Device and method for verifying x-ray image

Abstract

This image verification device is an X-ray image verification device comprising one or more processors, and a memory for storing one or more programs executed by the one or more processors. The X-ray image verification device comprises: an image acquisition module for acquiring an X-ray image for an object of interest; a detection module for detecting whether an external object is included in the X-ray image and whether there is distortion caused by a movement of the object of interest to acquire detection information; and a determination module for determining, on the basis of the detection information, whether the acquired X-ray image satisfies a criterion for use in X-ray medical diagnosis.

Description

Device and method for verifying X-ray images

[0001] Embodiments of the present invention relate to a device and method for verifying X-ray images.

[0002] In veterinary medicine, the importance of imaging studies is significant due to the limitations of communication caused by the living patient. Imaging studies such as computed tomography (CT) and magnetic resonance imaging (MRI) are extremely useful tools for examining patients for disease. However, because veterinary patients are constantly moving, CT and MRI require general anesthesia. However, the risks and high costs of general anesthesia limit their use. In contrast, X-ray medical diagnostics generally do not require anesthesia and are reasonably priced. Most veterinary hospitals are equipped with X-ray medical diagnostic equipment, making them highly versatile. X-ray medical diagnostics are useful for diagnosing skeletal, cardiac, pulmonary, and abdominal diseases, monitoring ongoing conditions, and assessing treatment responses. However, accurate X-ray medical diagnostics are only possible with normal images that meet established criteria for use in X-ray medical diagnostics. Accordingly, there is a need for a technology for a device and method for verifying X-ray images to determine whether a captured X-ray image is recognized as a normal image.

[0003] An embodiment of the present invention is intended to provide a device and method for verifying an X-ray image for determining whether a photographed X-ray image satisfies a preset standard for being recognized as a normal image.

[0004] An image verification device according to one embodiment disclosed is an X-ray image verification device comprising one or more processors; and a memory storing one or more programs executed by the one or more processors, the device including an image acquisition module for acquiring an X-ray image of an object of interest, a detection module for detecting whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest, thereby acquiring detection information, and a judgment module for determining whether the acquired X-ray image satisfies a standard for use in X-ray medical diagnosis based on the detection information.

[0005] If the above detection information indicates that distortion due to movement of the external object or the object of interest is detected in the X-ray image, the judgment module,

[0006] It can be determined that the above X-ray image does not meet the standards for use in X-ray medical diagnosis.

[0007] The above judgment module can obtain information on whether the posture of the object of interest satisfies the criteria required for performing the X-ray medical diagnosis based on the part of the object of interest captured in the X-ray image, and can determine whether the obtained X-ray image satisfies the criteria for use in the X-ray medical diagnosis based on the information on whether the posture is normal.

[0008] If the above-mentioned information on whether the posture is normal indicates that the X-ray image does not include a region of interest preset to capture a part of the object of interest to be diagnosed, the judgment module may determine that the X-ray image does not meet the criteria for use in X-ray medical diagnosis.

[0009] The above-described preset region of interest may include, if the X-ray image is an image of the chest of the object of interest taken in a dorsal direction, the entire lungs and the 13th rib area of ​​the object of interest; if the X-ray image is an image of the chest of the object of interest taken in a lateral direction, the region of two vertebrae in front of the manubrium of the object of interest and two vertebrae behind the diaphragm; and if the X-ray image is an image of the abdomen of the object of interest, the region of the liver margin and the coxofemoral joint of the object of interest.

[0010] If the above-mentioned posture normality information indicates that the X-ray image is not photographed by overlapping two or more bones that are preset to be photographed by overlapping, the judgment module may determine that the X-ray image does not meet the standards for use in X-ray medical diagnosis.

[0011] The two or more preset bones may include, if the X-ray image is an image of the chest of the object of interest taken in a dorsal direction, the sternum and vertebra of the object of interest; if the X-ray image is an image of the chest of the object of interest taken in a lateral direction, the left rib and right rib of the object of interest; if the X-ray image is an image of the abdomen of the object of interest taken in a lateral direction, the left pelvis and right pelvis of the object of interest; and if the X-ray image is an image of the musculoskeletal system of the object of interest, the femur condyle and the tibia.

[0012] If the above-mentioned information on whether the posture is normal indicates that the X-ray image was taken of the object of interest in a distorted state, the judgment module may determine that the X-ray image does not meet the criteria for use in X-ray medical diagnosis.

[0013] If the above judgment module determines that the X-ray image does not meet the criteria for use in X-ray medical diagnosis, it can display and provide a portion of the X-ray image that does not meet the criteria for use in X-ray medical diagnosis.

[0014] The above detection module learns a labeled X-ray image as to whether the X-ray image contains a foreign object and whether there is distortion caused by movement of the object of interest, and when a new X-ray image is acquired based on the learned result, it can detect whether the new X-ray image contains the foreign object and whether there is distortion caused by movement of the object of interest.

[0015] The above judgment module learns an X-ray image labeled with respect to the information on whether the posture is normal, and when a new X-ray image is acquired based on the learned result, the information on whether the posture is normal for the new X-ray image can be acquired.

[0016] The above image acquisition module can adjust one or more of the size and color of the X-ray image according to preset adjustment criteria, and remove noise of the X-ray image through a pre-learned model.

