Iconography auxiliary evaluation system used after spine surgery in orthopedics department

Abstract

The invention relates to the technical field of medical care information, in particular to an orthopaedic spine post-operation imaging auxiliary evaluation system, which comprises a region acquisition module, a region correction module, a callus growth coefficient acquisition module and a network training module, the method comprises the following steps: acquiring an initial vertebra fusion region in a spinal image of a historical patient after a spinal fusion operation, and correcting the initial vertebra fusion region to obtain a real vertebra fusion region; determining a plurality of first intervertebral gradient regions and second intervertebral gradient regions in the real vertebra fusion region, and determining a callus growth coefficient of the real vertebra fusion region according to position distribution and pixel gray levels of the first intervertebral gradient regions and the second intervertebral gradient regions; and forming a training set by utilizing the callus growth coefficients of all historical patients and the postoperative risk grades calibrated in the initial vertebra fusion region, and training the postoperative risk classification network. According to the invention, the classification accuracy of the postoperative risk classification network is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of healthcare information technology, and particularly to an imaging-assisted evaluation system for orthopedic spinal surgery. Background Art

[0002] When a patient's spinal lesion causes pain or dysfunction, spinal fusion surgery is usually performed to fuse two or more spinal bones together, so as to limit their movement, relieve pain, restore spinal stability or improve function. A series of complications may occur after orthopedic spinal surgery, such as loosening or displacement of the metal device in the spine, spinal fracture, bone infection, etc. In order to ensure good surgical results, postoperative evaluation is usually required.

[0003] In the prior art, in order to evaluate the spinal fusion condition of a patient after spinal surgery, machine vision-assisted evaluation is usually performed by collecting partial images of the patient's spine after surgery, that is, the vertebral fusion area image in the partial image of the patient's spine after surgery is input into a classification model, and the classification model performs classification to obtain an assisted evaluation result. However, in the process of obtaining the classification model through training, since a fusion device is usually implanted in the patient during the surgery, the continuous change of bone mass generated during spinal fusion will be confused with the fusion gray difference generated by the fusion device in the intervertebral space, and the imaging performances of different fusion devices between vertebrae are also different, resulting in the classification model being difficult to accurately extract the characteristics of the intervertebral fusion area of the patient's spine, and thus the classification accuracy of the classification model is poor. Summary of the Invention

[0004] In order to solve the above technical problem that in the process of machine vision-assisted evaluation of partial images of a patient's spine after surgery, the classification accuracy of the classification model is poor because the classification model is difficult to accurately extract the characteristics of the intervertebral fusion area of the patient's spine, the purpose of the present invention is to provide an imaging-assisted evaluation system for orthopedic spinal surgery, and the specific technical solution adopted is as follows: In a first aspect, the present invention provides an imaging-assisted evaluation system for orthopedic spinal surgery, and the system includes: A region acquisition module, configured to acquire an initial vertebral fusion region in a spinal image of a historical patient after spinal fusion surgery; A region correction module, configured to correct the initial vertebral fusion region according to the gradient and position distribution of pixel points in the low-change region of the initial vertebral fusion region to obtain a true vertebral fusion region; A callus growth coefficient acquisition module, configured to determine a plurality of first intervertebral gradient regions and second intervertebral gradient regions in the true vertebral fusion region, and determine the callus growth coefficient of the true vertebral fusion region according to the positional distribution and grayscale values of the pixels in the first intervertebral gradient regions and the second intervertebral gradient regions, wherein the gradient of the pixels in the first intervertebral gradient regions is higher than the gradient of the pixels in the second intervertebral gradient regions; A network training module, configured to use the callus growth coefficients of all historical patients and the postoperative risk grades calibrated by the initial vertebral fusion regions to form a training set, and use the training set to train a postoperative risk classification network to obtain a trained postoperative risk classification network.

[0005] Combined with the first aspect above, in some possible implementation manners, the region correction module includes: An intervertebral space coefficient acquisition unit, configured to determine the intervertebral space coefficient of the low-change region according to the gradient and positional distribution of the pixels in the low-change region; A region correction unit, configured to screen out the intervertebral space regions in all the low-change regions according to the intervertebral space coefficient, and remove the intervertebral space regions from the initial vertebral fusion region to obtain a true vertebral fusion region, wherein the intervertebral space coefficient of the intervertebral space region is greater than a set coefficient threshold.

