Ray equipment processing method and device based on image data and storage medium

Abstract

The invention provides a ray equipment processing method and device based on image data and a storage medium, and relates to the X-ray equipment detection technology, the body information corresponding to a user is determined according to the image data of the user, a standard model is updated based on the body information to obtain a customized model corresponding to the user, and the posture image of the user is obtained. Synchronizing a model attitude of the customized model according to attitude information of the attitude image; generating reference information based on a comparison result of the model attitude and a standard attitude; starting ray equipment based on completion information of the reference information; obtaining a brightness parameter of a target area according to a regulation and control module; the regulation and control module is controlled to adjust the radiation dose in real time, the radiation dose of the X-rays can be accurately adjusted, and the accuracy during image imaging is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to X-ray equipment detection technology, and in particular to a ray equipment processing method and device based on image data and a storage medium. BACKGROUND

[0002] Medical X-ray equipment plays a vital role in modern medical diagnosis. By taking advantage of the absorption difference of X-rays penetrating different tissues in the human body, it can generate images reflecting the internal condition of the human body, allowing doctors to distinguish different tissues and identify lesions, providing a key basis for doctors to understand the internal structure of the human body and accurately diagnose diseases. In this process, accurate adjustment of the radiation dose of X-rays is a key link when medical X-ray equipment generates images reflecting the internal structure of the human body.

[0003] In the prior art, medical X-ray equipment often relies on preset fixed parameters to adjust the radiation dose of X-rays, which is difficult to accurately adjust the dose according to the specific situation of the user. However, due to significant differences in body size, tissue density, etc. of different users, the absorption of X-rays by different users is different. For example, for users with larger body size or higher tissue density, a higher dose may be required to obtain a clear image. However, fixed dose adjustment parameters may result in insufficient radiation dose of X-rays, which may affect the accuracy of image imaging. For users with smaller body size or lower tissue density, the dose may be too high, causing excessive exposure of the user's tissue, which may mask some details that can be clearly displayed, thereby affecting the accuracy of image imaging.

[0004] Therefore, how to accurately adjust the radiation dose of X-rays and improve the accuracy of image imaging has become a problem to be solved. SUMMARY

[0005] The present application provides a ray equipment processing method and device based on image data and a storage medium, which can accurately adjust the radiation dose of X-rays and improve the accuracy of image imaging.

[0006] In a first aspect of the present application, a ray equipment processing method based on image data is provided, comprising: determining the body information corresponding to the user according to the image data of the user, updating the standard model based on the body information to obtain the customized model corresponding to the user; obtaining the posture image of the user, synchronizing the model posture of the customized model according to the posture information of the posture image, and generating reference information based on the comparison result of the model posture and the standard posture; starting the ray equipment based on the completion information of the reference information, and obtaining the brightness parameter of the target area according to the regulation module; According to the comparison result of the brightness parameter and the target brightness parameter, the control module is controlled to adjust the radiation dose in real time.

[0007] Optionally, in a possible implementation of the first aspect, the body information corresponding to the user is determined according to the image data of the user, and the standard model is updated to obtain the customized model corresponding to the user based on the body information, including: extracting a human body contour in the image data, and determining a contour specification of the human body contour in a length direction and a width direction; calling a positioning template corresponding to the contour specification and the detection attribute of the user, and positioning a center point of the positioning template to a center point of the human body contour; determining a key mapping area corresponding to a preset key area of the positioning template in the human body contour, and extracting a part key point in the key mapping area; obtaining a length ratio between the associated part key points, determining a target key point corresponding to the part key point in the standard model, and proportionally updating a part module corresponding to the corresponding target key point to obtain the customized model according to the length ratio, wherein the body information includes the length ratio.

[0008] Optionally, in a possible implementation of the first aspect, an attitude image of the user is obtained, a model attitude of the customized model is synchronized according to an attitude parameter of the attitude image, and reference information is generated based on a comparison result of the model attitude and a standard attitude, including: dividing feature points with associated attributes in the attitude image into a same determination set; obtaining a geometric quantization index corresponding to each feature point in the determination set and a target quantization index corresponding to a target point mapped with each feature point in the customized model; determining an attitude adjustment parameter of a corresponding part module in the customized model according to a quantization index difference between the geometric quantization index and the target quantization index, and updating the model attitude of the customized model according to the attitude adjustment parameter, wherein the attitude adjustment parameter includes the geometric quantization index; when the model attitude is different from the standard attitude, generating adjustment display data of the model attitude to the standard attitude, and the reference information includes the adjustment display data.

[0009] Optionally, in a possible implementation of the first aspect, the geometric quantization index corresponding to each feature point in the determination set and the target quantization index corresponding to the target point mapped with each feature point in the customized model are obtained, including: calculating a first distance between adjacent feature points in the determination set, and determining a first distance ratio corresponding to the corresponding feature point according to a ratio between the first distances; sequentially obtaining a preset number of adjacent feature points as selected targets, calculating a first angle corresponding to the corresponding feature point according to a vector of the selected target, and obtaining the geometric quantization index according to the first distance ratio and the first angle. a second distance between adjacent target points is calculated, and a second distance ratio corresponding to the adjacent target points is determined according to a ratio between the second distances; a second angle corresponding to the corresponding target point is calculated according to a vector of the target point corresponding to the selected target, and the target quantitative index is obtained according to the second distance ratio and the second angle.

[0010] Optionally, in a possible implementation manner of the first aspect, a posture adjustment parameter of a corresponding part module in the customized model is determined according to a quantitative index difference between the geometric quantitative index and the target quantitative index, and a model posture of the customized model is updated according to the posture adjustment parameter, and the posture adjustment parameter includes the geometric quantitative index, and includes: a proportional difference and an angular difference of the corresponding feature point and the target point are obtained, and a target point with a proportional difference greater than or equal to a proportional threshold and / or an angular difference greater than or equal to an angular threshold is determined as an adjustment node; an adjustment direction of a part module corresponding to the adjustment node is determined according to a bidirectional attribute of the angular difference, and the bidirectional attribute includes a positive attribute and a negative attribute; an adjustment parameter corresponding to the corresponding part module is determined based on the adjustment direction and the angular difference, and a model posture of the customized model is updated according to the adjustment parameter.

[0011] Optionally, in a possible implementation manner of the first aspect, when the model posture is different from the standard posture, adjustment display data from the model posture to the standard posture is generated, and the adjustment display data includes: standard angles of each part module of the customized model in the standard posture are determined, and a current angle of each part module in the model posture is compared with the standard angle; if there is a part module with an absolute value of a difference between the current angle and the standard angle greater than a standard angle threshold, it is determined that the model posture is different from the standard posture, and the corresponding part module is determined as an abnormal module; a calibration direction is determined according to a calibration angular difference between the current angle and the standard angle, a dynamic display video is obtained by rotating the part module based on the calibration direction and the calibration angular difference, and a change displacement between the same target points in the part module in a rotation process is obtained; a length corresponding to a maximum change displacement is determined as a length of an indication mark, a direction of the indication mark is determined according to the calibration direction, and the adjustment display data is obtained according to the indication mark and the dynamic display video.

[0012] Optionally, in a possible implementation manner of the first aspect, a ray device is started based on completion information of the reference information, and a brightness parameter of a target region is obtained according to the regulation module, and the brightness parameter includes: an interaction request of a user based on a user terminal is obtained, and the interaction request includes a single interaction request and a synchronous interaction request; According to the interactive request, a target part selected by the user is determined, a corresponding calibration image of the user is collected, a part contour corresponding to the target part in the calibration image is extracted, and a current posture of the target part is determined. When the current postures of all target parts are equal to the standard postures, completion information of the reference information is responded to, and the ray device is started; A real-time image collected by the ray device is acquired, a target region in the real-time image is extracted according to the detection template, and a brightness parameter of the target region is acquired according to the regulation module.

[0013] Optionally, in a possible implementation manner of the first aspect, according to the interactive request, a target part selected by the user is determined, a corresponding calibration image of the user is collected, a part contour corresponding to the target part in the calibration image is extracted, and a current posture of the target part is determined. Based on a single interactive request, an abnormal module currently selected by the user is determined as a target part, and an abnormal range corresponding to the target part is determined according to a length range of the target part in a length direction and a width direction. A reference positioning template is called, a part reference point corresponding to the target part in the reference positioning template is determined, a region corresponding to the abnormal range is determined as a part positioning area with the part reference point as a center, and the reference positioning template includes part reference points corresponding to multiple human body parts. The reference positioning template is positioned to the calibration image, and a part contour in the part positioning area in the calibration image is extracted. If there is no complete part contour in the part positioning area, the part positioning area is expanded according to a preset magnification ratio until a complete part contour is extracted in the part positioning area. According to a synchronous interactive request, it is determined that all abnormal modules are target parts, and a part contour corresponding to the target part in the calibration image is extracted.