[0017] According to one disclosed embodiment, it is automatically determined whether a captured X-ray image satisfies specific criteria for being recognized as a normal image, thereby improving the accuracy of X-ray medical diagnosis. Whether a captured X-ray image is a normal image is essential for X-ray medical diagnosis in order to exclude misdiagnosis that may affect the health of a patient.

[0018] Additionally, according to one embodiment disclosed, if the captured X-ray image is abnormal, the reliability of diagnosis can be increased by re-capturing it.

[0019] In addition, according to one embodiment disclosed, since it automatically analyzes whether an X-ray image taken is normal and provides the results, medical staff can quickly and accurately determine whether an X-ray image taken is normal, thereby increasing the efficiency of hospital operations. In other words, time and cost can be saved in determining whether an X-ray image is normal. In addition, as the efficiency of hospital operations is increased, more time can be devoted to examining and treating patients.

[0020] In addition, according to one embodiment disclosed, by utilizing artificial intelligence that analyzes a large amount of data, errors due to human judgment can be reduced, and even non-experts can easily obtain normal images by determining whether an X-ray image is normal.

[0021] In addition, according to one embodiment disclosed, by displaying and providing problematic areas in a captured X-ray image, the number of unnecessary re-captures is reduced, thereby reducing the number of times medical staff taking X-ray images and objects of interest are exposed to radiation, thereby providing a safe treatment environment.

[0022] Additionally, according to one embodiment disclosed, medical staff can obtain X-ray images judged to be normal images and abnormal X-ray images showing problematic areas, and use these as research and educational materials.

[0023] In addition, according to one embodiment disclosed, since it analyzes whether an automatically captured X-ray image is a normal image and provides the result, it is possible to practice capturing X-ray images for X-ray medical diagnosis without an instructor, even for unskilled students.

[0024] The present disclosure can be readily understood by the combination of the following detailed description and the accompanying drawings, wherein reference numerals refer to structural elements.

[0025] FIG. 1 is a schematic diagram for explaining the operation of an X-ray image verification device according to one embodiment of the present invention.

[0026] FIG. 2 is a flowchart showing the operation of an X-ray image verification device according to one embodiment of the present invention.

[0027] FIG. 3 is a drawing showing an example in which an external object other than the object of interest that is the subject of diagnosis is photographed in an X-ray image according to one embodiment of the present invention.

[0028] FIG. 4 is a drawing showing an example of an X-ray image taken in a shaking state of an object of interest according to one embodiment of the present invention.

[0029] FIG. 5 is a drawing showing examples of normal images and abnormal images in a chest X-ray image of an object of interest according to one embodiment of the present invention.

[0030] FIG. 6 is a drawing showing examples of normal images and abnormal images in an abdominal X-ray image of an object of interest according to one embodiment of the present invention.

[0031] FIG. 7 is a drawing showing examples of normal images and abnormal images in a musculoskeletal X-ray image of an object of interest according to one embodiment of the present invention.

[0032] FIG. 8 is a block diagram illustrating how a detection module learns whether an X-ray image contains a foreign object and whether there is distortion caused by movement of the object of interest, according to one embodiment of the present invention.

[0033] FIG. 9 is a block diagram illustrating how a judgment module learns information about whether an X-ray image is in a normal posture, according to one embodiment of the present invention.

[0034] FIG. 10 is a block diagram illustrating a computing environment including a computing device suitable for use in exemplary embodiments.

[0035] An image verification device is an X-ray image verification device having one or more processors; and a memory storing one or more programs executed by the one or more processors, the device including an image acquisition module for acquiring an X-ray image of an object of interest, a detection module for detecting whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest, thereby acquiring detection information, and a judgment module for determining, based on the detection information, whether the acquired X-ray image satisfies a standard for use in X-ray medical diagnosis.

[0036] Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to facilitate a comprehensive understanding of the methods, devices, and / or systems described herein. However, these are merely examples and the present invention is not limited thereto.

[0037] In describing embodiments of the present invention, if a detailed description of a known technology related to the present invention is judged to unnecessarily obscure the gist of the present invention, the detailed description will be omitted. In addition, the terms described below are terms defined in consideration of their functions in the present invention, and this may vary depending on the intention or custom of the user or operator. Therefore, the definitions should be made based on the contents throughout this specification. The terminology used in the detailed description is only for the purpose of describing embodiments of the present invention and should not be limited in any way. Unless clearly used otherwise, the singular form includes the plural form. In this description, expressions such as "comprises" or "having" are intended to indicate certain features, numbers, steps, operations, elements, parts or combinations thereof, and should not be construed to exclude the presence or possibility of one or more other features, numbers, steps, operations, elements, parts or combinations thereof other than those described.

[0038] Additionally, while terms such as "first" and "second" may be used to describe various components, these components should not be limited by these terms. These terms may be used to distinguish one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a "second component," and similarly, a second component may also be referred to as a "first component."

[0039] In this specification, the "X-ray image verification device (100)" can obtain an X-ray image photographed for an object of interest that is a diagnosis target, and determine whether the obtained X-ray image satisfies the criteria for use in X-ray medical diagnosis. The X-ray image verification device (100) can determine whether the X-ray image satisfies the criteria for use in X-ray medical diagnosis for each part of the object of interest photographed in the X-ray image.