[0006] Combined with the first aspect above, in some possible implementation manners, the intervertebral space coefficient acquisition unit includes: A pixel clustering determination unit, configured to cluster the pixels in the low-change region according to the gradient of the pixels in the low-change region, and divide the low-change region into a plurality of pixel clusters; A sequence distance determination unit, configured to determine each pixel sequence of the pixel clusters in the vertebral arrangement direction, and a first distance in the direction perpendicular to the vertebral arrangement of the pixel clusters to the initial vertebral fusion region; An intervertebral space coefficient determination unit, configured to determine the intervertebral space coefficient of the low-change region according to the number of pixels in the pixel sequence in the pixel clusters, the first distance corresponding to the pixel clusters, and the gradient of the pixels in the pixel clusters.

[0007] Combined with the first aspect above, in some possible implementation manners, the intervertebral space coefficient determination unit includes: A first arrangement determination unit, configured to arrange all the pixel clusters in the low-change region in ascending order of the first distance to obtain a first arrangement; A second arrangement determination unit, configured to determine an average gradient of pixel clusters in the pixel point clusters, and arrange all the pixel point clusters in the low-change region in ascending order of the average gradient to obtain a second arrangement; A first arrangement difference determination unit, configured to determine a total number of pixel point clusters with arrangement differences in the first arrangement and the second arrangement; A third arrangement determination unit, configured to arrange the number of pixels in the pixel point sequence in the pixel point clusters in ascending order to obtain a third arrangement; A second arrangement difference determination unit, configured to determine a total number of pixel point sequences with arrangement differences between the arrangement order of the pixel point sequence corresponding to the third arrangement and the original arrangement of the pixel point sequence in the pixel point clusters; An intervertebral space coefficient calculation unit, configured to determine a pixel number increase slope according to the third arrangement, and determine an intervertebral space coefficient of the low-change region according to the total number of pixel point clusters with arrangement differences corresponding to the low-change region, the pixel number increase slope, and the total number of pixel point sequences.

[0008] Combined with the first aspect above, in some possible implementation manners, the intervertebral space coefficient calculation unit includes: A first calculation unit, configured to determine an average value of a first product of the pixel number increase slope of all the pixel point clusters and the total number of pixel point sequences with arrangement differences, to obtain a pixel point cluster change index; A second calculation unit: determines a second product of the total number of pixel point clusters with arrangement differences corresponding to the low-change region and the pixel point cluster change index, and performs normalization processing on the second product to obtain the intervertebral space coefficient of the low-change region.

[0009] Combined with the first aspect above, in some possible implementation manners, the region correction module further includes: A low-change region acquisition unit, configured to perform Otsu threshold segmentation on the gray values of pixel points in the initial vertebral fusion region to obtain a plurality of first gray value regions and second gray value regions, and use the second gray value region as the low-change region of the initial vertebral fusion region, where the gray value of the first gray value region is higher than the gray value of the second gray value region.

[0010] Combined with the first aspect above, in some possible implementation manners, the callus growth coefficient acquisition module includes: An information value acquisition unit, configured to determine a second distance in a direction perpendicular to the vertebral arrangement direction from the first intervertebral gradient region to the real vertebral fusion region, and determine a distance distribution difference information value according to the distribution differences of all the second distances; A grayscale difference value acquisition unit is configured to determine an adjacent second intervertebral gradient region of the first intervertebral gradient region in a target direction, and determine a grayscale difference value between the first intervertebral gradient region and its adjacent second intervertebral gradient region, where the target direction refers to the direction in which the average grayscale value of each pixel point sequence in the vertebral arrangement direction of the first intervertebral gradient region gradually decreases in the direction perpendicular to the vertebral arrangement direction; An adjacent region quantity acquisition unit is configured to determine the quantity of adjacent first intervertebral gradient regions of the adjacent second intervertebral gradient region of the first intervertebral gradient region; A callus growth coefficient determination unit is configured to determine a callus growth coefficient of the true vertebral fusion region according to the distance distribution difference information value, and the grayscale difference values and the quantity of adjacent first intervertebral gradient regions corresponding to all the first intervertebral gradient regions.

[0011] Combined with the above first aspect, in some possible implementation manners, the callus growth coefficient determination unit includes: A third calculation unit is configured to determine an average value of third products of the grayscale difference values and the quantities of adjacent first intervertebral gradient regions corresponding to all the first intervertebral gradient regions to obtain a callus growth index; A fourth calculation unit is configured to determine a fourth product of the distance distribution difference information value and the callus growth index, and perform normalization processing on the fourth product to obtain the callus growth coefficient of the true vertebral fusion region.

