A bone joint replacement surgery auxiliary system

Abstract

The present invention relates to the field of medical computer-aided technology, and specifically to a bone joint replacement surgery auxiliary system. The present invention screens effective X-ray images; performs three-dimensional reconstruction on the effective X-ray images to obtain a three-dimensional model of the bone joint, obtains an abnormality index based on the position distribution difference and quantity of coordinate points in each spatial area of ​​the three-dimensional model of the bone joint, and selects a spatial abnormal area based on the abnormality index; then, obtains a construction misalignment rate based on the unevenness and abnormality index of the spatial abnormal area, and the degree of separation between the spatial abnormal area and its adjacent area, and adjusts the three-dimensional model of the bone joint in combination with the position distribution difference of the coordinate points in the spatial abnormal area, and assists bone joint replacement surgery based on the adjusted model. The present invention only adjusts the spatial abnormal area caused by the construction error in the three-dimensional reconstruction process, reduces the error between the adjusted bone joint model and the real bone joint, and effectively improves the auxiliary effect on bone joint replacement surgery.

Description

Technical Field

[0001] The present invention relates to the field of medical computer-aided technology, and in particular to a bone joint replacement surgery auxiliary system. Background Art

[0002] When a patient's joints lose their normal function due to severe trauma or chronic infection, joint replacement can usually be used for treatment. Joint replacement is a bone repair surgery that uses artificial joints to replace the original damaged joints. It can effectively improve the pain, movement disorders and other discomfort caused by joint injuries, and has a positive significance for improving the quality of life of patients.

[0003] Three-dimensional reconstruction technologies, such as three-dimensional image fusion technology, deep learning methods and geometric model methods, fuse multiple two-dimensional images into a three-dimensional model, allowing doctors to more intuitively observe and analyze the spatial position, size, severity and relationship with surrounding tissues of the lesion, thereby improving the accuracy of diagnosis and the effectiveness of treatment.

[0004] The bone joints that have been reconstructed in three dimensions may be abnormal due to the following reasons: abnormal construction during the three-dimensional reconstruction process, abnormal lesions such as bone spurs and tumors growing on the bone surface, which will affect the subsequent doctor's operational judgment during bone and joint replacement surgery, so the bone and joint model needs to be adjusted. The existing technology usually adjusts the bone and joint model according to the difference between the bone and joint model that has been reconstructed in three dimensions and the standard bone and joint model; however, due to diseases, bone spurs and tumors growing on the bone surface cause the patient's own bone abnormalities, and the bone abnormalities caused by the lesions are also different from the standard bone and joint model. Directly adjusting the bone and joint model based on the standard bone and joint model is likely to ignore the abnormalities of the bones themselves, so that there is an error between the adjusted bone and joint model and the patient's actual bone and joint, and the auxiliary effect of bone and joint replacement surgery is poor. Summary of the invention

[0005] In order to solve the technical problem that the adjustment effect of the bone joint model is poor due to the location of the bone joint lesions, and the auxiliary effect of the bone joint replacement surgery is poor, the purpose of the present invention is to provide a bone joint replacement surgery auxiliary system, and the technical solution adopted is as follows:

[0006] The present invention provides a bone joint replacement surgery auxiliary system, the system comprising:

[0007] A data acquisition module, used for acquiring at least two X-ray images of the patient's bone joints to be tested;

[0008] An image screening module is used to divide the X-ray image into sub-regions, and screen the effective X-ray image according to the difference in the overall direction between the sub-region of each X-ray image and its adjacent region, as well as the grayscale uniformity of the sub-region;

[0009] An analysis module is constructed to perform three-dimensional reconstruction of the effective X-ray image to obtain a three-dimensional model of the bone joint; the coordinate points on the three-dimensional model of the bone joint are divided into different spatial regions, and the abnormal index of each spatial region is obtained according to the position distribution difference and the number of the coordinate points in each spatial region, and the spatial abnormal region is selected based on the abnormal index; the construction dislocation rate of each spatial abnormal region is obtained according to the unevenness of each spatial abnormal region and the abnormal index, as well as the degree of separation of each spatial abnormal region and its adjacent region;

[0010] The model adjustment module is used to adjust the three-dimensional model of the bone joint according to the position distribution difference between each coordinate point in the spatial abnormal area and the remaining coordinate points and the constructed dislocation rate, and assist the bone joint replacement surgery based on the adjusted three-dimensional model of the bone joint.

[0011] Furthermore, the screening of effective X-ray images includes:

[0012] For each sub-region of each X-ray image, the principal eigenvector obtained by principal component analysis of the sub-region is recorded as the overall direction vector of the sub-region; the mean of the angles between the sub-region and the overall direction vectors of all its adjacent regions is calculated as the initial occlusion index of the sub-region; the edge pixel points on the edge of the adjacent region are located on the edge of the sub-region;

[0013] The accumulated sum of the absolute values ​​of the grayscale values ​​of each pixel in the sub-region and the pixels in its preset neighborhood range is used as the grayscale difference index of each pixel in the sub-region; the accumulated sum of the grayscale difference indexes of all pixels in the sub-region is calculated to obtain the grayscale uniformity index of the sub-region;

[0014] The accumulated grayscale values ​​of all pixels in the sub-region are used as the overall grayscale index of the sub-region;

[0015] According to the initial occlusion index, the grayscale uniformity index and the overall grayscale index, an external occlusion index of the sub-region is obtained; the initial occlusion index and the overall grayscale index are both positively correlated with the external occlusion index, and the grayscale uniformity index is negatively correlated with the external occlusion index;

[0016] For each X-ray image, if the external occlusion index of all sub-regions in the X-ray image is less than a preset occlusion threshold, the X-ray image is recorded as an X-ray valid image.