[0040] In addition, if the X-ray image verification device (100) determines that the captured X-ray image does not meet the standards for use in X-ray medical diagnosis, it can display the portion of the X-ray image that does not meet the standards for use in X-ray medical diagnosis and provide it to the user.

[0041] At this time, the X-ray image verification device (100) can use a model acquired through machine learning to determine whether the captured X-ray image meets the criteria for use in X-ray medical diagnosis.

[0042] An X-ray image verification device (100) includes one or more processors and a computer-readable recording medium connected to the processor to determine whether an acquired X-ray image is a normal image, and may further include a database for storing data. The computer-readable recording medium may be internal or external to the processor and may be connected to the processor by various well-known means. The processor in the X-ray image verification device (100) may cause the X-ray image verification device (100) to operate according to the exemplary embodiments described herein. For example, the processor may execute instructions stored in the computer-readable recording medium, and the instructions stored in the computer-readable recording medium may be configured to cause the X-ray image verification device (100) to perform operations according to the exemplary embodiments described herein when executed by the processor.

[0043] In this specification, an "object of interest" may be a living organism that is the subject of an X-ray medical diagnosis using an X-ray image. In an exemplary embodiment, the object of interest may be a mammal, a dog, a cat, etc., but is not limited to these examples.

[0044] In this specification, an "external object" may be an object that is not the subject of an X-ray medical diagnosis via an X-ray image. In an exemplary embodiment, the external object may be, but is not limited to, a person's hand holding an object of interest, an object near an X-ray imaging device, etc.

[0045] In this specification, an "X-ray image" may be an image taken to show the inside of the body of an object of interest using X-rays as a light source for X-ray medical diagnosis. The taken X-ray image may show one or more of the bones and internal organs of the object of interest.

[0046] In this specification, a “chest X-ray image” may be an X-ray image taken of a chest portion of an object of interest for the purpose of performing an X-ray medical diagnosis on the chest portion of the living organism, which is the object of interest.

[0047] In this specification, an "abdominal X-ray image" may be an X-ray image taken of the abdominal part of an object of interest for the purpose of performing an X-ray medical diagnosis on the abdominal part of the living body that is the object of interest.

[0048] In this specification, a “musculoskeletal X-ray image” may be an X-ray image of the musculoskeletal system of an object of interest taken to perform an X-ray medical diagnosis on the leg portion of the living organism, which is the object of interest.

[0049]

[0050] FIG. 1 is a schematic diagram for explaining the operation of an X-ray image verification device according to one embodiment of the present invention, and FIG. 2 is a flowchart showing the operation of an X-ray image verification device according to one embodiment of the present invention. The method illustrated in FIG. 2 may be performed, for example, by the aforementioned X-ray image verification device (100). In the illustrated flowchart, the method is described by dividing it into a plurality of steps, but at least some of the steps may be performed in a reversed order, combined with other steps and performed together, omitted, divided into detailed steps and performed, or one or more steps not illustrated may be added and performed.

[0051] Referring to FIG. 1, an X-ray image verification device (100) according to one embodiment may include an image acquisition module (110), a detection module (120), and a judgment module (130).

[0052] Referring to FIGS. 1 and 2, in step S202, the image acquisition module (110) can acquire an X-ray image of an object of interest.

[0053] Specifically, the image acquisition module (110) can acquire an X-ray image by photographing an object of interest with an X-ray image photographing device. Alternatively, the image acquisition module (110) can acquire an X-ray image from a database storing X-ray images.

[0054] At this time, the object of interest may be a living organism that is the subject of an X-ray medical diagnosis using an X-ray image. In an exemplary embodiment, the object of interest may be a mammal, a dog, a cat, etc., but is not limited to these examples.

[0055] Additionally, an X-ray image may be an image taken using X-rays as a light source to show the inside of the body of an object of interest for X-ray medical diagnosis. The X-ray image may show one or more of the bones and internal organs of the object of interest.

[0056] The X-ray image acquired by the image acquisition module (110) may be one of a chest X-ray image, an abdominal X-ray image, and a musculoskeletal X-ray image. At this time, the chest X-ray image may be an X-ray image taken of the chest part of the object of interest, which is the object of interest, in order to perform an X-ray medical diagnosis on the chest part of the object of interest. In addition, the abdominal X-ray image may be an X-ray image taken of the abdominal part of the object of interest, which is the object of interest, in order to perform an X-ray medical diagnosis on the abdominal part of the object of interest. In addition, the musculoskeletal X-ray image may be an X-ray image taken of the musculoskeletal system of the object of interest, which is the object of interest, in order to perform an X-ray medical diagnosis on the leg part of the object of interest.

[0057] The image acquisition module (110) can preprocess the X-ray image to use it as input for the detection module (120) and judgment module (130) that determine whether the acquired X-ray image meets the criteria for use in X-ray medical diagnosis.

[0058] Specifically, the image acquisition module (110) can adjust one or more of the size and color of the X-ray image according to preset adjustment criteria, and remove noise of the X-ray image through a pre-learned model. In an exemplary embodiment, the size of the X-ray image can be adjusted to one of 256x256, 512x512, and 1024x1024. In addition, in an exemplary embodiment, the image acquisition module (110) can adjust the color of the X-ray image to a 266 scale.