[0012] Combined with the above first aspect, in some possible implementation manners, the information value acquisition unit includes: An information value calculation unit is configured to determine the information entropy of the second distances of all the first intervertebral gradient regions, and use the information entropy as the distance distribution difference information value.

[0013] Combined with the above first aspect, in some possible implementation manners, the callus growth coefficient acquisition module includes: A threshold segmentation region acquisition unit is configured to perform Otsu threshold segmentation on the gradients of pixel points in the true vertebral fusion region to obtain a plurality of first intervertebral gradient regions and second intervertebral gradient regions, where the gradients of the first intervertebral gradient regions are higher than the gradients of the second intervertebral gradient regions.

[0014] In a second aspect, the present invention further provides an imaging-assisted evaluation method for orthopedic spine surgery after operation, and the method includes the following steps: Obtain an initial vertebral fusion region in a spine image of a historical patient after spine fusion surgery; Modify the initial vertebral fusion region according to the gradient and position distribution of pixel points in the low-variation region of the initial vertebral fusion region to obtain the true vertebral fusion region; Determine a number of first intervertebral gradient regions and second intervertebral gradient regions in the true vertebral fusion region, and determine the callus growth coefficient of the true vertebral fusion region according to the position distribution and gray values of pixel points in the first intervertebral gradient region and the second intervertebral gradient region, where the gradient of pixel points in the first intervertebral gradient region is higher than the gradient of pixel points in the second intervertebral gradient region; Use the callus growth coefficients of all historical patients and the postoperative risk grades calibrated by the initial vertebral fusion region to form a training set, and use the training set to train the postoperative risk classification network to obtain a trained postoperative risk classification network.

[0015] In a third aspect, the present invention also provides an imaging-assisted evaluation device for orthopedic spine surgery, including a memory, a processor, and computer program code stored in the memory. The processor is configured to call and run the computer program code from the memory, so that the device executes the method in the first aspect or any possible implementation manner of the first aspect.

[0016] In a fourth aspect, the present invention also provides a computer program product, which includes: computer program code. When the computer program code runs on a computer, the computer is caused to execute the method in the first aspect or any possible implementation manner of the first aspect.

[0017] In a fifth aspect, the present invention also provides a computer-readable storage medium storing computer program code. When the computer program code runs on a computer, the computer is caused to execute the method in the first aspect or any possible implementation manner of the first aspect.

[0018] The present invention has the following beneficial effects: By obtaining the initial vertebral fusion region in the spinal image of historical patients after spinal fusion surgery and correcting the initial vertebral fusion region, the true vertebral fusion region is obtained to avoid the influence of the vertebral space region on the evaluation of the fusion condition of the vertebral fusion region. According to the determination of several first intervertebral gradient regions and second intervertebral gradient regions in the true vertebral fusion region, that is, the high-gradient region and the low-gradient region, based on the position distribution and the gray value of the pixel points of the high-gradient region and the low-gradient region, the bone callus growth condition of the true vertebral fusion region is accurately evaluated to determine the bone callus growth coefficient; finally, the bone callus growth coefficients of all historical patients are used to train the postoperative risk classification network. By extracting the bone callus growth coefficients related to the postoperative risk in the vertebral fusion region to train the postoperative risk classification network, the postoperative risk classification network can more accurately learn the relationship between the bone callus growth condition and the postoperative risk, thereby effectively improving the classification accuracy of the postoperative risk classification network. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is a schematic structural diagram of an imaging-assisted evaluation system for orthopedic spinal surgery after the operation of an embodiment of the present invention; Figure 2 It is a flowchart of the steps of an imaging-assisted evaluation method for orthopedic spinal surgery after the operation of an embodiment of the present invention; Figure 3 It is a schematic diagram of the non-fusion region in the initial vertebral fusion region of an embodiment of the present invention; Figure 4 It is a schematic structural diagram of an imaging-assisted evaluation device for orthopedic spinal surgery after the operation of an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] In order to clearly illustrate the technical features of the present solution, the present invention will be described in detail below through specific embodiments in combination with the drawings.

