 57, respectively.

[0079] When overlapping is performed according to the position of the markers determined for each radiographic image 16, etc., such as Figure 15 As shown, the photographic condition recognition unit 13 may include a position recognition unit 61. The position recognition unit 61 uses the radiographic image 16 to identify the position where the mark overlap portion 14 should overlap with the mark in the radiographic image 16. The mark overlap portion 14 overlaps with or moves the mark at the position identified by the position recognition unit 61.

[0080] The position recognition unit 61 preferably identifies positions on the radiographic image 16 that will not cause problems for doctors when using it for diagnosis and that doctors will not miss. Therefore, the position recognition unit 61 can identify various positions based on the subject or other objects shown in the radiographic image 16. For example, the positions identified by the position recognition unit 61 may be the four corners of the radiographic image 16 or portions of the radiographic image 16 where the subject is not shown.

[0081] As for the method by which the position recognition unit 61 uses the radiation image 16 to identify the position of the overlapping marker, any known method can be used as long as it can identify the subject photographed in the radiation image 16. For example, a method that pre-establishes a corresponding association between the photography menu 17a, the radiation image 16, and the position of the overlapping marker can be used. That is, the position of the overlapping marker is estimated using the radiation image 16 and the corresponding information, and the estimated overlapping position is set as the overlapping position of the marker identified by the position recognition unit 61. Known image analysis techniques, image recognition techniques, or image processing techniques can be used in the estimation. Specifically, for example, methods for extracting feature points through image processing of the radiation image 16 or machine learning-based methods can be used.

[0082] As a machine learning-based method, a learning model is generated, and this learning model can be used as corresponding information. For example, after using the radiographic image 16 and corresponding information to infer the position of the overlapping marker, the inference result is evaluated, and evaluation information is assigned to the inferred position of the overlapping marker. Then, a learning model is generated based on corresponding information that establishes a corresponding association between the photographic menu 17a (with a certain amount of evaluation information assigned), the radiographic image 16, and the position of the overlapping marker. Furthermore, the radiographic image 16 in this corresponding information is preferably a radiographic image 16 that is not a shooting error.

[0083] By pre-establishing corresponding information linking the photography menu 17a, the radiation image 16, and the positions to be overlapped, and then generating a learning model and setting it as the corresponding information, for example, in the radiation image 16, instead of simply setting the parts of the subject Obj that are not photographed as the overlapped marking positions, the markings can be overlapped at appropriate overlapped marking positions according to each photography menu 17a. Furthermore, in the method of the position recognition unit 61 using the radiation image 16 to identify the positions to be overlapped, in addition to the radiation image 16, the photography menu 17a and camera image 18a can also be used.

[0084] like Figure 16As shown, when using the radiographic image 16 acquired in Example 41, and identifying the upper right end 56 and lower left end 57 as the positions of the overlapping marks 14, the radiographic image 58 becomes a radiographic image with the directional imaging condition "R, P→A" overlapping at the lower left end 57 and the directional imaging condition "standing posture" overlapping at the upper right end 56. Furthermore, in the radiographic image 51 where the marks have been overlapped by the overlapping marks 14, when the position recognition unit 61 identifies the position, the marks already overlapped on the radiographic image 51 can be moved to perform overlap, depending on the situation. For example, in Example 41, if the radiographic image 51 (see reference...) Figure 11 When there are overlapping marks on the upper left end 55 and the upper right end 56, after the position recognition unit 61 recognizes the lower left end 57 as the mark overlap position, the mark overlap unit 14 moves the mark on the upper left end 55 from the upper left end 55 to the position of the lower left end 57.

[0085] As described above, according to the image inspection device or radiographic system, markings indicating the photographic direction and / or laterality of the subject can be automatically overlaid on the radiographic image. Furthermore, during the image inspection process of overlaying the markings, human errors are suppressed, and the markings can be accurately overlaid on the radiographic image 16 using the recognition results from the photographic condition recognition unit 13. Therefore, the workload of the image inspection process can be reduced.

