Orthopedic image auxiliary detection method and system
Through orthopedic image-assisted detection methods, including image denoising, correction and feature analysis, the problem of orthopedic disease diagnosis in the prior art requires a large number of examinations, improve the accuracy and convenience of detection, and reduce the pain and examination costs of patients.
Patent Information
- Application Number
- CN202411811596.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-10
- Publication Date
- 2025-05-16
AI Technical Summary
In the prior art, medical personnel need to undergo a large number of medical examinations when diagnosing orthopedic diseases, which increases the pain and consultation costs of patients and is difficult to meet the examination needs.
An orthopedic image-assisted detection method is adopted, including determining the patient's initial orthopedic detection area, performing image shooting and denoising processing, correcting image data, analyzing bone characteristics and connection relationships, dividing blocks, calculating similarity, determining orthopedic abnormalities, and sending undetermined abnormalities to the expert platform for manual testing.
By improving image quality and accuracy, the pain and examination costs for patients are reduced, and the accuracy and convenience of detection are improved.
Smart Images
Figure CN120014311A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of orthopedic imaging detection, and in particular to an orthopedic imaging-assisted detection method and system. Background Art
[0002] The main component of bones is a large amount of calcium, which has a high density and a sharp natural contrast with the surrounding soft tissues. Orthopedic imaging is an important means for orthopedic doctors to diagnose bone, joint and soft tissue lesions. Through different imaging examination methods, doctors can clearly observe the structure, morphology and functional status of patients' bones, joints and soft tissues, providing strong support for the diagnosis and treatment of diseases.
[0003] In the prior art, when determining which orthopedic disease a patient suffers from, medical personnel, without much experience in diagnosis and treatment, often need to conduct a large number of medical examinations on the patient to obtain sufficient diagnosis and treatment data before accurately identifying the orthopedic disease the patient suffers from. These examination methods increase the patient's pain on the one hand, and increase the patient's consultation costs on the other, making it difficult to meet the examination needs. Summary of the invention
[0004] In order to solve the above technical problems, an orthopedic image-assisted detection method and system are provided. This technical solution solves the problem in the prior art proposed in the above background technology that, when determining the orthopedic disease suffered by a patient, medical personnel often need to conduct a large number of medical examinations on the patient without a lot of diagnosis and treatment experience, and only after obtaining sufficient diagnosis and treatment data can the orthopedic disease suffered by the patient be accurately identified. These examination methods increase the pain of the patient on the one hand, and increase the cost of the patient's consultation on the other hand.
[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is: An orthopedic image-assisted detection method, comprising: Determine the patient's initial orthopedic inspection area, and use medical imaging equipment to photograph the initial inspection area to obtain orthopedic images; Perform denoising on orthopedic images to obtain denoised orthopedic image data; Identify the denoised orthopedic image data, perform rectangular positioning and tilt angle correction on it, and obtain the rectangular orthopedic image data; Analyze the bone characteristics and bone connection relationship of the rectangular orthopedic image data to determine the bone distribution characteristics, divide the blocks based on the bone distribution characteristics, and obtain multiple block image information; Retrieve the patient orthopedic abnormality types that have been counted in the detection system, and perform similarity calculation on multiple block image information; The calculated similarity values are analyzed to determine whether there are orthopedic abnormalities in the patient; The imaging data of orthopedic abnormality types that cannot be determined in the system are sent to the expert platform, and the imaging data are manually inspected by experts.
[0006] Preferably, the denoising process of the orthopedic image to obtain denoised orthopedic image data specifically includes the following steps: Decompose high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.
[0007] Preferably, the step of identifying the orthopedic denoised image data and performing rectangular positioning and tilt angle correction on the orthopedic denoised image data specifically comprises the following steps: Perform edge detection on the image in the horizontal and vertical directions, and sum the detection results in the horizontal and vertical directions; The orthopedic part of the image is processed based on the first morphological closing operation to obtain the access area, and then the single line interference on the image is removed by the second morphological closing operation; Fill the holes of the image after the first morphological closing operation; Calculate the minimum bounding rectangle according to each white area in the image after the holes are filled, and obtain information of each bounding rectangle, wherein the rectangle information includes coordinates of the four vertices of the rectangle, an inclination angle, and the number of white pixels in the rectangular area; According to the calculated rectangular tilt angles, their distribution within the right angle range is counted, and the angles with more concentrated distribution are used to perform preliminary tilt angle correction.
[0008] Preferably, the calculation formula of the minimum enclosing rectangle is:
[0009] In the formula, is the area of several groups of bounding rectangles.