[0022] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0023] It should be understood that the various steps recited in the method embodiments of the present invention can be executed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0024] The term "comprising" and its variations used herein are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0025] It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0026] In the embodiments of the present invention, although the operations or steps are described in a specific order in the drawings, it should not be understood that these operations or steps are required to be performed in the specific order shown or in a serial order, or that all the operations or steps shown are required to be performed to obtain the desired result. In the embodiments of the present invention, these operations or steps can be performed serially; they can also be performed in parallel; or a part of these operations or steps can be performed.

[0027] At the same time, it can be understood that the data involved in the technical solution of the present invention (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of the corresponding laws, regulations and related provisions. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs, and all parameters or indicators in the formulas involved in the present invention are numerical values after normalization that eliminate the influence of dimensions.

[0028] In order to solve the above technical problem that in the process of machine vision-assisted evaluation of the postoperative spinal part images of patients, it is difficult for the classification model to accurately extract the features of the spinal intervertebral fusion area of patients, resulting in poor classification accuracy of the classification model, the embodiment of the present invention provides an imaging-assisted evaluation system for orthopedic spine surgery after operation. This system is essentially a software system, which is composed of various modules that implement corresponding functions, and its corresponding structural schematic diagram is as Figure 1 shown. The core of this system is to implement an imaging-assisted evaluation method for orthopedic spine surgery after operation. Each module in this system corresponds to each step in the method, and the flowchart corresponding to this method is as Figure 2 shown. The following will introduce each module of this system in detail in combination with the specific steps in this method.

[0029] The region acquisition module 101 is used to acquire the initial vertebral fusion region in the spinal image of historical patients after spinal fusion surgery.

[0030] Specifically, in order to perform imaging-assisted evaluation for orthopedic spine surgery after operation, for several historical patients who have undergone orthopedic spine fusion surgery, X-ray images of the postoperative spinal part of them are taken by using an X-ray device. When taking the X-ray image, the patient needs to adjust the body position according to the shooting requirements. When the patient stands or lies down, ensure that their back is aligned with the center of the X-ray machine. When the patient lies on the side, ensure that their spine is in a straight line. At the same time, reasonably set the device parameters of the X-ray device (such as the exposure amount) to ensure that the taken X-ray image is clear and not overexposed. Take this X-ray image as the spinal image of the historical patient after spinal fusion surgery, and upload the taken spinal image to the assisted evaluation system.

[0031] The region acquisition module 101 in the assisted evaluation system uses a neural network to identify and segment the vertebral fusion region in the spinal image of historical patients after spinal fusion surgery, and obtains the initial vertebral fusion region in the spinal image. Among them, when the patient undergoes intervertebral disc fusion surgery, an intervertebral fusion cage will be used for vertebral fusion, and the fusion cage will clearly appear in the image, marking the fusion site. Therefore, the neural network can be used to identify and segment the vertebral fusion region in the spinal image, so as to determine the initial vertebral fusion region in the spinal image. The initial vertebral fusion region refers to the region of the entire fusion site where different vertebral spaces are fused through an intervertebral fusion cage.

[0032] The region correction module 102 is used to correct the initial vertebral fusion region according to the gradient and position distribution of pixel points in the low-change region of the initial vertebral fusion region, so as to obtain the real vertebral fusion region.

[0033] Specifically, the above-mentioned initial vertebral fusion region in the spinal image of the historical patient after spinal fusion surgery is obtained. Since the bone sclerosis around the implant may lead to an unsatisfactory fusion effect between the vertebrae after the patient's surgery, it is necessary to analyze the fusion condition of the vertebral fusion region. However, since the intervertebral width in the vertebral fusion region during the fusion process is not completely filled by growth, there will be some non-fused regions in the intervertebral space of the initially segmented vertebral fusion region. As Figure 3 shown, region a is the non-fused region in a certain initial vertebral fusion region. To avoid the influence of this non-fused region on the analysis of the fusion condition in the vertebral fusion region, the initial vertebral fusion region is corrected to remove this part of the non-fused region, thereby obtaining the true vertebral fusion region.

[0034] Further, in some possible implementation manners, the above-mentioned region correction module 102 includes: an intervertebral space coefficient acquisition unit, configured to determine the intervertebral space coefficient of the low-variation region according to the gradient and position distribution of the pixel points in the low-variation region; a region correction unit, configured to screen out the gap regions in all the low-variation regions according to the intervertebral space coefficient, and remove the gap regions in the initial vertebral fusion region to obtain the true vertebral fusion region, where the intervertebral space coefficient of the gap region is greater than a set coefficient threshold. Among them, the above-mentioned region correction module 102 further includes: a low-variation region acquisition unit, configured to perform Otsu threshold segmentation on the gray values of the pixel points in the initial vertebral fusion region to obtain a plurality of first gray value regions and second gray value regions, and use the second gray value region as the low-variation region of the initial vertebral fusion region, where the gray value of the first gray value region is higher than the gray value of the second gray value region.