[0086] Additionally, as an image detection process, a shooting error determination can be performed to determine whether a radiation image was captured incorrectly or needs to be re-captured. When a shooting error determination unit is provided to determine whether a radiation image needs to be re-captured, the photography condition recognition unit identifies the photography conditions of radiation images that are determined not to need to be re-captured in the shooting error determination, and the mark overlap unit overlaps the mark in radiation images that are determined not to need to be re-captured in the shooting error determination.

[0087] like Figure 17 As shown, when the image detection processing unit 12 includes a shooting error determination unit 71 that determines shooting errors in the radiation image 16, the overlapping marking unit 14 uses the recognition result based on the photography condition recognition unit 13 to mark the overlapping of the radiation image 16 that is determined not to be re-shot in the shooting error determination. The photography condition recognition unit 13 identifies the photography conditions for the radiation image 16 that is determined not to be re-shot in the shooting error determination before or after the acquisition of the radiation image 16.

[0088] In the acquisition of radiographic images, sometimes imaging failures (referred to as imaging errors) occur due to factors such as incorrect patient positioning, patient movement, insufficient breathing, incorrect imaging conditions, or foreign object detection. Known image analysis techniques, image recognition techniques, or image processing techniques can be used as methods for determining imaging errors. In this embodiment, for example, a learned model based on previously acquired radiographic images 51 is used for determination. By using the learned model, determination can be made based on a benchmark determined through learning. Furthermore, the determination result can be obtained in a short time.

[0089] Once the model is learned, it can utilize algorithms or libraries that provide good judgment results for image processing. Furthermore, algorithms or libraries that provide good judgment results for radiographic image 51 can also be constructed and used. Additionally, as training data, data pre-assigned information about whether previously acquired radiographic images 51 were captured by errors can be used. Furthermore, data that assigns any one of the accompanying information related to the radiographic image 51, such as photographic data or patient data, to the radiographic image 51 can also be used. Moreover, data that selects feature quantities based on the type of radiographic image 51 and assigns feature quantity information to the radiographic image 51 can also be used.

[0090] As long as a judgment can be made according to the prescribed criteria, other well-known machine learning techniques or image processing techniques other than machine learning techniques can be used in addition to the learned model. Furthermore, multiple image processing techniques other than the learned model and machine learning techniques can be used, and preferred techniques can be selected based on the type of location of the radiographic image 51 or the accuracy of the judgment result. The judgment criteria can be preset. For example, the criteria can be set strictly or loosely depending on the purpose of the radiographic image 51. More specifically, for example, when judging whether the radiographic image 51 was successfully captured relative to the imaging menu, a threshold is preset for the deviation in the depiction of the key parts. By further reducing this threshold, the criteria can be set strictly; on the other hand, by further increasing this threshold, the criteria can be set loosely. Therefore, the desired judgment criteria can be set according to the medical institution. Furthermore, even within the same medical institution, the judgment criteria can be set differently for each department, such as emergency medicine, internal medicine, or surgery; or even within the same internal medicine department, for each imaging area; or for training purposes such as judging imaging errors by operators. Furthermore, the setting of the judgment criteria can also be changed.

[0091] Furthermore, as an image detection process, steps such as adjusting the density and / or contrast of the radiation image can be performed. When an adjustment unit for adjusting the density and / or contrast of the radiation image is provided, the imaging condition recognition unit recognizes the imaging conditions of the radiation image with adjusted density and / or contrast, and the overlapping marking unit overlaps and marks the radiation image with adjusted density and / or contrast.

[0092] like Figure 17 As shown, when the radiation image 16 is equipped with an adjustment unit 72 for adjusting the concentration and / or contrast, the overlapping marking unit 14 marks the radiation image 16 with adjusted concentration and / or contrast by using the recognition result based on the photography condition recognition unit 13. Furthermore, the photography condition recognition unit 13 identifies the photography conditions of the radiation image 16 with adjusted concentration and / or contrast before or after acquisition.