[0017] Furthermore, obtaining the abnormal index of each spatial area includes:

[0018] For each spatial region, the coordinate points of the spatial region are respectively mapped to all coordinate planes of the coordinate system of the three-dimensional model of the bone joint, the number of mapped points on each coordinate plane is counted, and the coordinate plane corresponding to the maximum value of the number is selected as the reference plane of the spatial region;

[0019] Obtaining the range of the distance from the coordinate point of the spatial region to the reference plane, recorded as the distance distribution difference of the spatial region; counting the total number of coordinate points in the spatial region, recorded as the quantity index of the spatial region;

[0020] According to the distance distribution difference and the quantity index, an abnormality index of a spatial area is obtained; the distance distribution difference and the abnormality index are positively correlated, and the quantity index and the abnormality index are negatively correlated.

[0021] Furthermore, obtaining the construction dislocation rate of each spatial anomaly area includes:

[0022] For each spatial anomaly region, the variance of the distance between each coordinate point of the spatial anomaly region and the coordinate points of its preset spatial neighborhood range and the reference plane is obtained as the local concave-convex index of each coordinate point of the spatial anomaly region; the cumulative sum of the local concave-convex indexes of all coordinate points of the spatial anomaly region is recorded as the overall concave-convex index of the spatial anomaly region;

[0023] The mean value of the distance from the spatial anomaly region to the reference plane is obtained, and recorded as the analysis distance of the spatial anomaly region; the mean value of the distance from the coordinate points of all adjacent regions of the spatial anomaly region to the reference plane of the spatial anomaly region is taken as the neighborhood distance of the spatial anomaly region; the absolute value of the difference between the analysis distance and the neighborhood distance is taken as the neighborhood splitting index of the spatial anomaly region;

[0024] Based on the overall concave-convex index, the anomaly index and the neighborhood splitting index, the construction dislocation rate of the spatial anomaly area is obtained.

[0025] Furthermore, the three-dimensional model of the bone joint is adjusted according to the position distribution difference between each coordinate point in the spatial abnormal area and the remaining coordinate points and the constructed dislocation rate, and the bone joint replacement surgery is assisted based on the adjusted three-dimensional model of the bone joint, including:

[0026] Selecting a construction dislocation region from the spatial anomaly region based on the construction dislocation rate;

[0027] Projecting each constructed dislocation region onto the reference plane to obtain a projection region corresponding to the constructed dislocation region; recording the coordinate points on the edge of the projection region corresponding to the coordinate points in each constructed dislocation region as projection edge points of each constructed dislocation region;

[0028] According to the difference between each coordinate point of each constructed dislocation area and the distance from the projection edge point to the reference plane, and the constructed dislocation rate, the distance from the coordinate point of each constructed dislocation area to the reference plane is adjusted to obtain a corrected distance for each coordinate point of each constructed dislocation area;

[0029] Based on the correction distance, the position of the coordinate point of each constructed dislocation area is re-determined, and the position of the coordinate point of the remaining space area of ​​the bone joint three-dimensional model except the constructed dislocation area is kept unchanged to obtain the bone joint optimization model;

[0030] Bone joint replacement surgery is assisted based on the bone joint optimization model.

[0031] Furthermore, the correction distance of each coordinate point of each constructed dislocation area is expressed by the formula:

[0032]

[0033] In the formula, The correction distance of the i-th coordinate point of the k-th constructed dislocation area of ​​the bone joint three-dimensional model; The distance from the i-th coordinate point of the k-th constructed dislocation area of ​​the bone joint three-dimensional model to the reference plane; is the construction misalignment rate of the kth construction misalignment area of ​​the bone and joint three-dimensional model; The average value of the distances from all the projected edge points of the kth constructed dislocation area of ​​the bone joint three-dimensional model to the reference plane; is the absolute value function; Norm is the normalization function.

[0034] Further, the step of selecting a construction dislocation region from the spatial abnormal region based on the construction dislocation rate includes:

[0035] For all the spatial abnormal regions of the bone and joint three-dimensional model, the spatial abnormal region corresponding to the constructed misalignment rate greater than the preset misalignment threshold is recorded as the constructed misalignment region.

[0036] Furthermore, the selecting of the spatial abnormal area based on the abnormal index includes:

[0037] The mean value of the abnormal index of all spatial regions of the three-dimensional model of the bone joint is obtained; and the spatial region where the abnormal index is greater than the mean value is regarded as the spatial abnormal region.

[0038] Furthermore, the division of the coordinate points on the three-dimensional model of the bone joint into different spatial regions includes:

[0039] The coordinate points on the three-dimensional model of the bone joint are clustered to obtain different clusters; each spatial region is formed by the coordinate points in each cluster.

[0040] Furthermore, the method for three-dimensional reconstruction of the X-ray effective image is an SFM algorithm.

[0041] The present invention has the following beneficial effects:

[0042] In the embodiment of the present invention, external metal ornaments and other bones in the body will block the image of the bone joint to be measured. The difference in the overall direction between the sub-region and the adjacent region, as well as the grayscale uniformity of the sub-region, is combined to analyze the possibility that the sub-region is caused by external metal obstruction, and the effective X-ray image is screened to eliminate the obstruction effect of external metal ornaments, thereby improving the accuracy of the construction of the three-dimensional model of the bone joint; the effective X-ray image is three-dimensionally reconstructed to obtain a three-dimensional model of the bone joint, and the position distribution difference and the number of coordinate points of the spatial region of the model can reflect the degree of abnormality of the spatial region. A comprehensive analysis is performed to obtain an abnormality index, which is used to measure the degree of abnormality of the spatial region, and then the abnormal spatial region is screened to describe the construction abnormality and the bone spurs and tumors growing on the bone surface during the three-dimensional reconstruction process. and other lesions are abnormal; the degree of separation between the spatial abnormal area and its adjacent area is analyzed according to the growth law of lesions such as bone spurs and tumors. The coordinate points with construction errors in the three-dimensional reconstruction process have obvious concave and convex characteristics, and are analyzed in combination with abnormal indicators to obtain the construction dislocation rate of the spatial abnormal area, which measures the possibility that the spatial abnormal area is caused by construction errors in the three-dimensional reconstruction process; according to the position distribution difference and construction dislocation rate of each coordinate point in the spatial abnormal area and the other coordinate points, the spatial abnormal area caused by construction errors in the three-dimensional model of the bone and joint is adjusted, and the spatial abnormal area caused by lesions such as bone spurs and tumors growing on the bone surface is retained, so that the adjusted three-dimensional model of the bone and joint is more in line with the patient's real bone and joint, thereby improving the auxiliary effect of bone and joint replacement surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A system structure diagram of a bone joint replacement surgery auxiliary system provided by one embodiment of the present invention;