[0014] Optionally, in a possible implementation manner of the first aspect, according to a comparison result of the brightness parameter and the target brightness parameter, the regulation module is controlled to adjust the ray dose in real time, including: A brightness difference value of the brightness parameter and the target brightness parameter is acquired, when the brightness difference value is greater than a brightness difference threshold value, a dose adjustment amount is determined according to the brightness difference value; A regulation fluctuation direction is determined, the regulation fluctuation direction includes an upward direction and a downward direction, the regulation module is controlled to add the dose adjustment amount to obtain an updated ray dose based on the upward direction; The regulation module is controlled to subtract the dose adjustment amount to obtain an updated ray dose according to the downward direction.

[0015] In a second aspect of the present application, a ray device processing apparatus based on image data is provided, including: An updating module is configured to determine body information of a user according to image data of the user, and update a standard model based on the body information to obtain a customized model corresponding to the user. A synchronizing module is configured to obtain a posture image of the user, synchronize a model posture of the customized model according to posture information of the posture image, and generate reference information based on a comparison result of the model posture and a standard posture. A starting module is configured to start a radiographic device based on completion information of the reference information, and obtain a brightness parameter of a target region according to the regulating module. A control module is configured to control the regulating module to adjust a radiation dose in real time according to a comparison result of the brightness parameter and a target brightness parameter.

[0016] In a third aspect, the present application provides a readable storage medium, wherein the readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method of the first aspect and various possible aspects related to the first aspect.

[0017] The present application has the following advantages: 1. The present application can generate a customized model matching a user, and guide and adjust the posture of the user in an X-ray imaging process through the customized model, so that the posture of the user during imaging is as close as possible to a standard imaging posture, and the accuracy during imaging is improved.

[0018] 2. The present application can obtain a brightness parameter of a target region of a human body through a medical X-ray device, and adjust a radiation dose of the medical X-ray in real time according to a comparison result of the brightness parameter and a preset target brightness parameter, so that the image brightness of the target region is always maintained near the target brightness parameter, and a high-quality X-ray image is obtained, thereby providing a doctor with a more accurate diagnosis basis.

[0019] 3. The present application can extract a human body contour corresponding to a user from image data of the user, and update a corresponding part module in a standard model according to a length ratio between associated part key points in the human body contour corresponding to the user, so that a customized model highly consistent with actual physical characteristics of the user is obtained.

[0020] 4. When guiding the user to make corresponding posture adjustment through the customized model, the application can first adjust the model posture of the customized model to be fully synchronized with the current actual posture of the user, ensure that the customized model can truly reflect the current body posture of the user, compare the adjusted model posture with the pre-set standard posture, when the model posture is different from the standard posture, generate adjustment display data of adjusting the model posture to the standard posture, thereby being able to intuitively display the process of gradually adjusting from the current model posture to the standard posture, so that the user can clearly understand the posture adjustment action he needs to make by observing the visualized guidance information, thereby more accurately and quickly adjusting his posture to the standard posture, further ensure that the user is in the standard posture when taking X-ray, thereby improving the clarity and accuracy of imaging. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 is a flowchart of a ray equipment processing method based on image data provided by an embodiment of the application; Figure 2 is a schematic diagram of determining a first distance ratio provided by an embodiment of the application; Figure 3 is a schematic diagram of determining a first angle provided by an embodiment of the application; Figure 4 is a structural schematic diagram of a ray equipment processing device based on image data provided by an embodiment of the application. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be described below in a clear and complete manner with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the protection scope of the application.

[0023] The technical solutions of the application will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described in some embodiments.

[0024] Reference is made to Figure 1 is a ray equipment processing method based on image data provided by an embodiment of the application, Figure 1The execution subject of the method shown can be a software and / or hardware device. The execution subject of the present application can include, but is not limited to, at least one of the following: user equipment, network equipment, etc. Among them, the user equipment can include, but is not limited to, a computer, a smart phone, a personal digital assistant (PDA) and the above-mentioned electronic equipment, etc. The network equipment can include, but is not limited to, a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of computers or network servers based on cloud computing, wherein cloud computing is a kind of distributed computing, and a super virtual computer composed of a group of loosely coupled computers. The present embodiment does not make any limitation. It includes steps S1 to S4, as follows: S1, determining the body information corresponding to the user according to the image data of the user, and updating the standard model based on the body information to obtain a customized model corresponding to the user.

[0025] Among them, the image data refers to an image containing the body part information of the user, the body information refers to the parameters describing the size, shape and other characteristics of each part of the user's body extracted from the image data, the standard model refers to a pre-configured virtual human body model with universality, and the customized model refers to a virtual human body model corresponding to the user.

[0026] In actual application, the medical X-ray equipment can realize imaging of internal structure of human body according to different absorption and attenuation degrees of X-rays when penetrating different tissues of human body, and provide important diagnostic basis for medical diagnosis. For example, the tissue with higher density such as bone can absorb more X-rays, while the tissue such as muscle and fat relatively absorbs and attenuates less, thereby forming difference in intensity of X-rays after penetration, and when the medical X-ray equipment is used to take X-ray film of the user, inaccurate posture of the user can cause overlapping or shielding of tissue structure of the body part, thereby reducing accuracy of image imaging, and the incorrect posture can change the path of X-rays penetrating the human body, causing image distortion and distortion. For example, when the user takes X-ray film of the limbs, if the angle of the limbs is abnormally placed, the image of X-rays projected on the detector will be stretched or compressed, so that the length, angle and other measurements of the bone are inaccurate. The distorted image can mislead the doctor in judging the position of fracture and the direction of fracture line, thereby affecting the accuracy of diagnosis. Moreover, the X-rays penetrating the body of the user are captured by the detector installed on the C-shaped arm. The detector is various, and the common one is a digital flat panel detector. The digital flat panel detector can directly convert the energy of X-rays into an electrical signal, and then generate a digital image signal. The image data processed can be displayed on the display in real time, and the doctor can intuitively observe the internal anatomical structure of the user through the monitor. In this process, the radiation dose of X-rays also affects the accuracy of image imaging. When the radiation dose is too low, the number of X-ray photons reaching the detector can be insufficient, which can reduce the contrast and clarity of the image. When the radiation dose is too high, the image can be saturated, that is, the signal received by the detector is too strong, causing excessive exposure of the image in some areas, which appears as a white color, resulting in complete loss of tissue structure information in the corresponding area, thereby affecting the accuracy of image imaging.

[0027] The scheme can customize a virtual model matched with the user, guide and adjust the posture of the user in the X-ray shooting process through the virtual model, ensure that the posture of the user during imaging is as close as possible to the standard imaging posture, improve the accuracy of imaging, and obtain the brightness parameter of the target region of the human body which needs to be focused on through the medical X-ray equipment, adjust the radiation dose of the medical X-ray in real time according to the comparison result of the brightness parameter and the preset target brightness parameter, and ensure that the image brightness of the target region is always maintained near the target brightness parameter through the real-time and dynamic dose adjustment mechanism, so as to obtain high-quality X-ray images and provide more accurate diagnostic basis for doctors.

[0028] Specifically, during the X-ray shooting process, the user's posture may differ from the standard shooting posture, and the user's shooting posture needs to be adjusted accordingly. When adjusting, a customized model corresponding to the user's current posture can be generated. Subsequently, the posture adjustment process can be intuitively observed by the user through the corresponding adjustment of the customized model, so that the shooting posture can be quickly adjusted. When generating a customized model corresponding to the user, the user can be scanned by a 3D optical scanning device, such as a laser scanner, before X-ray examination. Image data of the user can be obtained, which can include the overall contour of the user's body and the detailed features of each part, such as the length and thickness of the user's legs. Through image processing techniques such as contour extraction techniques, body information such as part size and position relationship between parts can be extracted from the obtained image data. According to the corresponding body information of the user, the pre-configured standard model is updated accordingly, for example, if the length of the user's legs is different from the standard model, the length parameter of the legs in the standard model can be adjusted to obtain a customized model corresponding to the user.

[0029] In some embodiments, the specific implementation of step S1 can be: S11, extract the human body contour in the image data, and determine the corresponding contour specification according to the length in the length direction and the width direction.

[0030] Specifically, the human body contour corresponding to the user can be extracted from the image data, and the length in the length direction and the length in the width direction of the human body contour can be obtained. When determining the length in the length direction of the human body contour, the distance between the most distant pixel points in the length direction can be calculated to determine the corresponding length in the length direction. According to the length in the length direction, the height of the user can be roughly determined. When determining the length in the width direction of the human body contour, the distance between the most distant pixel points in the width direction can be calculated to determine the corresponding length in the width direction. According to the length in the length direction and the width direction, the specification size of the human body contour, i.e. the contour specification, can be determined. By specifying the contour specification, the customized model generated subsequently can better fit the actual body size of the user. The human body contour refers to the contour of the user extracted from the image data. The length direction can be the direction from the head to the feet of the human body, and the width direction can be the direction perpendicular to the length direction. The contour specification refers to the length, width, and other size of the human body contour of the user.