[0093] The density and / or contrast can be adjusted to the set density and / or contrast using known image analysis techniques, image recognition techniques, or image processing techniques, such as using a known transformation function on the radiographic image 16. Furthermore, the density and / or contrast can be adjusted to values ​​according to each setting of the subject Obj photographed in the radiographic image 16, the photography menu, or other information.

[0094] Furthermore, as an image detection process, a process for adjusting the angle of the subject illuminated in the radiation image can be performed. When a subject angle adjustment unit is provided that adjusts the angle of the subject within the radiation image, the photography condition recognition unit recognizes the photography conditions of the radiation image with the subject angle adjusted, and the mark overlap unit overlaps the mark in the radiation image with the subject angle adjusted.

[0095] like Figure 17 As shown, when the subject angle adjustment unit 73 is provided to adjust the angle of the subject within the radiographic image 16, the overlapping mark unit 14 marks the radiographic image 16 with the subject angle adjusted by using the recognition result based on the photography condition recognition unit 13. The photography condition recognition unit 13 recognizes the photography conditions of the radiographic image 16 with the subject angle adjusted before or after the acquisition of the radiographic image 16.

[0096] The process of adjusting the angle of the subject in the radiographic image 16 is a process of rotating the radiographic image 16 at any angle. Thus, for example, in the radiographic image 16 where a specific part is set as the subject according to the patient's condition at the time of the imaging, even if the radiographic image 16 is taken from a different imaging direction than usual, by adjusting the angle of the subject, it is possible to set the radiographic image 16 of the subject to be photographed in a direction that is easy for the doctor to examine.

[0097] The process of adjusting the angle of the subject Obj illuminated in the radiation image 16 can be accomplished using known image analysis techniques, image recognition techniques, or image processing techniques. Based on the identification of the subject illuminated in the radiation image 16, the radiation image 16 is rotated at a specific angle to make it the appropriate orientation for the subject.

[0098] Furthermore, for example, the angle of the subject illuminated in the radiation image 16 can be adjusted by using the camera image 18a and identifying the positional relationship between the subject Obj and the radiation image acquisition unit 11, i.e., the sensor panel, using known image recognition technology. Four cases can be cited as examples of the positional relationship between the subject Obj and the sensor panel: First, both the sensor panel and the subject Obj are in a normal orientation. In this case, the subject Obj is illuminated in the radiation image 16 with a normal orientation, so no angle adjustment process is performed. Second, the sensor panel is not in a normal orientation, but the subject Obj is in a normal orientation. In this case, the radiation image 16 is corrected so that the sensor panel is in a normal orientation. Thus, the angle adjustment process of the subject Obj can be performed first, using the correctly oriented sensor panel as a reference. Third, the sensor panel is in a normal orientation, but the subject Obj is not in a normal orientation. In this case, the angle of the subject Obj can be adjusted using the sensor panel as a reference. Fourth, a combination of cases 2 and 3. That is, the sensor panel and / or the subject Obj are not normal. In this case, for example, a reference can be set to the normal orientation of the sensor panel, and the angle of the subject in the radiographic image 16 can be adjusted accordingly. Furthermore, normal orientation refers to the orientation conventionally used in the radiographic image 16 that illuminates the subject Obj. Orientation refers to not only the orientation in the two-dimensional direction relative to the longitudinal or lateral direction of the radiographic image 16, but also the three-dimensional direction relative to the depth direction of the radiographic image 16. Therefore, it also includes cases where the subject Obj is illuminated obliquely towards the depth direction of the radiographic image 16.

[0099] Furthermore, the process of adjusting the angle of the subject in the radiographic image 16 is preferably performed using machine learning techniques. That is, by using pre-established correspondence information relating the shape of the sensor panel in the camera image 18a, the normal orientation of the sensor panel, and the radiographic image 16, the angle of the subject (Obj) is adjusted by comparing the shape of the sensor panel in the radiographic image 16 with the corresponding information, aligning the subject (Obj) in the radiographic image 16 to the normal orientation of the sensor panel. Moreover, the information of the radiographic image 16 of a specific part in the acquisition command can also be used as the correspondence information. By creating a learned model using this correspondence information, the angle of the subject can be automatically adjusted in the radiographic image 16 according to the acquisition command.