[0045] Figure 2 A structural diagram of a model adjustment module provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of a computer device of a bone joint replacement surgery auxiliary device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a bone joint replacement surgery auxiliary system proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0049] The specific scheme of the bone joint replacement surgery auxiliary system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0050] Embodiment 1:

[0051] See also Figure 1 , which shows a system block diagram of a bone joint replacement surgery auxiliary system provided by an embodiment of the present invention. The system includes: a data acquisition module 110, an image screening module 120, a construction analysis module 130, and a model adjustment module 140.

[0052] The data acquisition module 110 is used to obtain at least two X-ray images of the patient's bone joints to be tested.

[0053] Joint replacement surgery is a treatment method that replaces a damaged joint with an artificial joint through surgery. It is suitable for a variety of joint diseases, including the hip, knee, shoulder, elbow, wrist and ankle.

[0054] Use an X-ray machine to take X-ray images of the patient's bone joints at different angles or positions, which are recorded as initial X-ray images; each angle or position corresponds to an initial X-ray image. Semantic segmentation is used to remove irrelevant background areas in the initial X-ray image, retain the bone tissue area, and obtain an X-ray image.

[0055] It should be noted that the X-ray image is a grayscale image; semantic segmentation is a well-known technology for those skilled in the art and will not be described in detail here.

[0056] The image screening module 120 is used to divide the X-ray image into sub-regions, and screen the effective X-ray images according to the difference in overall direction between the sub-region of each X-ray image and its adjacent region, as well as the grayscale uniformity of the sub-region.

[0057] In order to improve the accuracy of the analysis, the X-ray image is divided into different sub-regions; the sub-region division method in this embodiment is: based on the gray value of the pixel point, the K-means clustering algorithm is used to cluster the pixel points in the X-ray image to obtain multiple clusters, and the closed area formed by the pixel points in each cluster is used as the sub-region. Among them, the K-means clustering algorithm is a well-known technology for those skilled in the art and will not be described in detail here.

[0058] Other embodiments may also input the X-ray image into a neural network and divide the sub-regions through the neural network.

[0059] When performing X-ray examination on the bone joints to be tested, the bone tissue can absorb a large amount of X-rays and leave an image on the photographic film. If there are metals or other materials that absorb X-rays in the bone joints to be tested, the metal image will appear in the X-ray image, thereby blocking the bone joint tissue structure and destroying the X-ray image. The shape distribution of the blocked area and its surrounding areas will be more obviously different, affecting the accuracy of the construction of the three-dimensional model of the bone joints.

[0060] In addition to the metal ornaments worn by the patient, the other possible reasons for the obstruction of the bone joints to be measured are the unreasonable position of the patient's own movements during the X-ray examination, which causes other bone joints to obstruct the bone joints to be measured. The former has a greater impact on the construction of the three-dimensional model and needs to be discarded, while the latter can be retained. Therefore, it is necessary to filter out the effective X-ray images from the X-ray images to exclude the influence of the metal ornaments on the patient's body on the obstruction of the bone joints to be measured.

[0061] The density of metal accessories is relatively consistent, but the density of bones is not uniform. The density of objects is presented in the grayscale of X-ray images. The grayscale of the occluded area caused by metal accessories outside the body is relatively uniform, while the grayscale uniformity of the occluded area caused by other bone joints inside the body is relatively low. The overall direction of the sub-region reflects the shape distribution of the sub-region. The overall direction of the occluded area caused by metal accessories outside the body is relatively different from that of the surrounding area, while the overall direction of the occluded area caused by other bone joints inside the body is relatively different from that of the surrounding area.

[0062] Therefore, combining the difference in the overall direction between the sub-region and the adjacent region, as well as the grayscale uniformity of the sub-region, the possibility of the sub-region being caused by external metal occlusion is analyzed, and the external occlusion index is obtained to eliminate the occlusion effect of external metal accessories and improve the accuracy of the construction of the three-dimensional model of the bone joint.

[0063] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the in vitro occlusion index includes: for each sub-region of each X-ray image, the main eigenvector obtained by principal component analysis of the sub-region is recorded as the overall direction vector of the sub-region; the mean of the angle between the overall direction vectors of the sub-region and all its adjacent regions is calculated as the initial occlusion index of the sub-region; there are edge pixels on the edge of the adjacent region and are on the edge of the sub-region; the sum of the absolute values ​​of the difference between the grayscale values ​​of each pixel in the sub-region and the pixels in its preset neighborhood range is used as the grayscale difference index of each pixel in the sub-region; the sum of the grayscale difference indexes of all pixels in the sub-region is calculated to obtain the grayscale uniformity index of the sub-region; the sum of the grayscale values ​​of all pixels in the sub-region is used as the overall grayscale index of the sub-region; according to the initial occlusion index, the grayscale uniformity index and the overall grayscale index, the in vitro occlusion index of the sub-region is obtained. Among them, the principal component analysis technology is a well-known technology for those skilled in the art and will not be repeated here.