[0031] S12, retrieve the positioning template corresponding to the contour specification and the detection attribute of the user, and position the center point of the positioning template to the center point of the human body contour.

[0032] wherein the detection attribute refers to a preset attribute corresponding to the detection part and the detection purpose of the user, different combinations of the detection part and the detection purpose have different detection attributes, and the positioning template refers to a template that can position the part to be detected.

[0033] In actual application, according to the medical information of the user when the user visits, including the detection part such as the chest, limbs, head, abdomen, etc., and the detection purpose such as pneumonia diagnosis, fracture diagnosis, tumor screening, etc., the detection attribute of the user is determined, different detection parts and purposes correspond to different detection attributes, for example, when the detection part is the chest and the detection purpose is pneumonia diagnosis, the detection attribute focuses on the lung texture, inflammation area and other features, if the detection part is the limbs and the detection purpose is fracture diagnosis, the detection attribute focuses on the continuity of the skeleton, the position of the fracture line and other information.

[0034] After the contour specification and the detection attribute are obtained, the positioning template corresponding to the contour specification and the detection attribute of the user can be called, for example, for tumor screening of the head, the positioning template will focus on marking the position of each lobe of the brain, the direction of important blood vessels and nerves, etc., and for organ function evaluation of the abdomen, the positioning template will highlight the contours and relative positions of the main organs such as the liver, gallbladder and pancreas, after the corresponding positioning template is called, the center point of the positioning template can be aligned with the center point of the human body contour, which can ensure that the positioning template and the human body contour have a relatively accurate corresponding relationship in position, facilitating the positioning of the key area and the key point in the subsequent process.

[0035] S13, determining the key mapping area corresponding to the preset key area of the positioning template in the human body contour, and extracting the part key point in the key mapping area.

[0036] Specifically, some key areas are preset on the positioning template, i.e., preset key areas, the preset key areas are determined according to medical anatomy knowledge and detection requirements, therefore, the preset key areas in the positioning template are different for different detection attributes, for example, if the user needs to diagnose knee joint diseases, the preset key areas can be the areas where the structures such as bones, ligaments and the like around the knee joint are located, for the diagnosis of chest pneumonia, the preset key areas can include each lobe of the lung, high-risk areas where inflammation can occur and the like, since the center point of the positioning template has been aligned with the center point of the human body contour, the preset key areas on the positioning template are mapped to the human body contour according to the relative position relationship between the positioning template and the human body contour, to obtain corresponding key mapping areas, for example, when the preset key areas are the knee joint areas, the corresponding key mapping areas on the human body contour are also the knee joint areas, in the key mapping areas, the corresponding key points of the human body parts, i.e., part key points, can be extracted by using image processing and feature extraction algorithms, for different key mapping areas, different key points are extracted, for example, for the knee joint area, the corresponding joint points of the medial and lateral condyles of the knee joint and the tibial plateau can be extracted as the part key points.

[0037] The preset key areas refer to the key areas preset on the positioning template, the key mapping areas refer to the corresponding areas obtained after the preset key areas on the positioning template are mapped to the human body contour, and the part key points refer to the key points representing the characteristics of the corresponding human body parts extracted in the key mapping areas.

[0038] S14, obtaining the length proportion between the associated part key points, determining the target key points corresponding to the part key points in the standard model, and performing proportional updating on the part modules corresponding to the corresponding target key points according to the length proportion to obtain the customized model, the body information including the length proportion.

[0039] Specifically, the extracted part key points can be analyzed to find a plurality of part key points associated with each other, generally, the key points on adjacent joints or the same limb are associated with each other, for example, for the upper limb, the shoulder key point, the elbow key point and the wrist key point are associated with each other, the distances between the associated part key points are calculated, for example, the distance between the shoulder key point and the elbow key point and the distance between the elbow key point and the wrist key point can be calculated, according to the calculated distances between the key points, the length proportion between the associated part key points can be determined, for example, by ratio calculation of the distance between the shoulder key point and the elbow key point and the distance between the elbow key point and the wrist key point, the length proportion between the associated shoulder key point, elbow key point and wrist key point can be obtained.

[0040] After the length ratio is obtained, the human body can be divided into multiple part modules according to anatomical parts in the standard model, each part module corresponds to a group of associated key points, for example, the upper limb module includes shoulder, elbow and wrist key points, the lower limb module includes hip, knee and ankle key points, and the target key points corresponding to each part key point can be determined in the standard model. According to the length ratio, the part module corresponding to the corresponding target key point in the standard model is proportionally updated, for example, if the length ratio of the actual human body upper limb shoulder to elbow and elbow to wrist is different from the ratio in the standard model, the upper limb module of the standard model can be adjusted accordingly, so that the length ratio of the module is consistent with the actual human body. By proportionally updating the part module corresponding to each target key point, the standard model can be gradually adjusted to a customized model that better fits the user's physical characteristics. Wherein, the part module refers to the module corresponding to each part of the human body in the model.

[0041] Through the above implementation, a customized model highly consistent with the actual physical characteristics of the user can be obtained.

[0042] S2, obtain the posture image of the user, synchronize the model posture of the customized model according to the posture parameter of the posture image, and generate reference information based on the comparison result of the model posture and the standard posture.

[0043] Wherein, the posture image refers to an image obtained by image shooting of the current posture of the user, the posture parameter refers to the posture corresponding to the customized model, and the reference information refers to relevant indication information for guiding the user to adjust the posture to the standard posture.

[0044] After obtaining the customized model corresponding to the user, the real-time posture images of the user can be obtained from multiple angles by a special image acquisition device such as a camera when the user is preparing for X-ray shooting. These posture images can clearly record the actual posture of the user when X-ray shooting is performed, including the placement position of the limbs and other detailed information. The obtained posture images can be analyzed, and the posture parameters such as the positions of the joints and the angles of the body parts can be extracted by using image recognition and posture estimation techniques such as feature point detection techniques. The obtained posture parameters are applied to the customized model, and the model posture of the customized model is completely synchronized with the actual posture of the user by adjusting each part of the customized model, for example, if the user bends the left arm at a certain angle when shooting, the left arm part of the customized model can be adjusted accordingly, so that the bending angle of the left arm of the customized model is consistent with the actual bending angle of the user, and the customized model can accurately reflect the current body posture of the user.

[0045] After synchronizing the model posture of the customized model with the actual posture of the user, the synchronized model posture can be compared with a pre-set standard posture. The standard posture can be a rational posture that enables the medical X-ray device to obtain the best imaging effect. Different examination sites and examination purposes can correspond to different standard postures. For example, when performing chest frontal X-ray imaging, the standard posture can require the user to relax the shoulders and be isometric, naturally drop the arms, straighten the chest, and keep the head in the central position. When imaging the legs, the standard posture can require the user to straighten the legs and point the toes forward. If the model posture deviates from the standard posture, corresponding reference information can be generated according to the specific circumstances of the deviation. The reference information can be a visualized guide video that clearly shows the specific steps and action details for adjusting the current model posture to the standard posture.

[0046] The specific implementation of step S2 based on the above embodiment can be: S21, dividing the feature points with correlation attributes in the posture image into the same determination set.

[0047] Specifically, after obtaining the posture image corresponding to the user, a feature point detection algorithm, such as a Harris corner detection algorithm, can be used to extract a plurality of representative feature points from the human body contour in the posture image. These feature points usually correspond to the joint parts of the human body, such as the shoulder joint, elbow joint, knee joint, wrist joint, etc., which can effectively describe the posture information of the human body. According to the anatomical structure of the human body, the correlation attributes between the extracted feature points can be analyzed. For example, the feature points corresponding to the shoulder joint, elbow joint, and wrist joint on the same arm have obvious correlation attributes, and can be considered as belonging to the same body part. Therefore, the feature points with correlation attributes can be divided into the same determination set. The feature points in each determination set can represent a specific part of the human body, which facilitates subsequent analysis of the posture of each part of the human body. The correlation attribute refers to a characteristic that can reflect the mutual relationship between feature points. The feature points can be joint points of the human body, such as shoulder joint points, elbow joint points, etc. The determination set refers to a set formed by dividing the feature points with correlation attributes.

[0048] S22, obtaining the geometric quantization index corresponding to each feature point in the determination set and the target quantization index corresponding to the target point mapped with each feature point in the customized model.