[0100] like Figure 18 As shown, as an example of the process of adjusting the angle of the subject in a radiographic image, for example, when the radiographic image 81 before subject angle adjustment (without image detection) is an image of the left hand taken with the fingertips pointing downwards, subject angle adjustment 83, i.e., rotation processing, is automatically performed. As a result, the image 81 before subject angle adjustment becomes the image 82 after subject angle adjustment, which is the orientation used by the doctor during routine examination, i.e., the image 82 after subject angle adjustment of the left hand taken with the fingertips pointing upwards.

[0101] Furthermore, as an image inspection process, a trimming process can be performed on portions of the radiographic image related to cropping and diagnosis. When a trimming unit is included to trim a portion of the radiographic image, the imaging condition recognition unit identifies the imaging conditions of the trimmed radiographic image and marks overlapping portions in the trimmed radiographic image. As a part related to the diagnosis of radiographic images, examples include identifying areas of interest related to diagnosis or removing portions that are unclear in the radiographic image due to lack of X-ray irradiation and therefore unusable for diagnosis.

[0102] like Figure 17 As shown, when the trimming unit 75 is equipped with a trimming unit that performs trimming processing on a portion of the radiation image 16, the overlapping marking unit 14 marks the overlap of the trimmed radiation image 16 using the recognition result based on the photography condition recognition unit 13. The photography condition recognition unit 13 recognizes the photography conditions of the trimmed radiation image 16 before or after the acquisition of the radiation image 16.

[0103] The trimming process can be performed using known image analysis, image recognition, or image processing techniques, but machine learning techniques are preferred. This is because, in image analysis techniques other than machine learning, the method of determining the boundary of the radiation field in the radiation image 16 and setting the trimming box based on the boundary information of the radiation field can lead to misidentification of the radiation field boundary when the boundary of the radiation field is unclear due to the influence of scattered rays, or when the concentration changes drastically due to artificial products in the body.

[0104] One method for retouching using machine learning techniques is to generate a learned model by pre-establishing correspondence information between the photography menu and the radiation image 16, which is not a shooting error. Furthermore, this retouching process does not involve changing the size of the radiation image 16, i.e., enlarging or reducing it. Enlarging or reducing can be done after the retouching process.

[0105] like Figure 19 As shown, as an example of retouching, when the center of a frontal chest image is specified via the photography menu, the specified portion is identified in the radiation image (image 84 before retouching), and retouching is automatically performed (86). As described above, a learned model is generated, and this model is used as corresponding information to infer and determine whether to retouch the image 84 before retouching or which portion to retouch. Therefore, it is possible to more accurately identify the radiation field boundary or the location specified in the photography menu for appropriate retouching. Through retouching (86), the image becomes the radiation image (image 85 after retouching) specified in the photography menu.

[0106] In addition, the first image detection step 91 includes determining shooting errors, adjusting density and / or contrast, and adjusting the angle of the subject. Figure 17 First, the first image detection step 91 is performed. Within the first image detection step 91, it is preferable to first determine if a shooting error has occurred. Regardless of the order in which the radiation image 16, which was not determined to be a shooting error in the shooting error determination, is adjusted for density and / or contrast and / or the angle of the subject within the first image detection step 91, one or both may be performed. The retouching process is performed after the first image detection step 91. Furthermore, the overlapping marking process is performed after the retouching process. It is also possible to perform or not perform any of these processes. For example, preferably, as the image detection step, the radiation image 16 is first determined for shooting error, then one or both of the density and / or contrast adjustment and the angle adjustment of the subject are performed in different orders, followed by the retouching process, and finally the overlapping marking process.