[0059] The image acquisition module (110) can transmit the preprocessed X-ray image to the detection module (120).

[0060] In step S204, the detection module (120) can detect whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest, thereby obtaining detection information. Specific examples regarding detection of distortion caused by movement of an external object or an object of interest in the X-ray image will be described later with reference to FIGS. 3 and 4. At this time, the detection module (120) can machine-learn an X-ray image labeled with detection information and obtain detection information based on the learning result. Specific details thereof will be described later with reference to FIG. 8.

[0061] The detection module (120) can acquire a new X-ray image that has undergone preprocessing from the image acquisition module (110). Based on the learned results, the detection module (120) can detect whether a foreign object is included in the new X-ray image and whether there is distortion caused by movement of the object of interest. The detection module (120) can acquire detection information based on the detection results and transmit the detection information to the judgment module.

[0062] If the detection information indicates that no distortion due to movement of an external object or an object of interest is detected in the X-ray image, the judgment module (130) may operate according to step S206, and if the detection information indicates that an external object and distortion are detected in the X-ray image, the judgment module (130) may operate according to step S208.

[0063] In step S206, the judgment module (130) can obtain information on whether the posture is normal by diagnosing whether the posture of the object of interest satisfies the criteria required for performing an X-ray medical diagnosis, based on the area of ​​the object of interest captured in the X-ray image. At this time, the judgment module (130) can obtain information on whether the posture is normal through one of the chest diagnosis module (132), the abdomen diagnosis module (134), and the musculoskeletal diagnosis module (136), based on the area of ​​the captured X-ray image.

[0064] In addition, the judgment module (130) can determine whether the acquired X-ray image satisfies the criteria for use in X-ray medical diagnosis based on information about whether the posture is normal. Specific examples of the posture of the object of interest required for performing X-ray medical diagnosis are described below with reference to FIGS. 5 to 7.

[0065] Specifically, at this time, the judgment module (130) can machine learn the X-ray image labeled with information on whether the posture is normal, and acquire information on whether the posture is normal based on the learned result. The specific details thereof will be described later with reference to FIG. 9. The information on whether the posture is normal can include not only whether the posture of the object of interest satisfies the criteria for performing an X-ray medical diagnosis, but also, if it does not, specific information on what is not satisfied.

[0066] Next, the judgment module (130) can obtain an X-ray image from the detection module (120) in which distortion due to movement of an external object and an object of interest is not detected. Based on the results learned regarding whether the posture is normal, the judgment module (130) can obtain information on whether the posture is normal for the X-ray image.

[0067] In step S208, the judgment module (130) can determine whether the acquired X-ray image satisfies the criteria for use in X-ray medical diagnosis based on the detection information and the posture normality information.

[0068] Specifically, if the detection information indicates that a distortion due to movement of an external object or an object of interest is detected in the X-ray image, the judgment module (130) may determine that the X-ray image does not meet the criteria for use in X-ray medical diagnosis. In addition, if the detection information indicates that an external object and distortion are not detected in the X-ray image, the judgment module (130) may determine that the X-ray image meets the criteria for use in X-ray medical diagnosis. In addition, if the posture normality information determines that the posture of the object of interest does not meet the criteria required for performing X-ray medical diagnosis, the judgment module (130) may determine that the X-ray image does not meet the criteria for use in X-ray medical diagnosis. In addition, if the posture normality information determines that the posture of the object of interest meets the criteria required for performing X-ray medical diagnosis, the X-ray image may be determined to meet the criteria for use in X-ray medical diagnosis.

[0069] In step S210, if the judgment module (130) determines that the X-ray image does not meet the criteria for use in X-ray medical diagnosis, it can display and provide a portion of the X-ray image that does not meet the criteria for use in X-ray medical diagnosis.

[0070] Specifically, the judgment module (130) can display the external object and distortion on the X-ray image in which the external object and distortion are detected based on the result of learning about the distortion caused by the movement of the external object and the object of interest on the X-ray image. The judgment module (130) can provide the X-ray image in which the external object and distortion are displayed to the user. In addition, the judgment module (130) can display the unsatisfactory portion on the X-ray image in which the posture of the object of interest does not meet the standard for performing the X-ray medical diagnosis based on the result of learning whether the posture of the object of interest on the X-ray image satisfies the standard for performing the X-ray medical diagnosis. The judgment module (130) can provide the X-ray image in which the unsatisfactory portion is displayed to the user.

[0071] If the judgment module (130) determines that the X-ray image does not meet the criteria for use in X-ray medical diagnosis, it can provide correction information regarding posture correction required in re-shooting so that the X-ray image meets the criteria for use in X-ray medical diagnosis.

[0072] Specifically, if the posture normality information indicates that the X-ray image does not include a region of interest preset to capture a portion of the object of interest to be diagnosed, the correction information may include information regarding in which direction the position of the object of interest should be moved.

[0073] Additionally, if the posture normality information indicates that two or more bones that are preset to be overlaid on the X-ray image are not overlaid, the correction information may be used to indicate in which direction (e.g., clockwise or counterclockwise, with the head-to-tail direction of the object of interest as the axis) the body of the object of interest is rotated to some degree (e.g., angle 0). o 90 inland o ) may contain information about whether to rotate.