[0049] Specifically, for each feature point in the determination set, a geometric quantification index corresponding to each feature point can be calculated, the geometric quantification index can include distance, angle and other information between feature points, for example, in the determination set including shoulder joint, elbow joint and wrist joint, the distance between the shoulder joint and the elbow joint, the distance between the elbow joint and the wrist joint, and the included angle between the upper arm and the forearm can be calculated, and the geometric quantification index can reflect the relative position and orientation of the body part corresponding to the determination set under the current posture of the human body. Since the customized model is a customized human body model corresponding to the user, in the customized model, the target points corresponding to each feature point in the posture image can be determined according to the corresponding relationship of human anatomy, for example, if the shoulder joint, elbow joint and wrist joint are extracted as feature points in the posture image, the corresponding shoulder joint, elbow joint and wrist joint can be found in the customized model as target points. For each target point in the customized model, the distance and angle between the target points can be calculated to obtain the target quantification index corresponding to the target points, and the target quantification index can reflect the relative position and orientation of the corresponding part of the customized model under the current posture.

[0050] wherein the geometric quantification index refers to an index that can quantify the positional relationship between each feature point in the determination set, which can include the distance and angle between each feature point, the target point refers to a point in the customized model corresponding to each feature point in the posture image, and the target quantification index refers to an index that can quantify the positional relationship between each target point in the customized model, which can include the distance between each target point.

[0051] In some embodiments, the specific implementation of step S22 can be: S221, calculating the first distance between adjacent feature points in the determination set, and determining the first distance ratio corresponding to the corresponding feature point according to the ratio between the first distances.

[0052] Specifically, the distance between adjacent feature points in the determination set, i.e. the first distance, can be obtained, and the first distance ratio corresponding to each feature point in the determination set can be determined according to the ratio between each first distance, as shown in Figure 2 , a schematic diagram for determining the first distance ratio provided by an embodiment of the present application, as shown in Figure 2As shown in , in the posture image, feature point 1, feature point 2 and feature point 3 can be divided into the same judgment set. In the judgment set, feature point 1 is adjacent to feature point 2, and the first distance between feature point 1 and feature point 2 can be obtained. Feature point 2 is adjacent to feature point 3, and the first distance between feature point 2 and feature point 3 can be obtained. The first distance between feature point 1 and feature point 2 and the first distance between feature point 2 and feature point 3 are calculated by ratio, and the first distance ratio corresponding to feature point 1, feature point 2 and feature point 3 can be obtained. By calculating the first distance ratio, the quantitative relationship between the relative distances of each feature point in the judgment set can be obtained. This quantitative relationship can reflect the morphological characteristics of the body part corresponding to the judgment set in the current posture.

[0053] The first distance refers to the distance between adjacent feature points in the determination set, and the first distance ratio refers to the ratio of the first distances between adjacent feature points.

[0054] S222 , sequentially obtaining a preset number of adjacent feature points as selected targets, calculating first angles corresponding to corresponding feature points according to vectors of the selected targets, and obtaining geometric quantization indicators according to the first distance ratio and the first angle.

[0055] The preset number may be 3, the selected targets refer to feature points that are adjacent in position, and the first angle refers to the angle between vectors corresponding to the adjacent selected targets.

[0056] Specifically, a preset number, for example, 3, of feature points adjacent to each other can be selected from the judgment set as selected targets, see Figure 3 , which is a schematic diagram of determining a first angle provided by an embodiment of the present invention, such as Figure 3 As shown in , three adjacent feature points can be selected as selected targets, namely selected target 1, selected target 2 and selected target 3. Vector 1 can be determined based on selected target 1 and selected target 2, and vector 2 can be determined based on selected target 2 and selected target 3. Since there is an angle between vector 1 and vector 2, the first angle corresponding to the feature points at three adjacent positions can be determined based on vector 1 and vector 2. The first angle can reflect the joint angle information of the corresponding part in the current posture. The geometric quantitative index corresponding to each feature point in the judgment set can be obtained based on the first distance ratio and the first angle.

[0057] When determining the vectors corresponding to adjacent selected targets, the coordinates corresponding to each contour point on the human body contour in the posture image can be obtained by coordinate processing of the posture image, and the coordinates of the feature points corresponding to the selected target can be obtained. According to the corresponding coordinates, the vectors corresponding to the adjacent selected targets can be determined.

[0058] S223, calculate the second distance between adjacent target points, and determine the second distance ratio of the adjacent target points according to the ratio between the second distances.

[0059] Specifically, in the customized model, the interval distance between adjacent target points, i.e., the second distance, can be calculated. By calculating the ratio of the second distances between adjacent target points, the second distance ratio of the adjacent target points can be obtained. For example, in the customized model, target point 1 and target point 2 are adjacent, and target point 2 and target point 3 are adjacent. Therefore, the second distance between target point 1 and target point 2, and the second distance between target point 2 and target point 3 can be obtained. By calculating the ratio of the second distance between target point 1 and target point 2 and the second distance between target point 2 and target point 3, the second distance ratio of target point 1, target point 2 and target point 3 can be obtained. By calculating the second distance ratio, the quantitative relationship of the relative distance between adjacent target points in the customized model can be obtained, which can reflect the morphological characteristics of the corresponding parts of the customized model in the current posture.

[0060] wherein the second distance refers to the distance between adjacent target points, and the second distance ratio refers to the ratio of the second distance of the adjacent target points.

[0061] S224, according to the vector of the target point corresponding to the selected target, calculate the second angle corresponding to the target point, and obtain the target quantitative index according to the second distance ratio and the second angle.

[0062] Specifically, in the customized model, the target points corresponding to the selected target in the posture image can be found, for example, target point 1, target point 2 and target point 3. The vectors corresponding to the adjacent target points can be determined according to the three-dimensional coordinate information of the adjacent target points in the customized model. The vector 1 can be determined according to the coordinates of target point 1 and target point 2, and the vector 2 can be determined according to the coordinates of target point 2 and target point 3. Then, the included angle between the two vectors can be calculated by using the included angle calculation formula of three-dimensional vector, such as the cosine theorem in three-dimensional space, to obtain the second angle corresponding to the target point. The second angle reflects the joint angle information of the corresponding parts in the customized model in the current posture.

[0063] When obtaining the three-dimensional coordinate information of the target points in the customized model, since when generating the customized model by modeling software, the modeling software will record the position information in the three-dimensional space every time a vertex, edge or face is added in the process of creating the model. For example, when creating a model simulating a human arm in 3dsMax, the modeling software will automatically record the three-dimensional coordinates (X, Y, Z) of the corresponding points in the data structure corresponding to the customized model when determining the position of the points at the joints of the arm. Therefore, the three-dimensional coordinate information of each target point can be determined by querying the modeling software when generating the customized model.

[0064] S23, determine the posture adjustment parameter of the corresponding part module in the customized model according to the difference between the geometric quantification index and the target quantification index, and update the model posture of the customized model according to the posture adjustment parameter, and the posture adjustment parameter includes the geometric quantification index.

[0065] Specifically, the difference between the geometric quantification index and the target quantification index, i.e. the quantification index difference, can be obtained, for example, the distance difference between the shoulder joint and the elbow joint in the posture image and the corresponding target points in the customized model, and the angle difference, etc. can be calculated to obtain the corresponding quantification index difference. According to the quantification index difference, the parameter for adjusting the posture of the corresponding part module in the customized model, i.e. the posture adjustment parameter, can be determined. The posture adjustment parameter can include the angle and the direction when adjusting the corresponding part module, etc. so that the model posture of the customized model can be updated accordingly according to the posture adjustment parameter.

[0066] Wherein, the quantification index difference refers to the difference between the geometric quantification index and the target quantification index, the part module refers to the module corresponding to the actual human body part in the customized model, and the posture adjustment parameter refers to the parameter for adjusting the posture of the corresponding part module.

[0067] In some embodiments, the specific implementation of step S23 can be: S231, obtain the proportion difference and the angle difference of the corresponding feature points and target points, and determine the target point whose proportion difference is greater than or equal to the proportion threshold and / or whose angle difference is greater than or equal to the angle threshold as the adjustment point.

[0068] Specifically, after obtaining the first distance ratio between the corresponding feature points and the second distance ratio between the corresponding target points, the difference between the first distance ratio and the second distance ratio can be calculated to obtain the ratio difference between the corresponding feature points and target points. After obtaining the first angle corresponding to the corresponding feature points and the second angle corresponding to the corresponding target points, the difference between the first angle and the second angle can be calculated to obtain the angle difference between the corresponding feature points and target points. If the ratio difference is greater than or equal to the ratio threshold, or the angle difference is greater than or equal to the angle threshold, the corresponding target point can be determined as the adjustment point. For example, when the ratio difference between the first distance ratio between the feature point 1, the feature point 2 and the feature point 3 and the second distance ratio between the target point 1, the target point 2 and the target point 3 is greater than or equal to the ratio threshold, the target point 1, the target point 2 and the target point 3 can be determined as the adjustment point. Or when the angle difference between the first angle corresponding to the feature point 1, the feature point 2 and the feature point 3 and the second angle corresponding to the target point 1, the target point 2 and the target point 3 is greater than or equal to the angle threshold, the target point 1, the target point 2 and the target point 3 can also be determined as the adjustment point. And if the ratio difference is greater than or equal to the ratio threshold at the same time, the angle difference is also greater than or equal to the angle threshold, in this case, the corresponding target point can also be determined as the adjustment point. Subsequently, the posture of the customized model can be adjusted according to the adjustment point.