[0107] Additionally, the image detection apparatus 10 may include an image detection history display unit that displays the history of an image detection process including overlap processing based on the marker overlap portion 14. For example... Figure 20 As shown, the image inspection apparatus 10 may include an image inspection history display unit 92. The image inspection history display unit 92 displays a series of processes for all image inspection procedures performed on the radiographic image 16 as a history. Therefore, by using the image inspection history display unit 92, for example, even if multiple image inspection procedures are automatically performed on the radiographic image 16, it is possible to know which procedure was performed in which order and how. Furthermore, it is also possible to know the types of image inspection procedures that were not performed on the radiographic image 16.

[0108] Furthermore, the image detection apparatus 10 may include an image re-detection receiving unit 93 that receives a correction command for at least a portion of the image detection process including overlap processing based on the mark overlap portion 14. It may also include an image detection control unit 94 that, when the image re-detection receiving unit 93 receives a correction command, automatically re-executes, based on the result of the image detection process that received the correction command, at least the image detection process performed after the image detection process that received the correction command, in addition to correcting the image detection process that received the correction command.

[0109] The image re-detection receiving unit 93 receives correction commands for at least a portion of the image detection history displayed on the image detection history display unit 92. Since various image detection processes have a priority order, when a correction command is received, the processes up to the one with the correction command are sequentially deactivated. Then, the correction for the image detection process that received the correction command is performed. Finally, the image detection control unit 94 automatically re-executes the image detection process following the one that received the correction command.

[0110] like Figure 21As shown, the image detection history 95 is displayed via the image detection history display unit 92. The image detection history 95 is, for example, a data table displaying the image detection history. The first row displays a main title, "Image Detection History," and a mode display 97, "Automatic Image Re-detection Mode." The mode display 97 displays either "Automatic Image Re-detection Mode" (for automatic image re-detection) or "Manual Mode" (for automatic image re-detection). Next, as "Image Detection Processing," titles such as "(Initial Image)," "Shooting Error," "Density," "Contrast," and "Angle Adjustment" are displayed. Correspondingly, as "Processing Content," content such as "Not a Shooting Error," "No Adjustment," and "Adjustment, 500 / 1" are displayed. Similarly, as the corresponding "Image," thumbnail images of the radiometric image 16 during each image detection process are displayed. "UNDO" buttons for "Maintain" and "Correct" are displayed. The image detection process with the "Maintain" button set to "On" and its color inverted does not perform image re-detection even when automatic image re-detection is being performed. The image detection process that reverses the "Correct" button automatically performs image re-detection, including subsequent image detection processes. Additionally, when "Automatic Image Re-detection Mode" is selected, if the "Density" setting in the image detection processing is "No Adjustment," the density is adjusted by pressing the "Correct" button before processing.

[0111] As described above, the image detection apparatus 10 can automatically perform multiple image detection steps. Furthermore, it is possible to set the process so that none of the multiple image detection steps are performed. Moreover, each image detection step can be set to produce an accurate image detection result using machine learning techniques, etc. Therefore, with the image detection apparatus 10 and the radiography system 20 equipped with the above-described structure, markings indicating the photographic direction and / or laterality of the subject are more accurately superimposed on the radiographic image, for example, preventing the incorrect assignment of markings indicating the photographic direction and / or laterality of the subject in the radiographic image. Furthermore, by automatically superimposing the correct markings onto the radiographic image, the workload of the image detection steps can be significantly reduced.

[0112] Next, refer to Figure 22The flowchart below explains the operation based on the above structure. First, the radiation image acquisition unit 11 acquires the radiation image 16 (step ST100). First, the shooting error determination unit 71 determines whether a shooting error has occurred (step ST110). If a shooting error has occurred (in step ST110, "No"), the radiation image acquisition is performed again. Only if a shooting error has not occurred does the process proceed to the next image detection step (in step ST110, "Yes"). First, the density and / or contrast adjustment unit 72 adjusts the density and / or contrast (step ST120). Sometimes, adjustment is not performed depending on the situation. Next, the subject angle adjustment unit 73 adjusts the subject angle (step ST130). Next, if these image detection steps are completed, the trimming processing unit 75 performs trimming processing (step ST140). Next, the photography condition recognition unit 13 recognizes the photography conditions or acquires the recognition result of the photography conditions obtained before the radiation image acquisition (step ST150). Based on the recognition result, the overlap marking unit 14 overlaps the radiation image 16 that has undergone the image detection processing so far (step ST160). When an image detection history is displayed ("Yes" in step ST170), the image detection history is displayed (step ST180). If corrections are made, the image detection process is automatically corrected ("No" in step ST190). If no corrections are made ("Yes" in step ST190), the process ends. The process also ends when there is no image detection history ("No" in step ST170).