[0074] Additionally, if the posture normality information indicates that the X-ray image was taken of an object of interest in a distorted state, the correction information indicates which part of the object of interest (e.g., head or leg) is distorted and to what extent (e.g., angle 0) and in what direction (e.g., clockwise or counterclockwise with the head-to-tail direction of the object of interest as the axis) o 90 inland o ) may contain information about whether to rotate.

[0075]

[0076] FIG. 3 is a drawing showing an example in which an external object other than the object of interest to be examined is photographed in an X-ray image according to one embodiment of the present invention.

[0077] Referring to FIG. 3, in one embodiment, the X-ray image may be one of an image (301, 301) of the musculoskeletal system of the object of interest, an image (303) of the chest of the object of interest, and an image (304) of the abdomen of the object of interest. As indicated by the circled portions of images 301 to 304, the X-ray image may include a human hand holding the living object of interest as an external object.

[0078]

[0079] FIG. 4 is a drawing showing an example of an X-ray image taken in a shaking state of an object of interest according to one embodiment of the present invention.

[0080] Referring to FIG. 4, in one embodiment, the X-ray image may be one of an image (401, 403) of the chest of the object of interest, an image (402) of the abdomen of the object of interest, and an image of the musculoskeletal system of the object of interest. As shown in X-ray images 401 to 404, the X-ray image may express the boundary of the object of interest as unclear and blurry. The unclear and blurry boundary may occur as the object of interest moves during the imaging.

[0081]

[0082] FIG. 5 is a drawing showing examples of normal images and abnormal images in a chest X-ray image of an object of interest according to one embodiment of the present invention.

[0083] Referring to FIG. 5, in one embodiment, the X-ray image may be an image (510, 520) of the chest of the object of interest.

[0084] At this time, the X-ray image may be an image (510) taken of the chest of the object of interest in a dorsal ventral direction. The X-ray image 511 may be an image in which the posture of the object of interest satisfies the standards required for performing an X-ray medical diagnosis.

[0085] X-ray images 512 to 514 may be images in which the posture of the object of interest does not meet the standards required for performing an X-ray medical diagnosis. X-ray image 512 may be an image that does not include a region of interest preset to capture a part of the object of interest to be diagnosed. In this case, the preset region of interest may be an area including the entire lung and the 13th rib of the object of interest. X-ray image 513 may be an image in which two or more bones preset to be captured in an overlapping manner are not captured in an overlapping manner. In this case, the two or more bones preset may include the sternum and the vertebra of the object of interest. X-ray image 514 may be an image in which the object of interest is captured in a distorted state. In this case, the vertebral body of the object of interest in the X-ray image may be rotated.

[0086] Alternatively, the X-ray image may be an image (520) taken in a lateral direction of the chest of the object of interest. The X-ray image 521 may be an image in which the posture of the object of interest meets the criteria required for performing an X-ray medical diagnosis.

[0087] X-ray images 522 to 525 may be images in which the posture of the object of interest does not meet the standards required for performing an X-ray medical diagnosis. X-ray image 522 may be an image that does not include a region of interest preset to capture a part of the object of interest to be diagnosed. In this case, the preset region of interest may be an area including two vertebrae anterior to the manubrium of the object of interest and two vertebrae posterior to the diaphragm. X-ray image 523 may be an image in which two or more bones preset to be captured superimposed are not captured superimposed. In this case, the two or more preset bones may include the left rib and the right rib of the object of interest. X-ray image 524 may be an image in which the object of interest is captured in a distorted state. In this case, the rib of the object of interest in the X-ray image may be rotated. X-ray image 525 may be an image in which the front of the object of interest is not pulled toward the head.

[0088]

[0089] FIG. 6 is a drawing showing examples of normal images and abnormal images in an abdominal X-ray image of an object of interest according to one embodiment of the present invention.

[0090] Referring to FIG. 6, in one embodiment, the X-ray image may be an image (610, 620) of the abdomen of the object of interest.

[0091] At this time, the X-ray image may be an image (610) taken in a dorsal ventral direction of the abdomen of the object of interest. The X-ray image 611 may be an image in which the posture of the object of interest satisfies the standards required for performing an X-ray medical diagnosis.

[0092] X-ray images 612 to 614 may be images in which the posture of the object of interest does not meet the standards required for performing an X-ray medical diagnosis. X-ray images 612 and 613 may be images that do not include a region of interest preset to capture a part of the object of interest to be diagnosed. In this case, the preset region of interest may be a region including a liver margin and a coxofemoral joint region of the object of interest. X-ray image 612 may be an image that does not include the liver margin region of the object of interest, and X-ray image 613 may be an image that does not include the hip joint region of the object of interest. X-ray image 614 may be an image captured in which the object of interest is captured in a distorted state. In this case, the posture of the object of interest in the X-ray image may be distorted.

[0093] Alternatively, the X-ray image may be an image (620) taken in a lateral direction of the abdomen of the object of interest. The X-ray image 621 may be an image in which the posture of the object of interest meets the criteria required for performing an X-ray medical diagnosis.