[0035] Specifically, perform Otsu threshold segmentation on the gray values of the pixel points in the initial vertebral fusion region, and obtain a plurality of high-threshold regions and low-threshold regions in the initial vertebral fusion region. These two types of regions are respectively referred to as the first gray value region and the second gray value region. The low-threshold region contains several pixel clusters. The low-threshold region of the initial vertebral fusion region includes non-fused regions and some regions with relatively low bone density of the fused bone. Denote the low-threshold region as the low-variation region, and use the canny edge detection algorithm to obtain the gradient of the pixel points in each low-variation region. Since the pixel gradients and pixel positions of the pixel clusters in the non-fused regions and some regions with relatively low bone density of the fused bone in the low-variation region will present different variation characteristics, the intervertebral space coefficient of the low-variation region can be determined by analyzing the gradient and position distribution of the pixel points in the low-variation region to characterize the possibility that the corresponding low-variation region belongs to the non-fused region in the vertebral interspace, thereby realizing the identification and distinction of the non-fused regions and some regions with relatively low bone density of the fused bone in the low-variation region.

[0036] Further, in some possible implementation manners, the above intervertebral space coefficient obtaining unit includes: a pixel clustering determination unit, configured to cluster the pixels in the low-change region according to the gradients of the pixels in the low-change region, and divide the low-change region into a plurality of pixel clusters; a sequence distance determination unit, configured to determine each pixel sequence in the pixel cluster in the vertebral arrangement direction, and a first distance in the direction perpendicular to the vertebral arrangement from the pixel cluster to the initial vertebral fusion region; and an intervertebral space coefficient determination unit, configured to determine the intervertebral space coefficient of the low-change region according to the number of pixels in the pixel sequence in the pixel cluster, the first distance corresponding to the pixel cluster, and the gradients of the pixels in the pixel cluster.

[0037] Specifically, according to the gradients of the pixels in each low-change region, the pixels in each low-change region are clustered, and each low-change region is divided into a plurality of pixel clusters. The gradients of the pixels in the same pixel cluster are similar, and the gradients of the pixels in different pixel clusters are quite different.

[0038] Since the gradients in different pixel clusters in the low-change region are different, it may be that some low-density calluses during the intervertebral growth process in a part of the low-change region. Determine each pixel sequence in the pixel cluster in the vertebral arrangement direction in the obtained low-change region, and determine the number of pixels in each pixel sequence. In this embodiment, since the vertebral arrangement direction is the vertical direction, therefore, determine each column of pixels in the pixel cluster in the vertical direction, and determine the number of pixels in each column of pixels. At the same time, obtain the Euclidean distance between the centroid of each pixel cluster and the initial vertebral fusion region in the direction perpendicular to the vertebral arrangement direction, and record this distance as the first distance. In this embodiment, the direction perpendicular to the vertebral arrangement direction refers to the horizontal direction. Based on the number of pixels in each pixel sequence in the pixel cluster in each low-change region, the first distance corresponding to the pixel cluster, and the gradients of the pixels in the pixel cluster, determine the intervertebral space coefficient of each low-change region.

[0039] Further, in some possible implementation manners, the intervertebral space coefficient determination unit includes: a first arrangement determination unit, configured to arrange all the pixel point clusters in the low-change region in ascending order of the first distance to obtain a first arrangement; a second arrangement determination unit, configured to determine an average gradient of pixel points in the pixel point cluster, and arrange all the pixel point clusters in the low-change region in ascending order of the average gradient to obtain a second arrangement; a first arrangement difference determination unit, configured to determine a total number of pixel point clusters with arrangement differences in the first arrangement and the second arrangement; a third arrangement determination unit, configured to arrange the number of pixel points in the pixel point sequence in the pixel point cluster in ascending order to obtain a third arrangement; a second arrangement difference determination unit, configured to determine a total number of pixel point sequences with arrangement differences between the arrangement order of the pixel point sequence corresponding to the third arrangement and the original arrangement of the pixel point sequence in the pixel point cluster; an intervertebral space coefficient calculation unit, configured to determine a pixel point number increase slope according to the third arrangement, and determine the intervertebral space coefficient of the low-change region according to the total number of pixel point clusters with arrangement differences corresponding to the low-change region, the pixel point number increase slope, and the total number of pixel point sequences.