[0069] Wherein, the ratio difference refers to the difference between the first distance ratio between the feature points and the second distance ratio between the target points, the angle difference refers to the difference between the first angle corresponding to the feature points and the second angle corresponding to the target points, the ratio threshold refers to the threshold for judging whether the ratio difference is too large, and the ratio difference is greater than or equal to the ratio threshold, which means that the ratio difference is considered to be too large, and the position of the target point needs to be adjusted. The angle threshold refers to the threshold for judging whether the angle difference is too large, and if the angle difference is greater than or equal to the angle threshold, it can be considered that the angle difference is too large, and the direction of the target point may need to be adjusted accordingly. The adjustment point is the target point in the customized model that needs to be adjusted in distance or angle.

[0070] S232, determining the adjustment direction of the part module corresponding to the adjustment point according to the bidirectional property of the angle difference. The bidirectional property includes positive property and negative property.

[0071] Since the first angle corresponding to the feature points in the posture image and the second angle corresponding to the target points in the customized model may have size difference, the calculated angle difference may be positive or negative, i.e. the bidirectional property of the angle difference. When the first angle is greater than the second angle, the obtained angle difference is a positive value, and the angle difference has a positive property. Conversely, when the first angle is less than the second angle, the angle difference is a negative value, and the angle difference has a negative property.

[0072] The adjustment direction of the part module corresponding to the adjustment node is determined according to the bidirectional attribute of the angle difference value. When the angle difference value has a positive attribute, that is, the first angle is greater than the second angle, it indicates that the angle of the corresponding part in the actual human body posture is greater than the angle of the corresponding part in the customized model. In order to make the customized model more fit the actual human body posture, the angle of the corresponding part in the customized model needs to be adjusted to be larger. In some cases, the adjustment direction of the part module corresponding to the adjustment node can be determined as the clockwise direction. For example, in a customized model with the front of the human body as the reference, when the actual bending angle of the arm is greater than the bending angle in the customized model, in order to adjust the model, the arm part module in the customized model may need to be rotated in the clockwise direction to increase the angle. Conversely, when the angle difference value has a negative attribute, it indicates that the angle of the corresponding part in the customized model is greater than the angle of the corresponding part in the actual human body posture. The part module needs to be adjusted in the direction of reducing the angle, and the adjustment direction can be counterclockwise.

[0073] The bidirectional attribute refers to the angle difference value having two possible properties, that is, positive or negative. The adjustment direction refers to the direction in which the part module in the customized model is rotated in order to adjust the customized model to better fit the actual human body posture. The positive attribute refers to the case where the angle difference value is positive, that is, the angle of the corresponding part in the actual human body posture is greater than the angle of the corresponding part in the customized model. The negative attribute refers to the case where the angle difference value is negative, that is, the angle of the corresponding part in the actual human body posture is less than the angle of the corresponding part in the customized model.

[0074] S233, determining the posture adjustment parameter corresponding to the part module based on the adjustment direction and the angle difference value, and updating the model posture of the customized model according to the posture adjustment parameter.

[0075] Specifically, the parameter for adjusting the posture of the corresponding part module, that is, the posture adjustment parameter, can be determined according to the adjustment direction and the angle difference value. The posture adjustment parameter mainly includes the angle size and direction of adjustment. The angle size of adjustment can be directly determined according to the angle difference value. For example, if the adjustment direction is determined to be the clockwise direction and the angle difference value is 20 degrees, the posture adjustment parameter of the corresponding part module can be clockwise rotation 20°. According to the posture adjustment parameter of the corresponding part module, the model posture of the customized model can be updated accordingly, so that the model posture corresponding to the customized model is more fitted to the actual posture of the user.

[0076] S24, when the model posture and the standard posture are different, generating adjustment display data from the model posture to the standard posture, and the reference information includes the adjustment display data.

[0077] Specifically, the model posture of the updated customized model can be compared with the preset standard posture. In the comparison process, the geometric quantification indexes of the model posture and the standard posture at each part, such as joint angle, limb distance, etc., can be compared. If there is a difference between the model posture and the standard posture, the display data for adjusting the model posture to the standard posture, that is, the adjustment display data, can be generated. The adjustment display data can intuitively show the process of gradually adjusting the current model posture to the standard posture, and can describe in detail the direction and amplitude of the adjustment of the corresponding part. The adjustment display data refers to the display data for adjusting the model posture to the standard posture.

[0078] Through the above embodiments, the process of gradually adjusting the current model posture to the standard posture can be intuitively shown, so that the user can clearly understand the posture adjustment action he needs to make by observing the visualized guidance information, thereby more accurately and quickly adjusting his posture to the standard posture.

[0079] In some embodiments, the specific implementation of step S24 can be: S241, determining the standard angles of each part module of the customized model in the standard posture, and comparing the current angles of each part module in the model posture with the standard angles.

[0080] In actual application, the angle change is often the most intuitive and most critical manifestation in the posture change. For example, in human posture, the angle change of the joint directly determines the form of the action, such as the lifting and bending of the arm, the stretching and bending of the leg, etc. By adjusting the angle and determining the posture of the customized model according to the angle, the key posture features can be focused on, and other factors can not interfere with the accuracy of the posture adjustment. By comparing the difference in angle, it can be quickly judged whether the posture of the customized model meets the standard posture.

[0081] Specifically, the customized model has a pre-configured corresponding standard angle for each part module in the standard posture. The current angles of each part module in the corresponding model posture of the customized model can be obtained, and the current angles of each part module are compared with the corresponding standard angles. The standard angle is the angle corresponding to each part module of the customized model in the standard posture, and the current angle is the angle corresponding to each part module of the customized model in the current model posture.

[0082] S242, if there is a part module with an absolute value of the difference between the current angle and the standard angle greater than the standard angle threshold, it is determined that the model posture is different from the standard posture, and the corresponding part module is determined as an abnormal module.

[0083] Specifically, the absolute value of the difference between the current angle and the corresponding standard angle of each part module can be obtained. If there is a part module with a difference absolute value greater than the standard angle threshold, it can be considered that the current posture of the part module is greatly different from the standard posture and has exceeded the normal fluctuation range. At this time, it can be determined that the model posture of the entire customized model is different from the standard posture, and the part module with the difference absolute value greater than the standard angle threshold can be determined as an abnormal module. The difference absolute value refers to the absolute value of the difference between the current angle and the standard angle, and the standard angle threshold refers to a threshold for determining whether the difference absolute value has reached an abnormal degree. If the difference absolute value is greater than the standard angle threshold, it can be considered that the corresponding difference absolute value is too large, and the difference between the current angle and the standard angle is too large. The abnormal module refers to the part module with the difference absolute value greater than the standard angle threshold.

[0084] S243, determine a calibration direction according to the calibration angle difference between the current angle and the standard angle, rotate the part module based on the calibration direction and the calibration angle difference to obtain a dynamic display video, and obtain the change displacement between the same target points in the part module during the rotation.

[0085] Specifically, for the determined abnormal module, the difference between its current angle and standard angle, i.e., the calibration angle difference, can be calculated. The calibration direction can be determined according to the positive and negative of the calibration angle difference. If the calibration angle difference is positive, it can be considered that the current angle is greater than the standard angle. In order to make the part module reach the standard posture, the calibration direction can be counterclockwise. If the calibration angle difference is negative, it can be considered that the current angle is less than the standard angle, and the calibration direction can be clockwise. For example, for the part module corresponding to the arm, if the calibration angle difference between the current angle and the standard angle is positive, it can be considered that the angle of the current arm is greater than the angle in the standard posture, and the arm part module needs to be rotated in the counterclockwise direction to calibrate the angle. According to the calibration angle difference and the calibration direction, the part module can be rotated accordingly. During the rotation, the posture of the customized model is recorded at certain time intervals, such as every 0.1 seconds, the posture change of the part module is recorded, and these posture changes are played continuously. The corresponding dynamic display video can be generated, which can intuitively show the process of adjusting the part module from the current posture to the standard posture, so that the user can clearly see the specific steps of the posture adjustment.

[0086] In the process of generating the dynamic display video, the position changes of the same target points in the part module during the rotation are recorded, the coordinate difference values of the same target points at different times are calculated, the change displacements between them are obtained, the change displacements can reflect the movement of each point in the part module during the adjustment process, and the length and direction of the indication mark are determined subsequently. The calibration angle difference refers to the difference between the current angle and the standard angle, the calibration direction refers to the direction when the posture of the part module is calibrated, the dynamic display video refers to the video that can be used to visually display the posture adjustment process, and the change displacement refers to the distance of the position changes of the same target points during the rotation.