[0094] X-ray images 622 to 624 may be images in which the posture of the object of interest does not meet the standards required for performing an X-ray medical diagnosis. X-ray images 622 and 623 may be images that do not include a region of interest that is preset to capture a part of the object of interest to be diagnosed. At this time, the preset region of interest may be a region that includes a liver margin and a coxofemoral joint region of the object of interest. X-ray image 622 may be an image that does not include a liver margin region of the object of interest, and X-ray image 623 may be an image that does not include a hip joint region of the object of interest. X-ray image 624 may be an image in which two or more bones that are preset to be superimposed are not superimposed. At this time, the two or more bones that are preset may include a left pelvis and a right pelvis of the object of interest.

[0095]

[0096] FIG. 7 is a drawing showing examples of normal images and abnormal images in a musculoskeletal X-ray image of an object of interest according to one embodiment of the present invention.

[0097] Referring to FIG. 7, in one embodiment, the X-ray image may be an image (710, 720) of the musculoskeletal system of the object of interest.

[0098] At this time, the X-ray image may be an image (710) taken in a bent state of the musculoskeletal system of the object of interest. The X-ray image 711 may be an image in which the posture of the object of interest satisfies the standards required for performing an X-ray medical diagnosis.

[0099] X-ray image 712 may be an image in which the posture of the object of interest does not meet the criteria required for performing an X-ray medical diagnosis. X-ray image 712 may be an image in which two or more bones that are preset to be superimposed are not superimposed. In this case, the two or more preset bones may include the femur condyle and the tibia.

[0100] Alternatively, the X-ray image may be an image (720) taken while the musculoskeletal system of the object of interest is in a straight line. The X-ray image 721 may be an image in which the posture of the object of interest satisfies the criteria required for performing an X-ray medical diagnosis.

[0101] X-ray image 722 may be an image in which the pose of the object of interest does not meet the standards required for performing an X-ray medical diagnosis. X-ray image 722 may be an image in which the object of interest is captured in a distorted state. In this case, the object of interest in the X-ray image may be rotated so as not to conform to the alignment.

[0102]

[0103] FIG. 8 is a block diagram illustrating how a detection module learns whether an X-ray image contains a foreign object and whether there is distortion caused by movement of the object of interest, according to one embodiment of the present invention.

[0104] Referring to FIG. 8, the detection module (120) can obtain a labeled X-ray image as input data, whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest, and, based on the result of learning the input data, obtain a detection model (120a) that obtains a result as output data, whether an external object is included and whether there is distortion caused by movement of the object of interest.

[0105] In addition, the detection module (130) can adjust the number of X-ray images including external objects, which are learning data sets, and X-ray images with distortion to be equal in order to improve the accuracy of the detection model (120a). The detection module (120) can augment X-ray images with a relatively smaller number among X-ray images including external objects and X-ray images with distortion so that the number of X-ray images including external objects and X-ray images with distortion is equal. Augmentation of the X-ray images can be performed through one or more of rotation of the X-ray images, cropping of the X-ray images, flipping of the X-ray images, scaling of the X-ray images, translation of the X-ray images, color jittering of the X-ray images, adding noise to the X-ray images, and distortion of the X-ray images.

[0106] In addition, the detection module (120) can learn X-ray images that do not meet the criteria for use in X-ray medical diagnosis, and can generate new X-ray images that do not meet the criteria for use in X-ray medical diagnosis. In an exemplary embodiment, the detection module (120) can generate new X-ray images that do not meet the criteria for use in X-ray medical diagnosis through one or more of a variational vutoencoder (VAE) that converts an image into a latent vector and then restores it, a generative adversarial network (GAN) that generates a high-quality image through adversarial learning of an image generator and a discriminator, and a diffusion model that generates an image through a process of removing noise added to the image.

[0107] In addition, the detection module (120) can determine whether an external object is included in the labeled X-ray image and whether there is distortion caused by movement of the object of interest, and train a detection model so that the difference between the determined data and the labeled data is minimized.

[0108] The detection module (120) can obtain, as output data, a result of whether a foreign object is included in the new X-ray image and whether there is distortion caused by movement of the object of interest when a new X-ray image is obtained as input data.

[0109]

[0110] FIG. 9 is a block diagram illustrating how a judgment module learns information about whether an X-ray image is in a normal posture, according to one embodiment of the present invention.

[0111] Referring to FIG. 9, the judgment module (130) can obtain a labeled X-ray image as input data to determine whether the posture of the object of interest satisfies the criteria required for performing an X-ray medical diagnosis, and obtain a judgment model (130a) that obtains a result as output data as to whether the posture of the object of interest satisfies the criteria required for performing an X-ray medical diagnosis based on the result of learning the input data.

[0112] In addition, the judgment module (130) can adjust the number of X-ray images for each region of the object of interest, which is a learning data set, to an equal number in order to improve the accuracy of the judgment model (130a). The judgment module (130) can augment X-ray images of regions with a relatively small number of X-ray images so that the X-ray images are equally distributed across each region of the object of interest. Augmentation of the X-ray image can be performed through one or more of rotation of the X-ray image, cropping of the X-ray image, flipping of the X-ray image, scaling of the X-ray image, translation of the X-ray image, color jittering of the X-ray image, adding noise to the X-ray image, and distortion of the X-ray image.