[0010] Preferably, the obtaining of rectangular orthopedic image data specifically comprises the following steps: In the image after the initial tilt angle correction, the orthopedic area is located in the zero degree direction or the ninety degree direction, and the error is within the range of plus or minus ten degrees; Screen candidate rectangles of the orthopedic region at the zero-degree and ninety-degree directions respectively to determine the final position and final angle of the orthopedic region; A final rotation is performed to obtain the final corrected image of the orthopedic area.
[0011] Preferably, performing bone feature and bone connection relationship analysis on the rectangular orthopedic image data to determine the bone distribution features specifically includes the following steps: Analyze the bone characteristics and bone connection relationship based on the rectangular orthopedic images to determine the bone position, bone characteristics and bone connection relationship; Based on the bone positions and the bone features, determining the similarity of bone features at adjacent positions; The bones with high similarity are regarded as a region, the segmentation nodes are determined, and the segmentation results are determined; According to the skeletal connection relationship, the integrity of the skeletal connection relationship is analyzed for the segmentation result, and the segmentation result is corrected to obtain the plurality of block image information.
[0012] Preferably, the formula for calculating the similarity is:
[0013] In the formula, is the similarity between the output patient orthopedic type and the i-th feature of the patient orthopedic abnormality type that has been counted, is the jth feature index value of the i-th feature of the output patient orthopedic category, is the jth index value of the i-th feature of the patient's orthopedic abnormality type that has been counted, is the total number of feature index values of the i-th feature.
[0014] Preferably, the step of analyzing the calculated similarity value to determine whether the patient has orthopedic abnormalities specifically comprises the following steps: Compare the calculated similarity value with a preset threshold in the system; If the similarity value is greater than or equal to the preset threshold, the patient's orthopedic abnormality type that has been counted is directly matched and output; If the similarity value is less than the preset threshold, the image data of the orthopedic abnormality type is sent to the expert platform.
[0015] Preferably, the manual detection of the image data by experts specifically includes the following steps: Pack the image data of orthopedic abnormality types that cannot be determined in the system to form a re-examination orthopedic data package; Retrieving the data of all orthopedic doctors in the hospital, wherein the data of orthopedic doctors includes the doctors' diagnosis and treatment experience; Retrieve the contact information of the doctor with the most experience in diagnosis and treatment from the system; Based on the contact information, the re-examination orthopedic data package was sent to the physician with the most experience in diagnosis and treatment, who then manually inspected the imaging data.
[0016] An orthopedic image-assisted detection system, comprising: An orthopedic image acquisition module, which is used to determine the patient's initial orthopedic detection area and photograph the initial detection area through a medical imaging detection device to acquire an orthopedic image; A data processing module, wherein the data processing module internally integrates a first data processing unit and a second data processing unit, wherein the first data processing unit is used to perform denoising on orthopedic images to obtain denoised orthopedic image data, and the second data processing unit is used to identify the denoised orthopedic image data, and perform rectangular positioning and tilt angle correction on the orthopedic image data to obtain regular rectangular orthopedic image data; A data analysis module, the data analysis module is used to analyze the bone characteristics and bone connection relationship of the rectangular orthopedic image data, determine the bone distribution characteristics, and divide the blocks based on the bone distribution characteristics to obtain multiple block image information; A calculation module, the calculation module is used to retrieve the types of orthopedic abnormalities of patients that have been counted in the detection system, and perform similarity calculation on multiple block image information; The detection result determination module is used to analyze the calculated similarity value, determine whether the patient has orthopedic abnormalities, and send the image data of the orthopedic abnormality type that cannot be determined in the system to the expert platform for manual detection of the image data by experts.
[0017] Compared with the prior art, the present invention provides an orthopedic image-assisted detection method and system, which has the following beneficial effects: The present invention denoises orthopedic images to reduce or eliminate noise in orthopedic images, thereby improving the quality of orthopedic images, and then performs rectangular positioning and tilt angle correction on orthopedic images to make the data referenceable and researchable, performs bone feature and bone connection relationship analysis on orthopedic images, determines bone distribution features, performs block division based on the bone distribution features, obtains multiple block image information, calculates similarity between the image information and the statistically analyzed orthopedic abnormality types of patients, and the orthopedic abnormality type corresponding to a larger similarity value is the orthopedic disease of the patient, and a smaller similarity indicates that the system cannot determine the type of orthopedic abnormality and further manual re-examination is required. The present invention selects the most experienced doctor for re-examination, thereby improving the accuracy and convenience of detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 It is a schematic diagram of the orthopedic image-assisted detection method of the present invention; Figure 2 A schematic diagram of a method for obtaining orthopedic denoised image data in the present invention; Figure 3 A schematic diagram of a method for rectangular positioning and tilt angle correction in the present invention; Figure 4 A schematic diagram of a method for obtaining rectangular orthopedic image data in the present invention; Figure 5 A schematic diagram of the method for determining bone distribution characteristics in the present invention; Figure 6 A schematic diagram of a method for determining whether a patient has orthopedic abnormalities in the present invention; Figure 7 It is a schematic diagram of the method for manually detecting the image data by experts in the present invention. DETAILED DESCRIPTION
[0019] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.