[0087] S244, determining that the length corresponding to the maximum change displacement is the length of the indication mark, determining the direction of the indication mark according to the calibration direction, and obtaining the adjustment display data according to the indication mark and the dynamic display video.

[0088] Specifically, the change positions between the same target points in the obtained part module are analyzed, the maximum change displacement is determined, and the length corresponding to the maximum change displacement is determined as the length of the indication mark. The length can intuitively reflect the maximum position change amplitude of the part module during the adjustment process, so that the user can more clearly understand the degree of posture adjustment. The direction of the indication mark can be determined according to the calibration direction. For example, if the calibration direction is counterclockwise, the direction of the indication mark is also set to counterclockwise. If the calibration direction is clockwise, the direction of the indication mark is clockwise. In this way, the direction of the indication mark is consistent with the adjustment direction of the part module, which can more accurately guide the user to understand the direction during the posture adjustment. The determined indication mark and the dynamic display video can be combined to obtain the adjustment display data. The adjustment display data can present the adjustment process from the current model posture to the standard posture in a more intuitive way. The dynamic display video displays the specific adjustment steps and posture changes, and the indication mark highlights the adjustment direction and amplitude, so that the user can more clearly and accurately understand how to adjust the posture to the standard posture.

[0089] The indication mark refers to a mark used to highlight the relevant information during the posture adjustment in the dynamic display video, for example, an arrow mark.

[0090] Through the above implementation manner, it can be ensured that the user is in a standard posture when X-ray imaging is performed, so as to improve the clarity and accuracy of imaging and provide a more reliable basis for the doctor's diagnosis.

[0091] S3, starting the ray device based on the reference information completion information, and obtaining the brightness parameter of the target region according to the control module.

[0092] The completion information refers to information that the user completes the posture adjustment of the body part, the radiation device refers to a medical X-ray device, and the regulation and control module refers to a module configured on the medical X-ray device for controlling the dose of X-rays. The regulation and control module can realize automatic brightness control of the target region through a PID-based control algorithm. The target region can be a region that needs to be focused on. The brightness parameter refers to a numerical value that can reflect the intensity of X-rays received by the target region.

[0093] After obtaining the reference information, the user can adjust the posture according to the reference information. When the user completes the posture adjustment operation of all parts according to the reference information, the corresponding completion information can be sent. After receiving the completion information, the medical X-ray device can be started. The X-ray tube in the device can emit X-rays. The X-rays penetrate the user's body and start the imaging process. During the imaging process, the regulation and control module configured on the medical X-ray device can obtain the brightness parameter of the region that needs to be focused on, i.e., the target region. The target region can be a specific part of the user's body, such as a nodular region of the lung, a suspected fracture site of the bone, etc. The regulation and control module can monitor the brightness parameter of the target region in real time. The brightness parameter can reflect the intensity of X-rays received by the region and is closely related to the absorption and attenuation of X-rays by the tissue. The regulation and control module obtains the X-ray signal of the target region through the detector and converts it into a corresponding brightness value for recording and analysis. For example, when checking the lung, if the brightness parameter of the target region is high, it means that the region has less absorption of X-rays, which can be normal tissue of the lung. If the brightness value is low, it can indicate that the region has more absorption of X-rays, which can be a diseased region.

[0094] In some embodiments, the specific implementation of step S3 can be: S31, obtain a user's interactive request based on a user terminal. The interactive request includes a single interactive request and a synchronous interactive request.

[0095] The user terminal refers to a terminal held by the user, such as a mobile phone. The interactive request refers to the user's request for posture adjustment of the body part. The single interactive request refers to the user's request to adjust the posture of the body part one by one. The synchronous interactive request refers to the user's request to adjust the posture of multiple body parts at the same time.

[0096] Specifically, the user can send a corresponding posture adjustment request, i.e., an interaction request, through a user terminal such as a mobile phone held by the user. However, when the user has posture abnormalities in multiple parts, the user can choose to adjust the posture of each part one by one or to adjust the posture of multiple parts at a time. If the user wants to adjust the posture of each body part one by one, the user can send a single interaction request. For example, the user can want to adjust the posture of the arm part first and can send a single interaction request for the arm part through the user terminal. If the user wants to adjust the posture of multiple body parts at the same time, the user can send a synchronous interaction request to adjust the posture of multiple parts at the same time.

[0097] In S32, the target part selected by the user is determined according to the interaction request, a calibration image of the user is collected, a part contour corresponding to the target part in the calibration image is extracted, and a current posture of the target part is determined.

[0098] Specifically, according to the interaction request sent by the user, the target part selected by the user for posture adjustment can be determined. If the user sends a single interaction request, the part specified in the request can be directly determined as the target part selected by the user. If the user sends a synchronous interaction request, the synchronous interaction request can be analyzed to determine multiple target parts for posture adjustment. Then, the user can be photographed by a suitable image collection device such as a camera to collect a calibration image of the user. A part contour corresponding to the target part can be extracted from the calibration image by contour extraction technology such as edge detection technology. According to the extracted part contour, the current posture corresponding to the target part can be determined. The target part refers to the part selected by the user for posture adjustment. The calibration image refers to the image during posture adjustment. The part contour refers to the contour corresponding to the target part. The current posture refers to the posture corresponding to the target part.

[0099] In some embodiments, the specific implementation of step S32 can be: In S321, the abnormal module currently selected by the user is determined as the target part based on the single interaction request, and the abnormal range corresponding to the target part is determined according to the length range of the target part in the length direction and the width direction.

[0100] Specifically, when the part contour corresponding to the target part is obtained, the part of the human body corresponding to the abnormal module with posture abnormalities selected by the user can be determined as the target part according to the single interaction request. The abnormal range corresponding to the target part can be determined according to the length range of the target part in the length direction and the length range of the target part in the width direction. The length range includes the range of the target part in the length direction and the range of the target part in the width direction. The abnormal range refers to the range corresponding to the part with posture abnormalities.

[0101] S322, call the reference positioning template, determine the part reference point corresponding to the target part in the reference positioning template, take the part reference point as the center, and determine the area corresponding to the abnormal range in the part positioning area. The reference positioning template includes part reference points corresponding to multiple human body parts.

[0102] Specifically, the pre-configured reference positioning template can be called. The reference positioning template is constructed based on a large amount of human sample data and medical knowledge, and contains part reference points corresponding to multiple human body parts. These reference points are key reference points of each part of the human body. For example, the reference points of the shoulder may be specific anatomical landmark points such as acromion and greater tuberosity of humerus. In the reference positioning template, the part reference point corresponding to the target part can be located, for example, when the target part is the right shoulder, the part reference point related to the right shoulder such as the acromion point can be located in the reference positioning template. Taking the good part reference point as the center, combined with the determined abnormal range, the area corresponding to the abnormal range in the reference positioning template, i.e. the part positioning area, can be determined. The reference positioning template refers to a pre-configured template containing reference points corresponding to multiple human body parts. The part reference point refers to a key reference point of each part of the human body. The part positioning area refers to an area in the reference positioning template corresponding to the abnormal range.

[0103] S323, position the reference positioning template to the calibration image, and extract the part contour in the part positioning area in the calibration image.

[0104] Specifically, the reference positioning template can be positioned to the calibration image, and the part contour in the part positioning area can be extracted from the calibration image through the reference positioning template.

[0105] S324, if there is no complete part contour in the part positioning area, the part positioning area is enlarged according to a pre-configured magnification ratio until a complete part contour is extracted in the part positioning area.

[0106] Specifically, the part contour extracted from the part positioning area is checked to determine whether it is complete. The continuity, closeness and judgment of whether the extracted part contour contains the key features of the target part can be checked, for example, for the shoulder contour, if the key structures such as acromion and greater tuberosity of humerus are not completely displayed, the contour is considered to be incomplete. If there is no complete part contour in the part positioning area, the part positioning area can be enlarged according to the pre-configured magnification ratio, i.e. the pre-configured magnification ratio. When a complete part contour is extracted from the part positioning area, the enlargement operation of the part positioning area is stopped. The pre-configured magnification ratio refers to the pre-configured magnification ratio for enlarging the part positioning area.

[0107] S325, determine all abnormal modules as target parts according to the synchronous interaction request, and extract the part contours corresponding to the target parts in the calibration image.

[0108] Specifically, if the interaction request sent by the user is a synchronous interaction request, the human body parts corresponding to all abnormal modules can be determined as target parts, and then the part contours corresponding to each target part can be extracted from the calibration image.

[0109] S33, when the current poses of all target parts are equal to the standard pose, respond to the completion information of the reference information, and start the radiation device.

[0110] Specifically, when the current poses of all target parts are equal to the standard pose, it can be considered that the user has completed the pose adjustment operation of all parts according to the reference information, at this time, the completion information of the reference information can be responded to, and the radiation device can be started, and the X-ray tube in the device starts to emit X-rays, preparing for the imaging process.