[0064] The initial occlusion index shows the difference in the overall direction between the sub-region and the surrounding area, and further reflects the difference in shape distribution between the sub-region and the surrounding area; if the initial occlusion index is larger, the difference in shape distribution between the sub-region and the surrounding area is larger, indicating that the X-ray image of the sub-region is severely damaged by the external metal jewelry, and the sub-region is more seriously affected by the external metal jewelry. The surrounding area of ​​the sub-region refers to the adjacent area of ​​the sub-region.

[0065] The grayscale uniformity index shows the uniformity of the grayscale distribution of the sub-region; the grayscale difference index represents the local grayscale uniformity of the preset neighborhood range of the pixel point. If the grayscale difference index is larger, it means that the grayscale distribution of the local range of the pixel point is more uneven; the grayscale uniformity of the local range of all pixels in the sub-region is comprehensively analyzed to obtain the overall grayscale uniformity of the sub-region. If the grayscale uniformity index is smaller, it means that the overall grayscale distribution of the sub-region is more uniform, and the sub-region is more seriously affected by the metal jewelry outside the body. It should be noted that the preset neighborhood in the embodiment of the present invention is eight neighborhoods, and the implementer can set it according to the specific situation.

[0066] It is known that the density of metal is greater than that of bone, and the denser the object, the greater the grayscale in the X-ray image. The overall grayscale index reflects the overall grayscale distribution of the sub-region; if the overall grayscale index is larger, the overall grayscale of the sub-region is larger, and the sub-region is more seriously affected by the metal jewelry outside the body.

[0067] Therefore, the initial occlusion index and the overall grayscale index are both positively correlated with the external occlusion index, and the grayscale uniformity index is negatively correlated with the external occlusion index. In an embodiment of the present invention, a negative correlation mapping is performed on the grayscale uniformity index of the sub-region, and the product of the mapping result, the initial occlusion index and the overall grayscale index is normalized to obtain the external occlusion index of the sub-region.

[0068] In a specific implementation of the embodiment of the present invention, the external shielding index of the sub-region is expressed by the formula:

[0069]

[0070] Where, Q is the external occlusion index of the sub-region; A is the total number of adjacent regions of the sub-region; is the angle between the sub-region and the overall direction vector of its ath adjacent region; is the initial occlusion index of the sub-region; B is the total number of pixels in the sub-region; is the gray value of the b-th pixel in the sub-area; is the overall grayscale index of the sub-region; is the grayscale value of the cth pixel in the preset neighborhood of the bth pixel in the sub-region; C is the total number of pixels in the preset neighborhood of each pixel in the sub-region. Since the preset neighborhood in this embodiment is eight neighborhoods, C=8. The implementer can set it according to the specific situation; is the gray uniformity index of the sub-region; is the absolute value function; exp is the exponential function with a natural constant as the base; the Sigmoid function is the normalization function.

[0071] It should be noted that in the embodiment of the present invention, an exponential function with a natural constant as the base is selected to implement negative correlation mapping, and a Sigmoid function is used for normalization processing. Other negative correlation mapping methods and normalization methods can also be selected in the embodiment of the present invention, which are not limited here.

[0072] The sub-region with a larger external occlusion index is more likely to be occluded by external metal accessories, and the sub-region with a smaller external occlusion index is more likely to be occluded by other bones and joints in the body; external metal accessories have a greater impact on the construction of the three-dimensional model and need to be discarded and do not participate in the subsequent construction of the three-dimensional bone and joint model.

[0073] For each X-ray image, if the external occlusion index of all sub-regions in the X-ray image is less than the preset occlusion threshold, the X-ray image is recorded as an X-ray valid image. The sub-regions whose external occlusion index is less than the preset occlusion threshold are occluded by other bones and joints in the body. The X-ray valid image is not affected by external metal ornaments, and has little impact on the construction of the bone and joint three-dimensional model.

[0074] It should be noted that, in this embodiment, the preset occlusion threshold is an empirical value of 0.5, and the implementer can set it according to the specific situation.

[0075] The construction analysis module 130 is used to perform three-dimensional reconstruction of the X-ray effective image to obtain a three-dimensional model of the bone joint; the coordinate points on the three-dimensional model of the bone joint are divided into different spatial regions, and the abnormal index of each spatial region is obtained according to the position distribution difference and number of the coordinate points in each spatial region, and the spatial abnormal region is selected based on the abnormal index; the construction dislocation rate of each spatial abnormal region is obtained according to the unevenness and abnormal index of each spatial abnormal region, and the degree of separation of each spatial abnormal region and its adjacent regions.

[0076] The Structure from Motion (SFM) algorithm is used to perform three-dimensional reconstruction on the effective X-ray image of the bone joint to be tested, and a three-dimensional model of the bone joint is obtained. The SFM algorithm is a well-known technology for those skilled in the art and will not be described in detail here.

[0077] In order to extract the abnormal area of ​​the bone joint three-dimensional model, it is necessary to cluster the coordinate points on the bone joint three-dimensional model based on the distance between different coordinate points on the bone joint three-dimensional model to obtain different clusters; each spatial area is composed of the coordinate points in each cluster.

[0078] It should be noted that the embodiment of the present invention selects the DBSCAN algorithm to cluster the coordinate points on the three-dimensional model of bone joints. The neighborhood radius Eps is 20, and the minimum number of points MinPts is 20. The implementer can set them according to the specific situation.

[0079] The method for selecting the spatial anomaly area is as follows:

[0080] The reasons for abnormalities in the three-dimensional model of bone and joints are: abnormal construction during the three-dimensional reconstruction process, abnormal lesions such as bone spurs and tumors growing on the bone surface. Abnormal areas in the three-dimensional model of bone and joints usually cause irregular protrusions and depressions on the surface, resulting in a large distribution gap in the abnormal area. The normal surface is usually smoother, that is, the distribution gap is smaller; the coordinate points in the smooth area of ​​the model are relatively evenly distributed in space and tend to gather into larger clusters, that is, the number of coordinate points in the cluster is large; on the contrary, the distribution of coordinate points in the abnormal area of ​​the model has a large local variation, which is easier to be divided into different clusters, and the number of coordinate points in the cluster is small.