[0113] In addition, the judgment module (120) can learn X-ray images that do not meet the criteria for use in X-ray medical diagnosis, and can generate new X-ray images that do not meet the criteria for use in X-ray medical diagnosis. In an exemplary embodiment, the detection module (120) can generate new X-ray images that do not meet the criteria for use in X-ray medical diagnosis through one or more of a variational vutoencoder (VAE) that converts an image into a latent vector and then restores it, a generative adversarial network (GAN) that generates a high-quality image through adversarial learning of an image generator and a discriminator, and a diffusion model that generates an image through a process of removing noise added to the image.

[0114] In addition, the judgment module (130) can judge whether the posture of the object of interest in the labeled X-ray image satisfies the criteria required for performing an X-ray medical diagnosis, and train a detection model so that the difference between the judged data and the labeled data is minimized.

[0115] The judgment module (120) can obtain, as output data, a result of whether the posture of the object of interest in the new X-ray image satisfies the criteria required for performing an X-ray medical diagnosis when a new X-ray image is obtained as input data.

[0116]

[0117] According to one disclosed embodiment, it is automatically determined whether a captured X-ray image satisfies specific criteria for being recognized as a normal image, thereby improving the accuracy of X-ray medical diagnosis. Whether a captured X-ray image is a normal image is essential for X-ray medical diagnosis in order to exclude misdiagnosis that may affect the health of a patient.

[0118] Additionally, according to one embodiment disclosed, if the captured X-ray image is abnormal, the reliability of diagnosis can be increased by re-capturing it.

[0119] In addition, according to one embodiment disclosed, since it automatically analyzes whether an X-ray image taken is normal and provides the results, medical staff can quickly and accurately determine whether an X-ray image taken is normal, thereby increasing the efficiency of hospital operations. In other words, time and cost can be saved in determining whether an X-ray image is normal. In addition, as the efficiency of hospital operations is increased, more time can be devoted to examining and treating patients.

[0120] In addition, according to one embodiment disclosed, by utilizing artificial intelligence that analyzes a large amount of data, errors due to human judgment can be reduced, and even non-experts can easily obtain normal images by determining whether an X-ray image is normal.

[0121] In addition, according to one embodiment disclosed, by displaying and providing problematic areas in a captured X-ray image, the number of unnecessary re-captures is reduced, thereby reducing the number of times medical staff taking X-ray images and objects of interest are exposed to radiation, thereby providing a safe treatment environment.

[0122] Additionally, according to one embodiment disclosed, medical staff can obtain X-ray images judged to be normal images and abnormal X-ray images showing problematic areas, and use these as research and educational materials.

[0123] In addition, according to one embodiment disclosed, since it analyzes whether an automatically captured X-ray image is a normal image and provides the result, it is possible to practice capturing X-ray images for X-ray medical diagnosis without an instructor, even for unskilled students.

[0124]

[0125] FIG. 10 is a block diagram illustrating a computing environment (10) including a computing device suitable for use in exemplary embodiments. In the illustrated embodiment, each component may have different functions and capabilities other than those described below, and may include additional components other than those described below.

[0126] The illustrated computing environment (10) includes a computing device (12). In one embodiment, the computing device (12) may be an X-ray image verification device (100).

[0127] The image acquisition module (110) of FIG. 1 may correspond to the input / output device (24) of FIG. 10, and the detection module (120) and judgment module (130) of FIG. 1 may correspond to the processor (14) of FIG. 10.

[0128] A computing device (12) includes at least one processor (14), a computer-readable storage medium (16), and a communication bus (18). The processor (14) may cause the computing device (12) to operate according to the exemplary embodiments mentioned above. For example, the processor (14) may execute one or more programs stored in the computer-readable storage medium (16). The one or more programs may include one or more computer-executable instructions, which, when executed by the processor (14), may be configured to cause the computing device (12) to perform operations according to the exemplary embodiments.

[0129] A computer-readable storage medium (16) is configured to store computer-executable instructions or program code, program data, and / or other suitable forms of information. A program (20) stored in the computer-readable storage medium (16) includes a set of instructions executable by the processor (14). In one embodiment, the computer-readable storage medium (16) may be a memory (volatile memory such as random access memory, non-volatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, any other form of storage medium that can be accessed by the computing device (12) and store desired information, or a suitable combination thereof.

[0130] A communication bus (18) interconnects various other components of the computing device (12), including the processor (14) and computer-readable storage media (16).

[0131] The computing device (12) may also include one or more input / output interfaces (22) that provide interfaces for one or more input / output devices (24) and one or more network communication interfaces (26). The input / output interfaces (22) and the network communication interfaces (26) are connected to the communication bus (18). The input / output devices (24) may be connected to other components of the computing device (12) via the input / output interfaces (22). Exemplary input / output devices (24) may include input devices such as pointing devices (such as a mouse or a trackpad), a keyboard, a touch input device (such as a touchpad or a touchscreen), a voice or sound input device, various types of sensor devices and / or photographing devices, and / or output devices such as display devices, printers, speakers and / or network cards. The exemplary input / output devices (24) may be included within the computing device (12) as a component constituting the computing device (12), or may be connected to the computing device (12) as a separate device distinct from the computing device (12).