[0111] Through the above implementation, it can be ensured that the user is in a standard pose when performing X-ray imaging examination, which is beneficial to obtain clear and accurate images and avoid imaging deviation caused by inaccurate pose.

[0112] S34, acquire the real-time image collected by the radiation device, extract the target region in the real-time image according to the detection template, and acquire the brightness parameter of the target region according to the regulation and control module.

[0113] Specifically, after the medical X-ray device is started, it emits X-rays to penetrate the user's body, and collects image information of the X-rays after passing through the body through the detector on the radiation device to generate a real-time image. According to the pre-set detection template, the real-time image is analyzed and processed. The detection template is designed according to different detection requirements and target parts, and contains related information and characteristics of the target region. Through image recognition and analysis algorithm, the target region matched with the detection template is extracted from the real-time image, for example, when detecting the lung, the detection template may mark the approximate range and some key structures of the lung. According to the detection template, the target region corresponding to the lung can be accurately extracted in the real-time image. The regulation and control module configured on the medical X-ray device can acquire the X-ray signal of the target region through the detector and convert it into the corresponding brightness value for recording to obtain the brightness parameter corresponding to the target region.

[0114] Among them, the real-time image refers to the image reflecting the internal structure of the user's body generated by the radiation device in the detection process, and the detection template refers to the template that can be used to accurately extract the target region corresponding to the detection target in the real-time image.

[0115] S4, according to the comparison result of the brightness parameter and the target brightness parameter, controlling the regulating module to adjust the radiation dose in real time.

[0116] Since the image brightness mainly depends on the energy of X-rays, the main factors affecting the energy of X-rays are the tube voltage and the current-time product, the tube voltage refers to the voltage value applied between the two poles of the X-ray tube, and the current-time product refers to the product of the current value applied between the two poles of the X-ray tube within a certain time, and the tube voltage determines the intensity of X-rays, that is, the penetration, and the current-time product can determine the density of X-ray energy. After imaging, the regulating module can obtain the brightness parameter of the target area, compare the brightness parameter with the target brightness parameter, if the brightness parameter is less than the target brightness parameter, the image brightness can be improved by increasing the tube voltage or the current-time product, if the brightness parameter is greater than the target brightness parameter, the image brightness can be improved by reducing the tube voltage or the current-time product.

[0117] After obtaining the brightness parameter, the brightness parameter can be compared with the target brightness parameter, the target brightness parameter is a desired brightness level preset according to the standards and experience of medical imaging for different examination sites and purposes, the target brightness parameter is to ensure that the target area can obtain the best image quality, so that the doctor can clearly observe the details of the tissue and the pathological condition, according to the comparison result of the brightness parameter and the target brightness parameter, the regulating module will automatically adjust the radiation dose in real time. If the brightness parameter is lower than the target brightness parameter, the regulating module will increase the emission dose of X-rays, so that more X-rays penetrate the target area, thereby improving the image brightness, on the contrary, if the brightness parameter is higher than the target brightness parameter, the regulating module will reduce the X-ray emission dose, reduce the intensity of X-rays reaching the detector, so that the image brightness is reduced, through this real-time and dynamic dose adjustment mechanism, the image brightness of the target area is always kept near the target brightness parameter, so as to obtain high-quality X-ray images, and provide accurate and clear diagnostic basis for doctors.

[0118] In actual application, when the regulation module regulates the X-ray dose, automatic brightness control of the target region can be realized through the PID-based control algorithm. The PID-based control algorithm is an algorithm for realizing accurate control of brightness through the combination of three parts of proportion, integral and differential. When the dose is regulated through proportional control, it is mainly to determine whether the difference between the brightness parameter and the target brightness parameter is too large. If the difference is too large, the X-ray dose needs to be increased, so that the X-ray dose intensity can be quickly increased to rapidly increase or decrease the image brightness. When there is a continuous small deviation in a single direction, such as positive deviation or negative deviation, proportional control cannot completely eliminate it. Integral control can be used to accumulate these small deviations over time. When the accumulated amount reaches a certain level, an increment is generated to eliminate this trend. The increment is the dose adjustment amount, which can be used to eliminate the deviation between the image brightness. Meanwhile, the present application includes a method for optimizing and improving the PID control, including anti-integral saturation, integral separation, variable integral, etc.

[0119] On the basis of the above embodiment, the specific implementation of step S4 can be: S41, obtain the brightness difference between the brightness parameter and the target brightness parameter. When the brightness difference is greater than the brightness difference threshold, determine the dose adjustment amount according to the brightness difference.

[0120] Specifically, the difference between the brightness parameter and the target brightness parameter, i.e. the brightness difference, can be obtained. The increment of the image brightness caused by the minimum step length of the tube voltage of the X-ray tube can be taken as the brightness difference threshold. The minimum step length of the tube voltage refers to the minimum change amount of the tube voltage of the X-ray tube that can be adjusted. Specifically, the tube voltage of the X-ray tube can be adjusted. By changing the tube voltage, the energy and output dose of the X-ray can be adjusted, thereby affecting the image brightness and other characteristics of the image. However, the adjustment of the tube voltage is not continuous and unlimited. It can only be changed in a certain fixed minimum unit, which is the minimum step length of the tube voltage. When determining the brightness difference threshold, the increment of the image brightness caused by the minimum step length of the tube voltage of the X-ray tube is taken as the brightness difference threshold. This is to consider the influence of the minimum step length of the tube voltage on the image brightness, so as to set a reasonable judgment standard for determining whether the difference between the brightness parameter and the target brightness parameter needs to be adjusted.

[0121] If the luminance difference value is greater than or equal to the luminance difference threshold value, it can be considered that the difference between the luminance parameter and the target luminance parameter is too large, and the X-ray dose needs to be adjusted accordingly, and at this time the PID control of the control module can be started, and the dose adjustment amount that needs to be adjusted can be determined according to the luminance difference value. When determining the dose adjustment amount, a preconfigured adjustment amount reference table can be retrieved. In the adjustment amount reference table, there are a plurality of preconfigured luminance difference value intervals, and each luminance difference value interval has a corresponding preset dose adjustment amount. The luminance difference value interval corresponding to the luminance difference value can be determined by traversing the plurality of luminance difference value intervals according to the luminance difference value between the luminance parameter and the target luminance parameter. The preset dose adjustment amount corresponding to the luminance difference value interval where the luminance difference value is located can be determined as the dose adjustment amount corresponding to the luminance difference value.

[0122] In some other embodiments, when the dose adjustment amount is obtained, the increment caused by the proportional factor can be obtained by proportional control in the PID control algorithm, the increment caused by the integral factor can be obtained by integral control, and the increment caused by the differential factor can be obtained by differential control. The sum of the three increments can obtain the dose adjustment amount.

[0123] The luminance difference value refers to the difference between the luminance parameter and the target luminance parameter, the luminance difference threshold value refers to a threshold value that can be used to determine whether the difference between the luminance parameter and the target luminance parameter is too large, and if the luminance difference value is greater than or equal to the luminance difference threshold value, it can be considered that the difference between the luminance parameter and the target luminance parameter is too large. The dose adjustment amount refers to the amount of adjustment of the X-ray dose.

[0124] S42, determine the adjustment fluctuation direction, the adjustment fluctuation direction includes the rising direction and the falling direction, and the updated X-ray dose is obtained by controlling the control module to add the dose adjustment amount based on the rising direction.

[0125] Specifically, after obtaining the dose adjustment amount, the direction of adjusting the X-ray dose can be determined, that is, the adjustment fluctuation direction. If the target luminance parameter is greater than the luminance parameter, it can be considered that the image luminance needs to be increased at this time, and the adjustment fluctuation direction is the rising direction at this time. If the target luminance parameter is less than the luminance parameter, it can be considered that the image luminance needs to be reduced at this time, and the adjustment fluctuation direction is the falling direction at this time. If the adjustment fluctuation direction is the rising direction, the tube voltage PID control of the control module can be started at this time, and the updated X-ray dose can be obtained by controlling the control module to add the dose adjustment amount to the current X-ray dose.

[0126] The adjustment fluctuation direction refers to the direction of adjusting the X-ray dose to change the image luminance, the rising direction refers to the direction of increasing the X-ray dose, and the falling direction refers to the direction of reducing the X-ray dose.

[0127] S43, according to the falling direction control regulation module subtracts the dose adjustment amount to obtain the updated radiation dose.

[0128] Specifically, if the adjustment fluctuation direction is the falling direction, at this time, the current time integral PID control of the regulation module can be controlled to start, the regulation module can be controlled to subtract the dose adjustment amount from the current radiation dose to obtain the updated radiation dose, and the X-ray tube assembly can be controlled to adjust the dose of X-rays in real time each time the regulation module obtains the updated radiation dose, and the brightness difference, the brightness difference threshold and other parameters in the scheme can be adjusted accordingly in actual tests.