[0040] Specifically, each pixel point cluster in each low-change region is arranged in ascending order of the first distance between it and the centroid of the initial vertebral fusion region to obtain a first arrangement. The pixel point sequences of the pixel point clusters are arranged in ascending order of the number of pixel points to obtain a third arrangement, and the rate of increase of the number of pixel points from small to large in the third arrangement is obtained, that is, the slope corresponding to the number of pixel points in the third arrangement is determined, and the slope value is recorded as the pixel point number increase slope, and the difference in the number of pixel point sequences between the arrangement order of the pixel point sequence corresponding to the third arrangement and the original arrangement of the pixel point sequence in the pixel point cluster is obtained and recorded as the total number of pixel point sequences with arrangement differences. For example, if the pixel point sequences in the original arrangement of the pixel point cluster are abcde respectively, and the arrangement order of the pixel point sequences corresponding to the third arrangement is abdec respectively, then the total number of pixel point sequences with arrangement differences is 3. At the same time, the pixel point clusters in the low-change region are arranged in ascending order of the average gradient of the pixel points in the cluster to obtain a second arrangement, and it is compared with the first arrangement of the low-change region to obtain the total number of pixel point clusters with arrangement differences. For example, if the numbers of the pixel point clusters in the first arrangement are 123456 respectively, and the numbers of the pixel point clusters in the second arrangement are 134256 respectively, then the total number of pixel point clusters with arrangement differences is 3.

[0041] Furthermore, based on the total number of pixel clusters with arrangement differences corresponding to the low-variation region, the slope of increase in the number of pixel points, and the total number of pixel point sequences with arrangement differences, the intervertebral disc coefficient of the low-variation region is determined. Among them, when both the slope of increase in the number of pixel points of each pixel cluster in the low-variation region and the total number of pixel point sequences with arrangement differences are relatively large, it indicates that the change of pixel clusters in the low-variation region in the direction perpendicular to the vertebral arrangement direction is faster and its change pattern is less in line with normal growth. This is because the faster the change indicates the faster the change in the width of the intervertebral disc, and at this time, it is less in line with the slow change situation during normal growth, and less in line with the characteristic that the number of pixel points in the pixel point sequence in the pixel cluster gradually changes during normal growth; and the larger the value of the total number of pixel clusters with arrangement differences in the low-variation region, the less the distribution of pixel clusters in the low-variation region conforms to the situation of gradually increasing bone density. Therefore, it indicates that the low-variation region is more likely to belong to the intervertebral disc part, and its corresponding intervertebral disc coefficient should be larger.

[0042] Further, in some possible implementation manners, the above-mentioned intervertebral disc coefficient calculation unit includes: a first calculation unit, configured to determine an average value of a first product of the slope of increase in the number of pixel points of all the pixel clusters and the total number of pixel point sequences with arrangement differences, so as to obtain a pixel cluster change index; a second calculation unit: determine a second product of the total number of pixel clusters with arrangement differences corresponding to the low-variation region and the pixel cluster change index, and perform normalization processing on the second product to obtain the intervertebral disc coefficient of the low-variation region.

[0043] In this embodiment, the intervertebral disc coefficient of the low-variation region is determined by the following formula: ; In the formula: represents the intervertebral disc coefficient of the th low-variation region; represents the total number of pixel clusters with arrangement differences corresponding to the th low-variation region; represents the pixel cluster change index of the th low-variation region, that is, the average value of the first product of the slope of increase in the number of pixel points of all pixel clusters and the total number of pixel point sequences with arrangement differences in the th low-variation region; represents a hyperbolic function, used to perform normalization processing on .

[0044] By the above method, the intervertebral space coefficients of all low-variation regions can be obtained. A suitable set coefficient threshold is preset. In this embodiment, the value of the set coefficient threshold is set to 0.5, and the low-variation regions corresponding to the intervertebral space coefficients greater than 0.5 are determined as the interspace regions. All the interspace regions are removed from the initial vertebral fusion region, and the remaining region is the true vertebral fusion region.