[0081] The position distribution difference of the coordinate points in the spatial area reflects the distribution difference of the spatial area. The position distribution difference of the coordinate points in the spatial area and the number of coordinate points in the spatial area can both reflect the degree of abnormality of the spatial area. The abnormality index is obtained by combining the two factors to measure the degree of abnormality of the spatial area.

[0082] Preferably, in some possible implementation modes of the embodiments of the present invention, the method for obtaining the abnormal index of the spatial region includes: for each spatial region, mapping the coordinate points of the spatial region to all coordinate planes of the coordinate system in which the three-dimensional model of the bone joint is located, counting the number of mapped points on each coordinate plane, and selecting the coordinate plane corresponding to the maximum value in the number as the reference plane of the spatial region; obtaining the extreme difference of the distance from the coordinate point of the spatial region to the reference plane, which is recorded as the distance distribution difference of the spatial region; counting the total number of coordinate points in the spatial region, which is recorded as the quantity index of the spatial region; and obtaining the abnormal index of the spatial region based on the distance distribution difference and the quantity index.

[0083] It should be noted that when using the SFM algorithm to reconstruct bones and joints in three dimensions, the three-dimensional coordinate system is usually based on anatomical reference points (such as joint centers or specific parts of bones), combined with the camera coordinate system and projection matrix, and obtained through multi-view image reconstruction; the three-dimensional model obtained by three-dimensional reconstruction using the SFM algorithm contains a three-dimensional coordinate system. The points where the coordinate points on the three-dimensional model of bones and joints are mapped on the coordinate plane are called mapping points.

[0084] The coordinate system of the three-dimensional model of bone joints has three coordinate planes. The number of points mapped on each coordinate plane of the coordinate points in the analysis space area is analyzed. The larger the number of coordinate planes, the better they can show the distribution difference of the coordinate points in the space area, and the higher the accuracy of the abnormal analysis of the space area. Therefore, the coordinate plane corresponding to the maximum number of points mapped on all coordinate planes of the coordinate points in the space area is used as the reference plane of the space area.

[0085] The distance distribution difference reflects the distribution difference of the spatial area through the distance from the coordinate point of the spatial area to its reference plane. If the distance distribution difference index is larger, it means that the distribution difference of the spatial area is larger, and the possibility that the spatial area is an abnormal area of ​​the bone and joint three-dimensional model is greater; conversely, the possibility that the spatial area is a normal area of ​​the bone and joint three-dimensional model is greater.

[0086] The quantitative index directly presents the degree of abnormality of the spatial area. If the quantitative index is larger, it means that the spatial area is smoother and the possibility that the spatial area is a normal area of ​​the bone and joint three-dimensional model is greater; conversely, the possibility that the spatial area is an abnormal area of ​​the bone and joint three-dimensional model is greater.

[0087] Therefore, the distance distribution difference is positively correlated with the abnormal index, and the quantity index is negatively correlated with the abnormal index. In the embodiment of the present invention, the quantity index is negatively correlated, and the product of the mapping result and the distance distribution difference is normalized to obtain the abnormal index of the spatial region.

[0088] In the embodiment of the present invention, the correlation between the distance distribution gap and the quantity index and the abnormal index can also be constructed through other basic mathematical operations, which will not be limited or elaborated here.

[0089] It should be noted that, since the quantity index must be a positive integer, the inverse of the quantity index is taken in the embodiment of the present invention to achieve negative correlation mapping; the Sigmoid function is used for normalization processing. Other normalization methods can also be selected in the embodiment of the present invention, such as function transformation, maximum and minimum normalization, etc., which are not limited here.

[0090] Since the spatial region with a larger abnormal index is more likely to be an abnormal region of the bone and joint three-dimensional model, the average of the abnormal indexes of all spatial regions of the bone and joint three-dimensional model is obtained; the spatial region with an abnormal index greater than the average is regarded as a spatial abnormal region. The spatial abnormal region is an abnormal region caused by construction errors during the three-dimensional reconstruction process and lesions such as bone spurs and tumors growing on the bone surface.

[0091] The method to obtain the build misalignment rate is as follows:

[0092] The manifestations of abnormal spatial areas caused by lesions and construction errors are different, specifically:

[0093] Since the current 3D reconstruction algorithm is relatively mature, the spatial range of the spatial abnormal area caused by construction errors is relatively small; construction errors are mainly caused by the SFM algorithm's incorrect matching of feature points in different X-ray effective images, which makes the position difference between the incorrectly matched coordinate points and the surrounding coordinate points more obvious. Therefore, compared with the spatial abnormal area caused by lesions such as bone spurs and tumors growing on the bone surface, the abnormal index of the spatial abnormal area caused by construction errors in the 3D reconstruction process is larger.

[0094] Lesions such as bone spurs and tumors grow irregularly, and their surfaces are usually uneven. Compared with the unevenness of the lesion surface, the position difference between the coordinate points with construction errors and the surrounding coordinate points during the 3D reconstruction process is more obvious, that is, the unevenness is more obvious. Therefore, compared with the spatial abnormal areas caused by lesions such as bone spurs and tumors growing on the bone surface, the spatial abnormal areas caused by construction errors during the 3D reconstruction process have more obvious unevenness.

[0095] Since the lesion area grows out of the bone, the connection between the lesion area and the bone area is relatively smooth, so the degree of separation between the lesion area and its adjacent areas is small; while the construction error area is due to the incorrect determination of the position of some pixel points, which does not conform to the growth law of the bone, resulting in a large degree of separation between the construction error area and its adjacent areas.