[0113] In the above embodiments, the hardware structure of the processing units that perform various processes, such as the radiation image acquisition unit 11, the photography menu acquisition unit 17, the camera image acquisition unit 18, the shooting error judgment unit 71, the density adjustment unit 72, the subject angle adjustment unit 73, the trimming processing unit 75, the photography condition recognition unit 13, the mark overlap unit 14, the image detection history display unit 92, the image re-detection receiving unit 93, or the image detection control unit 94, is as shown below. These processors include general-purpose processors that execute software (programs) and function as various processing units, such as CPUs (Central Processing Units), GPUs (Graphical Processing Units), FPGAs (Field Programmable Gate Arrays), and other processors whose circuit structures can be modified after manufacturing, i.e., Programmable Logic Devices (PLDs), as well as processors with circuit structures specifically designed for performing various processes, i.e., dedicated electrical circuits.

[0114] A processing unit can be composed of one of these various processors, or it can be composed of a combination of two or more processors of the same or different types (e.g., multiple FPGAs, a combination of CPUs and FPGAs, or a combination of GPUs and CPUs). Furthermore, multiple processing units can also be composed of a single processor. As examples of assembling multiple processing units with a single processor, firstly, as exemplified by computers such as client or server computers, a processor is composed of a combination of one or more CPUs and software, and this processor functions as multiple processing units. Secondly, as exemplified by System-on-Chip (SoC), a processor is used to implement the functions of the entire system containing multiple processing units using a single IC (Integrated Circuit) chip. Thus, various processing units are constructed as hardware structures using one or more of the aforementioned processors.

[0115] Moreover, more specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements. Another aspect of the present invention is an image detection device in which the processor acquires a radiation image of a subject taken using radiation, identifies photographic conditions related to the photographic direction and / or laterality of the subject illuminated in the radiation image, and uses the identification results to overlay a mark indicating the photographic direction and / or laterality of the subject illuminated in the radiation image onto the radiation image.

[0116] This invention is not limited to the embodiments described above. Of course, various structures can be adopted as long as they do not depart from the spirit of this invention. Moreover, in addition to the program, this invention also relates to a storage medium for storing the program.

[0117] Symbol Explanation

[0118] 10-Image detection device, 11-Radiation image acquisition unit, 12-Image detection and processing unit, 13-Photographic condition recognition unit, 14-Mark overlap unit, 15-Image server, 16, 51, 52, 53, 54, 58-Radiation images, 17-Photographic menu acquisition unit, 17a-Photographic menu, 18-Camera image acquisition unit, 18a-Camera image, 20-Radiation radiography system, 21-Radiation source, 22-Radiation radiography unit, 23-Camera, 24-Control console, 25-Photographic menu setting unit, 26-Operation unit, 31-RIS, 32-HIS, 33-Photographic command, 41, 42, 43, 44-Examples, 55-Upper left end, 56-Upper right end, 57-Lower left end, 61-Position recognition unit, 71 - Shooting error judgment unit, 72- Concentration adjustment unit, 73- Subject angle adjustment unit, 74- First image detection and processing unit, 75- Trimming processing unit, 81- Image before subject angle adjustment, 82- Image after subject angle adjustment, 83- Subject angle adjustment, 84- Image before trimming, 85- Image after trimming, 86- Trimming processing, 91- First image detection process, 92- Image detection history display unit, 93- Image re-detection receiving unit, 94- Image detection control unit, 95- Image detection history, 96- Cursor, 97- Mode display, Obj- Subject, Ra- Radiation, SR- Imaging range, F- Foot direction, R- Right direction, H- Head direction, L- Left direction, ST100~ST190- Steps.