[0045] The callus growth coefficient obtaining module 103 is configured to determine a plurality of first intervertebral gradient regions and second intervertebral gradient regions in the true vertebral fusion region, and determine the callus growth coefficient of the true vertebral fusion region according to the positional distribution of the first intervertebral gradient regions and the second intervertebral gradient regions and the gray values of the pixel points, wherein the gradient of the pixel points in the first intervertebral gradient regions is higher than the gradient of the pixel points in the second intervertebral gradient regions.

[0046] Specifically, since the true vertebral fusion region may not be completely fused and there is also interference from the fusion device, it is necessary to further analyze the callus growth situation therein. Considering that when the vertebrae are not successfully fused, inconspicuous calluses will appear, and these inconspicuous calluses usually cause the irregular shape of the vertebral edge, and since the bone tissue density is lower than that of normal bone, the gray value of this part of the inconspicuous callus is also relatively low. Based on this characteristic, the evaluation of the callus growth situation in the true vertebral fusion region can be realized.

[0047] Further, in some possible implementation manners, the above callus growth coefficient obtaining module 103 includes: a threshold segmentation region obtaining unit, configured to perform Otsu threshold segmentation on the gradient of the pixel points in the true vertebral fusion region to obtain a plurality of first intervertebral gradient regions and second intervertebral gradient regions, wherein the gradient of the first intervertebral gradient regions is higher than the gradient of the second intervertebral gradient regions; an information value obtaining unit, configured to determine a second distance from the first intervertebral gradient regions to the vertical direction of the vertebral arrangement direction of the true vertebral fusion region, and determine a distribution difference information value according to the distribution difference of all the second distances; a gray difference value obtaining unit, configured to determine adjacent second intervertebral gradient regions of the first intervertebral gradient regions in the target direction, and determine the gray difference value between the first intervertebral gradient regions and their adjacent second intervertebral gradient regions, wherein the target direction refers to the direction in which the average gray value of each pixel point sequence of the first intervertebral gradient regions in the vertebral arrangement direction gradually decreases in the vertical direction of the vertebral arrangement direction; an adjacent region number obtaining unit, configured to determine the number of adjacent first intervertebral gradient regions of the adjacent second intervertebral gradient regions of the first intervertebral gradient regions; and a callus growth coefficient determining unit, configured to determine the callus growth coefficient of the true vertebral fusion region according to the distribution difference information value, and all the gray difference values and the number of adjacent first intervertebral gradient regions corresponding to the first intervertebral gradient regions.

[0048] Specifically, the pixel points in the real vertebral fusion region are segmented by Otsu threshold with their gradients, so that the real vertebral fusion region is segmented into several high-gradient regions and low-gradient regions, and these two types of regions are respectively called the first intervertebral gradient region and the second intervertebral gradient region. Since the callus generated by incomplete fusion in the vertebra is uneven, at this time, the high-gradient regions and low-gradient regions in the real vertebral fusion region are irregularly arranged. Therefore, the fusion condition of the vertebral fusion region can be obtained by analyzing the distribution of the high-gradient regions and low-gradient regions therein.

[0049] Obtain the centroids of the high-gradient region, that is, the first intervertebral gradient region and the real vertebral fusion region, determine the Euclidean distance between the centroid of the first intervertebral gradient region and the centroid of the real vertebral fusion region in the vertical direction of the vertebral arrangement direction, and record this Euclidean distance as the second distance. Obtain the information entropy of the second distances of all the first intervertebral gradient regions, and record this information entropy as the distance distribution difference information value.

[0050] Obtain each pixel point sequence in the first intervertebral gradient region according to the arrangement mode of the vertebrae, that is, obtain each pixel point sequence in the first intervertebral gradient region in the vertebral arrangement direction, and record the two directions in the vertical direction of the vertebral arrangement direction as the first direction and the second direction. In this embodiment, since the vertebral arrangement direction is the vertical direction, each pixel point sequence is the pixel points of each column in the first intervertebral gradient region in the vertical direction. At this time, the first direction and the second direction respectively refer to the directions from left to right and from right to left of each column of pixel points in the first intervertebral gradient region.

[0051] Obtain the average gray value of each pixel point sequence in the first intervertebral gradient region. If the average gray values of each pixel point sequence as a whole show a gradual decrease in the first direction, then take the first direction as the target direction of the first intervertebral gradient region. If the average gray values of each pixel point sequence as a whole show a gradual decrease in the second direction, then take the second direction as the target direction of the first intervertebral gradient region.