[0096] The construction misalignment rate is obtained by combining the unevenness and abnormal indicators of the spatial abnormal area, as well as the degree of separation between the spatial abnormal area and its adjacent areas, which measures the possibility that the spatial abnormal area is caused by construction errors in the three-dimensional reconstruction process.

[0097] Preferably, in some possible implementation modes of the embodiments of the present invention, the method for obtaining the construction misalignment rate includes: for each spatial anomaly area, obtaining the variance of the distance from each coordinate point of the spatial anomaly area and the coordinate points of its preset spatial neighborhood range to the reference plane as the local concave-convex index of each coordinate point of the spatial anomaly area; recording the cumulative sum of the local concave-convex indexes of all coordinate points of the spatial anomaly area as the overall concave-convex index of the spatial anomaly area; obtaining the mean of the distance from the spatial anomaly area to its reference plane, recorded as the analysis distance of the spatial anomaly area; taking the mean of the distance from the coordinate points of all adjacent areas of the spatial anomaly area to the reference plane of the spatial anomaly area as the neighborhood distance of the spatial anomaly area; taking the absolute value of the difference between the analysis distance and the neighborhood distance as the neighborhood splitting index of the spatial anomaly area; and obtaining the construction misalignment rate of the spatial anomaly area based on the overall concave-convex index, the anomaly index and the neighborhood splitting index.

[0098] The local concavity index reflects the degree of concavity in the local range of the coordinate point. The overall concavity index is obtained by summing up the local concavity index of all coordinate points in the spatial abnormal area and considering the overall concavity of the spatial abnormal area. The larger the overall concavity index, the greater the possibility that the spatial abnormal area is caused by construction errors during the 3D reconstruction process.

[0099] The analysis distance reflects the overall level of the distance from the spatial anomaly area to its reference plane, and the neighborhood distance reflects the overall level of the distance from the adjacent area of ​​the spatial anomaly area to the reference plane of the spatial anomaly area. If the difference between the analysis distance and the neighborhood distance, that is, the neighborhood splitting index, is larger, it means that the degree of splitting between the spatial anomaly area and its adjacent area is greater, and the spatial anomaly area is more likely to be caused by construction errors in the three-dimensional reconstruction process.

[0100] Therefore, the overall concavity index, anomaly index and neighborhood splitting index are all positively correlated with the construction dislocation rate.

[0101] It should be noted that in the embodiment of the present invention, a spherical area with a radius of 3 and each coordinate point as the center is used as the preset spatial neighborhood of each coordinate point, and the implementer can set it according to the specific situation.

[0102] In a specific implementation of the embodiment of the present invention, the construction misalignment rate is expressed by the formula:

[0103]

[0104] Where R is the construction dislocation rate of each spatial abnormal area; W is the abnormal index of each spatial abnormal area; W_max is the maximum value of the abnormal indexes of all spatial abnormal areas of the bone and joint three-dimensional model; E is the total number of coordinate points of each spatial abnormal area; is the local concave-convex index of the e-th coordinate point in each spatial abnormal region; is the overall concavity and convexity index of each spatial abnormal area; is the distance from the e-th coordinate point of each spatial anomaly area to its reference plane; is the analysis distance of each spatial anomaly area; F is the total number of adjacent areas of each spatial anomaly area; is the total number of coordinate points of the fth adjacent area of ​​each spatial abnormal area; is the distance from the g-th coordinate point of the f-th adjacent area of ​​each spatial anomaly area to the reference plane of the spatial anomaly area; is the neighborhood distance of each spatial anomaly area; Neighborhood splitting indicator for each spatial abnormal area; is the absolute value function; Norm is the normalization function.

[0105] It should be noted that, if the construction misalignment rate R is larger, the possibility that each spatial abnormal area is caused by a construction error in the three-dimensional reconstruction process is greater; if the construction misalignment rate R is smaller, the possibility that each spatial abnormal area is caused by lesions such as bone spurs and tumors growing on the bone surface is greater.

[0106] The model adjustment module 140 is used to adjust the three-dimensional model of the bone joint according to the position distribution difference between each coordinate point in the spatial abnormal area and the remaining coordinate points and the constructed dislocation rate, and assist the bone joint replacement surgery based on the adjusted three-dimensional model of the bone joint.

[0107] In order to ensure the smooth progress of the patient's bone and joint replacement surgery, it is necessary to adjust the spatial abnormal areas caused by construction errors in the three-dimensional model of the bone and joint to obtain an optimized model of the bone and joint.

[0108] See also Figure 2 , which shows a structural diagram of a model adjustment module provided by an embodiment of the present invention, the model adjustment module includes: constructing a dislocation area screening unit 141, a projection edge point screening unit 142, a correction distance analysis unit 143, and a model correction unit 144.

[0109] The construction misalignment region screening unit 141 selects a construction misalignment region from the spatial abnormal region based on the construction misalignment rate.

[0110] For all the spatial abnormal regions of the bone and joint three-dimensional model, the spatial abnormal regions corresponding to the construction misalignment rate greater than the preset misalignment threshold are recorded as construction misalignment regions. The construction misalignment region is the spatial region caused by construction errors during the three-dimensional reconstruction process.

[0111] It should be noted that, in this embodiment, the preset misalignment threshold is an empirical value of 0.5, and the implementer can set it according to the specific situation.

[0112] The projection edge point screening unit 142 projects each constructed dislocation region onto a reference plane to obtain a projection region corresponding to the constructed dislocation region; and records the coordinate points on the edge of the projection region corresponding to the coordinate points in each constructed dislocation region as projection edge points of each constructed dislocation region.