[0132]

[0133] While representative embodiments of the present invention have been described in detail above, those skilled in the art will appreciate that various modifications to the above-described embodiments are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the claims set forth below but also by equivalents thereof.

Claims

1. One or more processors; and An X-ray image verification device having a memory for storing one or more programs executed by one or more processors, An image acquisition module for acquiring an X-ray image of an object of interest; A detection module that detects whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest to obtain detection information; and An X-ray image verification device including a judgment module that judges whether the acquired X-ray image satisfies the criteria for use in X-ray medical diagnosis based on the above detection information.

2. In claim 1, If the above detection information indicates that distortion due to movement of the external object or the object of interest has been detected in the X-ray image, The above judgment module, An X-ray image verification device that determines that the above X-ray image does not meet the standards for use in X-ray medical diagnosis.

3. In claim 1, The above judgment module, According to the part of the object of interest captured in the X-ray image, information on whether the posture of the object of interest is normal is obtained by diagnosing whether the posture of the object of interest satisfies the criteria required for performing the X-ray medical diagnosis, An X-ray image verification device that determines whether the acquired X-ray image meets the criteria for use in X-ray medical diagnosis based on the above-mentioned information on whether the posture is normal.

4. In claim 3, If the above posture normality information indicates that the X-ray image does not include a region of interest preset to capture the part of the object of interest to be diagnosed, The above judgment module, An X-ray image verification device that determines that the above X-ray image does not meet the standards for use in X-ray medical diagnosis.

5. In claim 4, The above preset area of ​​interest is, If the above X-ray image is an image taken in the dorsal direction of the chest of the object of interest, it includes the entire lung of the object of interest and the area of ​​the 13th rib, If the above X-ray image is an image taken in the lateral direction of the chest of the object of interest, it includes the area of ​​two vertebrae in front of the manubrium of the object of interest and two vertebrae in back of the diaphragm, An X-ray image verification device that includes the liver margin and the coxofemoral joint area of ​​the object of interest when the X-ray image is an image taken of the abdomen of the object of interest.

6. In claim 3, If the above posture normality information indicates that the X-ray image is not taken by overlapping two or more bones that are preset to be overlapping, The above judgment module, An X-ray image verification device that determines that the above X-ray image does not meet the standards for use in X-ray medical diagnosis.

7. In claim 6, The above two or more preset skeletons are, If the above X-ray image is an image taken in the dorsal direction of the chest of the object of interest, it includes the sternum and vertebra of the object of interest, If the above X-ray image is an image taken in a lateral direction of the chest of the object of interest, it includes the left rib and right rib of the object of interest, If the above X-ray image is an image taken in a lateral direction of the abdomen of the object of interest, it includes the left pelvis and right pelvis of the object of interest, If the above X-ray image is an image of the musculoskeletal system of the object of interest, An X-ray image verification device that includes the femur condyle and tibia.

8. In claim 3, If the above posture normality information indicates that the above X-ray image was taken of the object of interest in a distorted state, The above judgment module, An X-ray image verification device that determines that the above X-ray image does not meet the standards for use in X-ray medical diagnosis.

9. In claim 1, The above judgment module, If the above X-ray image is judged not to meet the criteria for use in X-ray medical diagnosis, An X-ray image verification device that provides an indication of a part of the X-ray image that does not meet the criteria for use in the X-ray medical diagnosis.

10. In claim 1, The above judgment module, If the above X-ray image is judged not to meet the criteria for use in X-ray medical diagnosis, An X-ray image verification device that provides correction information regarding posture correction required in re-examination so that the above X-ray image meets the standards for use in X-ray medical diagnosis.

11. In claim 1, The above detection module, An X-ray image verification device that learns a labeled X-ray image to determine whether a foreign object is included in the X-ray image and whether there is distortion caused by movement of the object of interest, and when a new X-ray image is acquired based on the learned result, detects whether the new X-ray image contains the foreign object and whether there is distortion caused by movement of the object of interest.

12. In claim 3, The above judgment module, An X-ray image verification device that learns an X-ray image labeled with respect to the above posture normality information and acquires a new X-ray image based on the learned result, and acquires the posture normality information for the new X-ray image.

13. In claim 1, The above image acquisition module, Adjusting one or more of the size and color of the above X-ray image according to preset adjustment criteria, An X-ray image verification device that removes noise from the above X-ray image using a learned model.

14. One or more processors, and A method performed on a computing device having a memory storing one or more programs executed by one or more processors, A step of acquiring an X-ray image of an object of interest; A step of detecting whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest to obtain detection information; and An X-ray image verification method, comprising a step of determining whether the acquired X-ray image satisfies the criteria for use in X-ray medical diagnosis based on the above detection information.

15. A computer program stored in a non-transitory computer readable storage medium, The computer program includes one or more instructions, which, when executed by a computing device having one or more processors, cause the computing device to: A step of acquiring an X-ray image of an object of interest; A step of detecting whether an external object is included in the X-ray image and whether there is distortion caused by movement of the object of interest to obtain detection information; and A computer program that performs a step of determining whether the acquired X-ray image satisfies the criteria for use in X-ray medical diagnosis based on the above detection information.

Images