Claims

1. An image detection device, comprising a processor, The processor is configured to, To obtain radiation images of a subject taken using radiation. Image detection is performed on the radiation image obtained using a radiography unit, wherein... The image detection includes multiple image detection steps with a priority order, wherein the last image detection step is a marker overlap processing, which includes: Identify photographic conditions related to the photographic orientation and / or laterality of the subject as seen in the radiographic image. Using the recognition results, a marker indicating the photographic direction and / or laterality of the subject as seen in the radiation image is overlaid onto the radiation image. The processor is further configured to, A correction command is received for one of the image detection processes, including the overlapping processing of the markers. When the correction command is received, in addition to correcting the image detection process performed by the processor after receiving the correction command, the image detection process performed at least after the processor received the correction command is automatically executed again according to the result of the image detection process performed by the processor after receiving the correction command.

2. The image detection device according to claim 1, wherein, The processor acquires a photography menu related to the acquisition of the radiation image. The photography conditions are identified using the photography menu and the radiation image.

3. The image detection device according to claim 1, wherein, The processor acquires camera images of the subject taken using a different method than radiography prior to obtaining the radiographic image. The camera images are used to identify the photographic conditions.

4. The image detection apparatus according to any one of claims 1 to 3, wherein, The processor uses the radiation image to identify the locations in the radiation image where the markers should overlap. The mark is superimposed on the identified position or moved to the identified position.

5. The image detection apparatus according to any one of claims 1 to 3, wherein, When the processor performs a shooting error determination on the radiation image to determine whether it needs to be retaken,... The marker is overlaid using the recognition result on the radiation image that is determined not to need to be reshot in the shooting error determination.

6. The image detection apparatus according to any one of claims 1 to 3, wherein, When the processor adjusts the density and / or contrast of the radiation image, The markers are overlaid using the recognition results on the radiation image with adjusted concentration and / or contrast.

7. The image detection apparatus according to any one of claims 1 to 3, wherein, When the processor adjusts the angle of the subject within the radiation image... The markers are overlaid using the recognition results on the radiation image of the subject, which has been adjusted at an angle.

8. The image detection apparatus according to any one of claims 1 to 3, wherein, When the processor performs trimming processing by cutting out a portion of the radiation image, The markers are overlaid using the recognition results on the modified radiation image.

9. The image detection apparatus according to any one of claims 1 to 3, wherein, The processor displays the history of the image detection process, including the overlapping processing of the markers.

10. A control console for controlling a radiography system and comprising an image detection device according to any one of claims 1 to 9, the radiography system comprising a radiation generating unit that generates radiation and a radiography unit that uses the radiation to photograph a subject.

11. A radiographic system comprising: Radiation-generating part, which produces radiation; A radiographic unit that uses said radiation to photograph a subject; and The console as described in claim 10.

12. A radiographic system comprising: Radiation-generating part, which produces radiation; A radiographic unit that uses said radiation to photograph a subject; and processor, The processor is configured to, Image detection is performed on the radiation images obtained using the radiography unit, wherein... The image detection includes multiple image detection steps with a priority order, wherein the last image detection step is a marker overlap processing, which includes: Identify the radiographic direction and / or lateral-related radiographic conditions of the subject as seen in the radiographic image obtained using the radiographic unit. Using the recognition results, a marker indicating the photographic direction and / or laterality of the subject as seen in the radiation image is overlaid onto the radiation image. The processor is further configured to, A correction command is received for one of the image detection processes, including the overlapping processing of the markers. When the correction command is received, in addition to correcting the image detection process performed by the processor after receiving the correction command, the image detection process performed at least after the processor received the correction command is automatically executed again according to the result of the image detection process performed by the processor after receiving the correction command.

Images