[0020] Example 1 Please refer to Figure 1-Figure 7 As shown, an orthopedic image-assisted detection method comprises: Determine the patient's initial orthopedic inspection area, and use medical imaging equipment to photograph the initial inspection area to obtain orthopedic images; Perform denoising on orthopedic images to obtain denoised orthopedic image data; Identify the denoised orthopedic image data, perform rectangular positioning and tilt angle correction on it, and obtain the rectangular orthopedic image data; Analyze the bone characteristics and bone connection relationship of the rectangular orthopedic image data, determine the bone distribution characteristics, divide the blocks based on the bone distribution characteristics, and obtain multiple block image information; Retrieve the patient orthopedic abnormality types that have been counted in the detection system, and perform similarity calculation on multiple block image information; The calculated similarity values are analyzed to determine whether there are orthopedic abnormalities in the patient; The imaging data of orthopedic abnormality types that cannot be determined in the system are sent to the expert platform, and the imaging data are manually inspected by experts.
[0021] It can be understood by those skilled in the art that by denoising orthopedic images, the noise in orthopedic images can be reduced or eliminated, thereby improving the quality of orthopedic images, and then rectangular positioning and tilt angle correction can be performed on orthopedic images to make the data referenceable and researchable, and the bone features and bone connection relationships of orthopedic images can be analyzed to determine the bone distribution features, and block division can be performed based on the bone distribution features to obtain multiple block image information, and the similarity between the image information and the statistically analyzed orthopedic abnormality types of patients can be calculated. The type of orthopedic abnormality corresponding to the larger similarity value is the patient's orthopedic disease, and the smaller similarity indicates that the system cannot determine the type of orthopedic abnormality and further manual re-examination is required. The present invention selects the most experienced doctor for re-examination, thereby improving the accuracy and convenience of detection.
[0022] De-noising orthopedic images and obtaining orthopedic denoised image data specifically include the following steps: Decompose high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.
[0023] It can be understood by those skilled in the art that, through wavelet transform, the high-frequency information and low-frequency information of the image are processed separately, and the low-frequency components are restored using the LPG-PCA method. In the process of restoring the high-frequency components, the natural image is sparsely represented in the group domain, and the inherent local sparsity and non-local self-similarity of the natural image are clearly and effectively characterized in a unified manner, and the high-frequency part is extracted from the image without being affected by noise. While effectively removing noise, the image details are retained, high sparsity and high restoration quality are achieved, and the visual performance of the image as well as the peak signal-to-noise ratio and structural similarity are improved.
[0024] The steps of identifying the denoised orthopedic image data and performing rectangular positioning and tilt angle correction are as follows: Perform edge detection on the image in the horizontal and vertical directions, and sum the detection results in the horizontal and vertical directions; The orthopedic part of the image is processed based on the first morphological closing operation to obtain the access area, and then the single line interference on the image is removed by the second morphological closing operation; Fill the holes of the image after the first morphological closing operation; Calculate the minimum bounding rectangle according to each white area in the image after the holes are filled, and obtain the information of each bounding rectangle, which includes the coordinates of the four vertices of the rectangle, the tilt angle, and the number of white pixels in the rectangular area; According to the calculated rectangular tilt angles, their distribution within the right angle range is counted, and the angles with more concentrated distribution are used to perform preliminary tilt angle correction.
[0025] The calculation formula for the minimum enclosing rectangle is:
[0026] In the formula, is the area of several groups of bounding rectangles.
[0027] The steps of obtaining rectangular orthopedic image data are as follows: In the image after the initial tilt angle correction, the orthopedic area is located in the zero degree direction or the ninety degree direction, and the error is within the range of plus or minus ten degrees; Screen candidate rectangles of the orthopedic region at the zero-degree and ninety-degree directions respectively to determine the final position and final angle of the orthopedic region; A final rotation is performed to obtain the final corrected image of the orthopedic area.