[0129] Through the above embodiments, it can be ensured that the image brightness of the target region is always maintained near the target brightness parameter, so that a high-quality X-ray image is obtained, and accurate and clear diagnostic basis is provided for doctors.

[0130] Referring to Figure 4 , a structure schematic diagram of a ray equipment processing device based on image data provided by the embodiment of the present application, the ray equipment processing device based on image data comprises: The updating module is configured to determine user corresponding body information according to image data of the user, and update a standard model based on the body information to obtain a customized model corresponding to the user. The synchronization module is configured to acquire a posture image of the user, synchronize a model posture of the customized model according to posture information of the posture image, and generate reference information based on a comparison result of the model posture and a standard posture. The starting module is configured to start the ray equipment based on completion information of the reference information, and acquire a brightness parameter of a target region according to the regulation module. The control module is configured to control the regulation module to adjust the radiation dose in real time according to a comparison result of the brightness parameter and a target brightness parameter.

[0131] Figure 4 The device of the embodiment shown can be used to execute Figure 1 The steps in the method embodiment shown have similar implementation principles and technical effects, and will not be described here.

[0132] The present application also provides a readable storage medium, the readable storage medium has a computer program, the computer program is executed by the processor and is used for realizing the method provided by various embodiments.

[0133] Wherein, the readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transfer of computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general or special purpose computer. For example, the readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). In addition, the ASIC can be located in the user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in the communication device. The readable storage medium can be a read only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk and an optical data storage device, etc.

[0134] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for processing a ray device based on image data, characterized in that: include: Determine the user's corresponding body information based on the user's image data, and update the standard model based on the body information to obtain a customized model corresponding to the user; Obtaining a user's posture image, synchronizing a customized model posture according to the posture information of the posture image, and generating reference information based on a comparison result between the model posture and the standard posture; Based on the completion information of the reference information, the ray equipment is started, and the brightness parameters of the target area are obtained according to the control module; According to the comparison results of the brightness parameters and the target brightness parameters, the control and regulation module adjusts the radiation dose in real time.

2. The method according to claim 1, characterized in that Determining the user's corresponding body information based on the user's image data, and updating the standard model based on the body information to obtain a customized model corresponding to the user, including: Extracting the human body contour from the image data, and determining the corresponding contour specifications according to the length and width of the human body contour; Retrieving a positioning template corresponding to the contour specification and the user's detection attributes, and locating the center point of the positioning template to the center point of the human body contour; Determine the key mapping area corresponding to the preset key area of ​​the positioning template in the human body contour, and extract the key points of the parts in the key mapping area; Obtain the length ratio between each associated part key point, determine the target key point corresponding to the part key point in the standard model, and update the part module corresponding to the corresponding target key point according to the length ratio to obtain a customized model. The body information includes the length ratio.

3. The method according to claim 1, characterized in that Obtain the user's posture image, synchronize the customized model's posture according to the posture parameters of the posture image, and generate reference information based on the comparison results of the model posture and the standard posture, including: Divide the feature points with associated attributes in the posture image into the same judgment set; Obtain the geometric quantitative index corresponding to each feature point in the judgment set and the target quantitative index corresponding to the target point mapped to each feature point in the customized model; Determine, based on a quantitative index difference between the geometric quantitative index and the target quantitative index, a posture adjustment parameter of a corresponding part module in the customized model, and update the model posture of the customized model based on the posture adjustment parameter, wherein the posture adjustment parameter includes the geometric quantitative index; When the model posture is different from the standard posture, adjustment display data for adjusting the model posture to the standard posture is generated, and the reference information includes the adjustment display data.

4. The method according to claim 3, characterized in that Obtain the geometric quantitative index corresponding to each feature point in the judgment set and the target quantitative index corresponding to the target point mapped to each feature point in the customized model, including: Calculate the first distances between adjacent feature points in the determination set, and determine the first distance ratios corresponding to the corresponding feature points according to the ratios between the first distances; Sequentially obtaining a preset number of adjacent feature points as selected targets, calculating first angles corresponding to the corresponding feature points based on vectors of the selected targets, and obtaining geometric quantitative indicators based on the first distance ratio and the first angle; Calculating the second distances between adjacent target points, and determining the second distance ratios corresponding to the adjacent target points according to the ratio between the second distances; A second angle corresponding to the target point is calculated according to the vector of the target point corresponding to the selected target, and a target quantitative index is obtained according to the second distance ratio and the second angle.

5. The method according to claim 3, characterized in that According to the quantitative index difference between the geometric quantitative index and the target quantitative index, the posture adjustment parameters of the corresponding part module in the customized model are determined, and the model posture of the customized model is updated according to the posture adjustment parameters. The posture adjustment parameters include the geometric quantitative index, including: Obtaining a proportional difference and an angle difference between the corresponding feature point and the target point, and determining a target point whose proportional difference is greater than or equal to a proportional threshold and / or whose angle difference is greater than or equal to an angle threshold as an adjustment point; Determine the adjustment direction of the part module corresponding to the adjustment point according to the bidirectional attribute of the angle difference, the bidirectional attribute includes a positive attribute and a negative attribute; The posture adjustment parameters corresponding to the corresponding part module are determined based on the adjustment direction and angle difference, and the model posture of the customized model is updated according to the posture adjustment parameters.

6. The method according to claim 5, characterized in that When the model pose is different from the standard pose, the adjustment display data from the model pose to the standard pose is generated, including: Determine the standard angles corresponding to the modules of each part of the customized model in the standard posture, and compare the current angles of the modules of each part in the model posture with the standard angles; If there is a part module whose absolute value of the difference between the current angle and the standard angle is greater than the standard angle threshold, the model posture is determined to be different from the standard posture, and the corresponding part module is determined to be an abnormal module; The calibration direction is determined according to the calibration angle difference between the current angle and the standard angle. The part module is rotated based on the calibration direction and the calibration angle difference to obtain a dynamic display video, and the change displacement between the same target points in the part module during the rotation process is obtained. The length corresponding to the maximum change displacement is determined as the length of the indicator mark, and the direction of the indicator mark is determined according to the calibration direction. Adjustment display data is obtained according to the indicator mark and the dynamic display video.

7. The method according to claim 1, characterized in that Based on the completion information of the reference information, the ray device is started, and the brightness parameters of the target area are obtained according to the control module, including: Obtaining user interaction requests based on user terminals, where the interaction requests include single interaction requests and synchronous interaction requests; Determine the target part selected by the user according to the interactive request, collect the calibration image corresponding to the user, extract the part contour corresponding to the target part in the calibration image, and determine the current posture of the target part; When the current postures of all target parts are equal to the standard posture, respond to the completion information of the reference information and start the radiation device; The real-time image captured by the X-ray device is obtained, the target area in the real-time image is extracted according to the detection template, and the brightness parameter of the target area is obtained according to the control module.

8. The method according to claim 7, characterized in that Determine the target part selected by the user according to the interactive request, acquire the calibration image corresponding to the user, and extract the part contour corresponding to the target part in the calibration image, including: Based on a single interactive request, the abnormal module currently selected by the user is determined as the target part, and the corresponding abnormal range is determined according to the length range of the target part in the length direction and the width direction; Retrieve the reference positioning template, determine the part reference point corresponding to the target part in the reference positioning template, and determine the area corresponding to the abnormal range as the part positioning area with the part reference point as the center. The reference positioning template includes part reference points corresponding to multiple human body parts; Positioning the reference positioning template to the calibration image, and extracting the part contour located in the part positioning area in the calibration image; If there is no complete part contour in the part positioning area, the part positioning area is enlarged according to a preset magnification ratio until a complete part contour is extracted from the part positioning area; According to the synchronous interaction request, it is determined that all abnormal modules are target parts, and the part contour corresponding to the target part in the calibration image is extracted.

9. The method according to claim 1, characterized in that Based on the comparison results of the brightness parameters and the target brightness parameters, the control and regulation module adjusts the radiation dose in real time, including: Obtaining a brightness difference between the brightness parameter and the target brightness parameter, and when the brightness difference is greater than a brightness difference threshold, determining a dose adjustment amount according to the brightness difference; Determine the adjustment fluctuation direction, which includes an upward direction and a downward direction, and obtain the updated radiation dose based on the upward direction control module plus the dose adjustment amount; The updated radiation dose is obtained by subtracting the dose adjustment amount from the descending direction control module.

10. A radiographic device processing apparatus based on image data, characterized in that: include: An updating module, configured to determine the user's corresponding body information based on the user's image data, and update the standard model based on the body information to obtain a customized model corresponding to the user; A synchronization module is used to obtain a posture image of the user, synchronize the model posture of the customized model according to the posture information of the posture image, and generate reference information based on the comparison result of the model posture and the standard posture; A starting module is used to start the ray equipment based on the completion information of the reference information and obtain the brightness parameters of the target area according to the control module; The control module is used to control the regulation module to adjust the radiation dose in real time according to the comparison result of the brightness parameter and the target brightness parameter.

Images