[0113] The constructed dislocation area appears as an obvious bulge or depression, and the projection edge points of the constructed dislocation area are usually more smoothly connected with its adjacent areas. By adjusting the projection edge points, the adjusted constructed dislocation area can be made to conform to the growth law of bones, thereby making the adjusted bone and joint three-dimensional model more consistent with the patient's actual bone and joint.

[0114] Corrected distance analysis unit 143: According to the difference between the distance between each coordinate point of each constructed misaligned area and the projection edge point to the reference plane, and the constructed misalignment rate, the distance between the coordinate point of each constructed misaligned area and the reference plane is adjusted to obtain the corrected distance of each coordinate point of each constructed misaligned area.

[0115] If the construction misalignment rate is larger, it means that the error degree of the construction misalignment area in the 3D construction process is greater, and the coordinate point of the construction misalignment area should be adjusted more. If the difference between the distances of the coordinate point of the construction misalignment area and the projection edge point to the reference plane is larger, in order to make the distances of the coordinate point and the projection edge point to the reference plane closer, the coordinate point needs to be adjusted more; otherwise, the coordinate point needs to be adjusted less.

[0116] The correction distance of each coordinate point in each constructed misaligned area is expressed by the formula:

[0117]

[0118] In the formula, The correction distance of the i-th coordinate point of the k-th constructed dislocation area of ​​the bone joint three-dimensional model; The distance from the i-th coordinate point of the k-th constructed dislocation area of ​​the bone joint three-dimensional model to the reference plane; is the construction misalignment rate of the kth construction misalignment area of ​​the bone and joint three-dimensional model; The average value of the distances from all the projected edge points of the kth constructed dislocation area of ​​the bone joint three-dimensional model to the reference plane; is the absolute value function; Norm is the normalization function.

[0119] It should be noted that when When the value is larger, it means that the i-th coordinate point of the k-th constructed dislocation area of ​​the three-dimensional model of the bone joint is in a more obvious concave position. In order to smooth the i-th coordinate point, the distance from the i-th coordinate point to the reference plane needs to be increased. The smaller it is, the more obvious that the i-th coordinate point is in a convex position. In order to smooth the i-th coordinate point, it is necessary to reduce the distance from the i-th coordinate point to the reference plane.

[0120] Model correction unit 144: re-determines the position of the coordinate point of each constructed dislocation area based on the correction distance, keeps the position of the coordinate point of the remaining spatial area of ​​the bone joint three-dimensional model except the constructed dislocation area unchanged, and obtains the bone joint optimization model.

[0121] It should be noted that the distance between the coordinate point corresponding to the coordinate point of the constructed dislocation area on the bone joint optimization model and the reference plane of the constructed dislocation area is equal to the correction distance.

[0122] The bone and joint optimization model is closer to the actual three-dimensional model of the patient's bone and joint to be tested. The Artificial Intelligence Knee (AIKNEE) system can accurately simulate the patient's bone morphology and joint structure through the bone and joint optimization model, determine the planning parameters to design the 3D osteotomy guide, select the appropriate artificial joint model based on the osteotomy guide, and reasonably cut the selected artificial joint based on the osteotomy guide. The cut artificial joint will replace the bone and joint to be tested in the subsequent bone and joint replacement surgery.

[0123] So far, the present invention is completed.

[0124] Embodiment 2:

[0125] Figure 3 A computer device schematic diagram of a bone joint replacement surgery auxiliary device provided by an embodiment of the present invention. Figure 3 As shown, the computer device includes: a memory 201, a processor 202, and a computer program 203 stored in the memory 201 and running on the processor 202, wherein when the processor 202 executes the computer program 203, the computer device can execute any one of the bone replacement surgery auxiliary systems introduced above.

[0126] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute a bone joint replacement surgery auxiliary system provided by an embodiment of the present application.

[0127] In this embodiment, the functional modules of the device can be divided according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0128] It should be understood that the device provided in this embodiment is used to implement the above-mentioned bone joint replacement surgery auxiliary system, and thus can achieve the same effect as the above-mentioned implementation method.

[0129] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the actions of the device. The storage module may be used to support the device to execute mutual program codes, etc.

[0130] The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits disclosed in the present application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0131] Embodiment 3:

[0132] This embodiment also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes the above-mentioned related method steps to implement a bone joint replacement surgery auxiliary system provided by the above embodiment.

[0133] Among them, the device and computer-readable storage medium provided in this embodiment are used to execute the corresponding system provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding system provided above, and will not be repeated here.