[0028] Analyzing the bone features and bone connection relationships of the rectangular orthopedic image data to determine the bone distribution features specifically includes the following steps: Analyze the bone characteristics and bone connection relationship based on the rectangular orthopedic images to determine the bone position, bone characteristics and bone connection relationship; Based on the bone positions and bone features, the similarity of bone features at adjacent positions is determined; The bones with high similarity are regarded as a region, the segmentation nodes are determined, and the segmentation results are determined; According to the skeletal connection relationship, the integrity of the skeletal connection relationship is analyzed for the segmentation results, and the segmentation results are corrected to obtain multiple block image information.
[0029] The formula for calculating similarity is:
[0030] In the formula, is the similarity between the output patient orthopedic type and the i-th feature of the patient orthopedic abnormality type that has been counted, is the jth feature index value of the i-th feature of the output patient orthopedic category, is the jth index value of the i-th feature of the patient's orthopedic abnormality type that has been counted, is the total number of feature index values of the i-th feature.
[0031] The calculated similarity values are analyzed to determine whether the patient's orthopedic abnormalities exist, specifically including the following steps: Compare the calculated similarity value with a preset threshold in the system; If the similarity value is greater than or equal to the preset threshold, the patient's orthopedic abnormality type that has been counted is directly matched and output; If the similarity value is less than the preset threshold, the image data of the orthopedic abnormality type is sent to the expert platform.
[0032] The manual detection of the image data by experts specifically includes the following steps: Pack the image data of orthopedic abnormality types that cannot be determined in the system to form a re-examination orthopedic data package; Retrieve the data of all orthopedic doctors in the hospital, including the doctors' diagnosis and treatment experience; Retrieve the contact information of the doctor with the most experience in diagnosis and treatment from the system; Based on the contact information, the re-examination orthopedic data package was sent to the physician with the most experience in diagnosis and treatment, who then manually inspected the imaging data.
[0033] An orthopedic image-assisted detection system, comprising: The orthopedic image acquisition module is used to determine the patient's initial orthopedic detection area and to capture the initial detection area through medical imaging detection equipment to obtain orthopedic images; A data processing module, wherein a first data processing unit and a second data processing unit are integrated inside the data processing module, wherein the first data processing unit is used to perform denoising on orthopedic images to obtain denoised orthopedic image data, and the second data processing unit is used to identify the denoised orthopedic image data, and perform rectangular positioning and tilt angle correction on the denoised orthopedic image data to obtain regular rectangular orthopedic image data; A data analysis module is used to analyze the bone characteristics and bone connection relationship of the rectangular orthopedic image data, determine the bone distribution characteristics, divide the blocks based on the bone distribution characteristics, and obtain multiple block image information; A calculation module, which is used to retrieve the types of orthopedic abnormalities of patients that have been counted in the detection system and perform similarity calculation on multiple block image information; The detection result determination module is used to analyze the calculated similarity value, determine whether there is any orthopedic abnormality in the patient, and send the image data of the orthopedic abnormality type that cannot be determined in the system to the expert platform for manual detection of the image data by experts.
[0034] The working principle and use process of this device: The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.
Claims
1. An orthopedic image-assisted detection method, characterized in that: include: Determine the patient's initial orthopedic inspection area, and use medical imaging equipment to photograph the initial inspection area to obtain orthopedic images; Perform denoising on orthopedic images to obtain denoised orthopedic image data; Identify the denoised orthopedic image data, perform rectangular positioning and tilt angle correction on it, and obtain the rectangular orthopedic image data; Analyze the bone characteristics and bone connection relationship of the rectangular orthopedic image data to determine the bone distribution characteristics, divide the blocks based on the bone distribution characteristics, and obtain multiple block image information; Retrieve the patient orthopedic abnormality types that have been counted in the detection system, and perform similarity calculation on multiple block image information; The calculated similarity values are analyzed to determine whether there are orthopedic abnormalities in the patient; The imaging data of orthopedic abnormality types that cannot be determined in the system are sent to the expert platform, and the imaging data are manually inspected by experts.
2. The orthopedic image-assisted detection method according to claim 1, characterized in that: The denoising of the orthopedic image to obtain denoised orthopedic image data specifically includes the following steps: Decompose high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.
3. The orthopedic image-assisted detection method according to claim 2, characterized in that: The identification of orthopedic denoised image data and the rectangular positioning and tilt angle correction thereof specifically include the following steps: Perform edge detection on the image in the horizontal and vertical directions, and sum the detection results in the horizontal and vertical directions; The orthopedic part of the image is processed based on the first morphological closing operation to obtain the access area, and then the single line interference on the image is removed by the second morphological closing operation; Fill the holes of the image after the first morphological closing operation; Calculate the minimum bounding rectangle according to each white area in the image after the holes are filled, and obtain information of each bounding rectangle, wherein the rectangle information includes coordinates of the four vertices of the rectangle, an inclination angle, and the number of white pixels in the rectangular area; According to the calculated rectangular tilt angles, their distribution within the right angle range is counted, and the angles with more concentrated distribution are used to perform preliminary tilt angle correction.
4. The orthopedic image-assisted detection method according to claim 3, characterized in that: The calculation formula of the minimum enclosing rectangle is: In the formula, is the area of several groups of bounding rectangles.
5. The orthopedic image-assisted detection method according to claim 4, characterized in that: The obtaining of rectangular orthopedic image data specifically comprises the following steps: In the image after the initial tilt angle correction, the orthopedic area is located in the zero degree direction or the ninety degree direction, and the error is within the range of plus or minus ten degrees; Screen candidate rectangles of the orthopedic region at the zero-degree and ninety-degree directions respectively to determine the final position and final angle of the orthopedic region; A final rotation is performed to obtain the final corrected image of the orthopedic area.
6. The orthopedic image-assisted detection method according to claim 5, characterized in that: The bone characteristics and bone connection relationship analysis of the rectangular orthopedic image data to determine the bone distribution characteristics specifically includes the following steps: Analyze the bone characteristics and bone connection relationship based on the rectangular orthopedic images to determine the bone position, bone characteristics and bone connection relationship; Based on the bone positions and the bone features, determining the similarity of bone features at adjacent positions; The bones with high similarity are regarded as a region, the segmentation nodes are determined, and the segmentation results are determined; According to the skeletal connection relationship, the integrity of the skeletal connection relationship is analyzed for the segmentation result, and the segmentation result is corrected to obtain the plurality of block image information.
7. The orthopedic image-assisted detection method according to claim 6, characterized in that: The formula for calculating the similarity is: In the formula, is the similarity between the output patient orthopedic type and the i-th feature of the patient orthopedic abnormality type that has been counted, is the jth feature index value of the i-th feature of the output patient orthopedic category, is the jth index value of the i-th feature of the patient's orthopedic abnormality type that has been counted, is the total number of feature index values of the i-th feature.
8. The orthopedic image-assisted detection method according to claim 7, characterized in that: The analysis of the calculated similarity value to determine whether the patient's orthopedic abnormality exists specifically includes the following steps: Compare the calculated similarity value with a preset threshold in the system; If the similarity value is greater than or equal to the preset threshold, the patient's orthopedic abnormality type that has been counted is directly matched and output; If the similarity value is less than the preset threshold, the image data of the orthopedic abnormality type is sent to the expert platform.
9. The orthopedic image-assisted detection method according to claim 8, characterized in that: The manual detection of the image data by experts specifically includes the following steps: Pack the image data of orthopedic abnormality types that cannot be determined in the system to form a re-examination orthopedic data package; Retrieving the data of all orthopedic doctors in the hospital, wherein the data of orthopedic doctors includes the doctors' diagnosis and treatment experience; Retrieve the contact information of the doctor with the most experience in diagnosis and treatment from the system; Based on the contact information, the re-examination orthopedic data package was sent to the physician with the most experience in diagnosis and treatment, who then manually inspected the imaging data.
10. An orthopedic image-assisted detection system, used to implement an orthopedic image-assisted detection method as claimed in any one of claims 1 to 9, characterized in that: include: An orthopedic image acquisition module, which is used to determine the patient's initial orthopedic detection area and photograph the initial detection area through a medical imaging detection device to acquire an orthopedic image; A data processing module, wherein the data processing module internally integrates a first data processing unit and a second data processing unit, wherein the first data processing unit is used to perform denoising processing on orthopedic images to obtain denoised orthopedic image data, and the second data processing unit is used to identify the denoised orthopedic image data, and perform rectangular positioning and tilt angle correction on the orthopedic image data to obtain regular rectangular orthopedic image data; A data analysis module, the data analysis module is used to analyze the bone characteristics and bone connection relationship of the rectangular orthopedic image data, determine the bone distribution characteristics, and divide the blocks based on the bone distribution characteristics to obtain multiple block image information; A calculation module, the calculation module is used to retrieve the types of orthopedic abnormalities of patients that have been counted in the detection system, and perform similarity calculation on multiple block image information; The detection result determination module is used to analyze the calculated similarity value, determine whether the patient has orthopedic abnormalities, and send the image data of the orthopedic abnormality type that cannot be determined in the system to the expert platform for manual detection of the image data by experts.