[0134] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0135] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0136] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A bone joint replacement surgery auxiliary system, characterized in that: The system includes: A data acquisition module, used for acquiring at least two X-ray images of the patient's bone joints to be tested; An image screening module is used to divide the X-ray image into sub-regions, and screen the effective X-ray image according to the difference in the overall direction between the sub-region of each X-ray image and its adjacent region, as well as the grayscale uniformity of the sub-region; An analysis module is constructed to perform three-dimensional reconstruction of the effective X-ray image to obtain a three-dimensional model of the bone joint; the coordinate points on the three-dimensional model of the bone joint are divided into different spatial regions, and the abnormal index of each spatial region is obtained according to the position distribution difference and the number of the coordinate points in each spatial region, and the spatial abnormal region is selected based on the abnormal index; the construction dislocation rate of each spatial abnormal region is obtained according to the unevenness of each spatial abnormal region and the abnormal index, as well as the degree of separation of each spatial abnormal region and its adjacent region; A model adjustment module, used to adjust the bone joint three-dimensional model according to the position distribution difference between each coordinate point in the spatial abnormal area and the remaining coordinate points and the constructed dislocation rate, and assist bone joint replacement surgery based on the adjusted bone joint three-dimensional model; The obtaining of abnormal indicators of each spatial area includes: For each spatial region, the coordinate points of the spatial region are respectively mapped to all coordinate planes of the coordinate system of the three-dimensional model of the bone joint, the number of mapped points on each coordinate plane is counted, and the coordinate plane corresponding to the maximum value of the number is selected as the reference plane of the spatial region; Obtaining the range of the distance from the coordinate point of the spatial region to the reference plane, recorded as the distance distribution difference of the spatial region; counting the total number of coordinate points in the spatial region, recorded as the quantity index of the spatial region; According to the distance distribution difference and the quantity index, an abnormality index of the spatial area is obtained; the distance distribution difference and the abnormality index are positively correlated, and the quantity index and the abnormality index are negatively correlated; The selecting of a spatial abnormal area based on the abnormal indicator includes: Obtaining the mean value of the abnormal index in all spatial regions of the three-dimensional model of the bone joint; taking the spatial region where the abnormal index is greater than the mean value as the spatial abnormal region; The method for three-dimensional reconstruction of the X-ray effective image is the SFM algorithm; The step of obtaining the construction misalignment rate of each spatial anomaly region includes: For each spatial anomaly region, the variance of the distance between each coordinate point of the spatial anomaly region and the coordinate points of its preset spatial neighborhood range and the reference plane is obtained as the local concave-convex index of each coordinate point of the spatial anomaly region; the cumulative sum of the local concave-convex indexes of all coordinate points of the spatial anomaly region is recorded as the overall concave-convex index of the spatial anomaly region; The mean value of the distance from the spatial anomaly region to the reference plane is obtained, and recorded as the analysis distance of the spatial anomaly region; the mean value of the distance from the coordinate points of all adjacent regions of the spatial anomaly region to the reference plane of the spatial anomaly region is taken as the neighborhood distance of the spatial anomaly region; the absolute value of the difference between the analysis distance and the neighborhood distance is taken as the neighborhood splitting index of the spatial anomaly region; Based on the overall concave-convex index, the anomaly index and the neighborhood splitting index, the construction dislocation rate of the spatial anomaly area is obtained.

2. The bone joint replacement surgery auxiliary system according to claim 1, characterized in that: The screening X-ray effective images include: For each sub-region of each X-ray image, the principal eigenvector obtained by principal component analysis of the sub-region is recorded as the overall direction vector of the sub-region; the mean of the angles between the sub-region and the overall direction vectors of all its adjacent regions is calculated as the initial occlusion index of the sub-region; the edge pixel points on the edge of the adjacent region are located on the edge of the sub-region; The accumulated sum of the absolute values ​​of the grayscale values ​​of each pixel in the sub-region and the pixels in its preset neighborhood range is used as the grayscale difference index of each pixel in the sub-region; the accumulated sum of the grayscale difference indexes of all pixels in the sub-region is calculated to obtain the grayscale uniformity index of the sub-region; The accumulated grayscale values ​​of all pixels in the sub-region are used as the overall grayscale index of the sub-region; According to the initial occlusion index, the grayscale uniformity index and the overall grayscale index, an external occlusion index of the sub-region is obtained; the initial occlusion index and the overall grayscale index are both positively correlated with the external occlusion index, and the grayscale uniformity index is negatively correlated with the external occlusion index; For each X-ray image, if the external occlusion index of all sub-regions in the X-ray image is less than a preset occlusion threshold, the X-ray image is recorded as an X-ray valid image.

3. The bone joint replacement surgery auxiliary system according to claim 1, characterized in that: The method of adjusting the three-dimensional bone joint model according to the position distribution difference between each coordinate point in the abnormal spatial area and the remaining coordinate points and the constructed dislocation rate, and assisting bone joint replacement surgery based on the adjusted three-dimensional bone joint model, comprises: Selecting a construction dislocation region from the spatial anomaly region based on the construction dislocation rate; Projecting each constructed dislocation region onto the reference plane to obtain a projection region corresponding to the constructed dislocation region; recording the coordinate points on the edge of the projection region corresponding to the coordinate points in each constructed dislocation region as projection edge points of each constructed dislocation region; According to the difference between each coordinate point of each constructed dislocation area and the distance from the projection edge point to the reference plane, and the constructed dislocation rate, the distance from the coordinate point of each constructed dislocation area to the reference plane is adjusted to obtain a corrected distance for each coordinate point of each constructed dislocation area; Based on the correction distance, the position of the coordinate point of each constructed dislocation area is re-determined, and the position of the coordinate point of the remaining space area of ​​the bone joint three-dimensional model except the constructed dislocation area is kept unchanged to obtain the bone joint optimization model; Bone joint replacement surgery is assisted based on the bone joint optimization model.

4. The bone joint replacement surgery auxiliary system according to claim 3, characterized in that: The correction distance of each coordinate point of each constructed dislocation area is expressed by the formula: In the formula, The correction distance of the i-th coordinate point of the k-th constructed dislocation area of ​​the bone joint three-dimensional model; The distance from the i-th coordinate point of the k-th constructed dislocation area of ​​the bone joint three-dimensional model to the reference plane; is the construction misalignment rate of the kth construction misalignment area of ​​the bone and joint three-dimensional model; The average value of the distances from all the projected edge points of the kth constructed dislocation area of ​​the bone joint three-dimensional model to the reference plane; is the absolute value function; Norm is the normalization function.

5. The bone joint replacement surgery auxiliary system according to claim 3, characterized in that: The selecting a construction dislocation region from the spatial abnormal region based on the construction dislocation rate comprises: For all the spatial abnormal regions of the bone and joint three-dimensional model, the spatial abnormal region corresponding to the constructed misalignment rate greater than the preset misalignment threshold is recorded as the constructed misalignment region.

6. The bone joint replacement surgery auxiliary system according to claim 1, characterized in that: The coordinate points on the three-dimensional model of the bone joint are divided into different spatial regions, including: The coordinate points on the three-dimensional model of the bone joint are clustered to obtain different clusters; each spatial region is formed by the coordinate points in each cluster.

Images