[0052] Since the bone density gradually approaches the normal vertebral bone density due to the callus growth in the vertebral fusion region, the intervertebral gradient region in the target direction of the first intervertebral gradient region should be a region with a lower gray level. Obtain the adjacent second intervertebral gradient regions of each first intervertebral gradient region in its target direction, obtain the difference in the average gray level between each first intervertebral gradient region and its adjacent second intervertebral gradient region, and determine the gray level difference value based on this difference in the average gray level. In this embodiment, the difference in the average gray level is used as the exponent of an exponential function with the natural constant e as the base, and the function value of the exponential function is the gray level difference value. At the same time, obtain the number of adjacent first intervertebral gradient regions of the adjacent second intervertebral gradient regions of each first intervertebral gradient region. For example, for the adjacent second intervertebral gradient region B1 of a certain first intervertebral gradient region A1, if the adjacent second intervertebral gradient region B1 is the adjacent second intervertebral gradient region of other 3 first intervertebral gradient regions, then the number of adjacent first intervertebral gradient regions corresponding to the first intervertebral gradient region A1 is 3.

[0053] Furthermore, based on the distance distribution difference information value, the gray level difference values corresponding to all the first intervertebral gradient regions, and the number of adjacent first intervertebral gradient regions, determine the callus growth coefficient of the true vertebral fusion region. Among them, when the value of the distance distribution difference information value is larger, and the values of the gray level difference values corresponding to all the first intervertebral gradient regions and the number of adjacent first intervertebral gradient regions are also larger, it indicates that the distribution of the first intervertebral gradient regions, that is, the high-gradient regions, in the true vertebral fusion region is more dispersed, and there are more second intervertebral gradient regions, that is, low-gradient regions, around the high-gradient regions. Therefore, it indicates that the growth condition of the callus is better, resulting in a larger callus density. At this time, the callus growth coefficient should be larger.

[0054] Further, in some possible implementation manners, the above callus growth coefficient determination unit includes: a third calculation unit, configured to determine the average value of the third products of the gray level difference values corresponding to all the first intervertebral gradient regions and the number of adjacent first intervertebral gradient regions, to obtain a callus growth index; a fourth calculation unit, configured to determine the fourth product of the distance distribution difference information value and the callus growth index, and perform normalization processing on the fourth product to obtain the callus growth coefficient of the true vertebral fusion region.

[0055] In this embodiment, the callus growth coefficient of the true vertebral fusion region is determined by the following formula: ; In the formula: represents the callus growth coefficient of the true vertebral fusion region; represents the distribution difference information value; It represents the callus growth index, that is, the average value of the third product of the grayscale difference values corresponding to all the first intervertebral gradient regions and the number of adjacent first intervertebral gradient regions; It represents the hyperbolic function and is used to perform normalization processing.

[0056] In the above manner, the callus growth coefficient of the vertebral fusion region in the spinal image of the historical patient after spinal fusion surgery can be determined. This callus growth coefficient reflects the callus growth situation in the vertebral fusion region of the historical patient after spinal fusion surgery. When the callus growth situation is better, the corresponding value of the callus growth coefficient is larger. Thus, the extraction of the callus growth characteristics of the vertebral fusion region is realized.

[0057] The network training module 104 is used to form a training set by using the callus growth coefficients of all historical patients and the postoperative risk grades calibrated for the initial vertebral fusion regions, and use the training set to train the postoperative risk classification network to obtain a trained postoperative risk classification network.

[0058] Specifically, a postoperative risk classification network is constructed, and the network structure of this postoperative risk classification network is Encoder-FC. Manually calibrate the postoperative risk grades for the vertebral fusion regions in the spinal images of all historical patients after spinal fusion surgery. The value range of this postoperative risk grade is [0,1], where the larger the value, the higher the postoperative risk. Use the callus growth coefficients of all historical patients and the postoperative risk grades calibrated for the initial vertebral fusion regions to form a training set, and use this training set to train the constructed postoperative risk classification network to obtain a trained postoperative risk classification network. During the training process, the input of the postoperative risk classification network is the callus growth coefficient of the historical patient, and the output is the postoperative risk grade. The loss function adopted by the postoperative risk classification network is the cross-entropy function, and the gradient descent method is used for training until the loss function converges, and finally the training of the postoperative risk classification network is completed. By training the postoperative risk classification network by extracting the callus growth coefficients related to the postoperative risk in the vertebral fusion region, the postoperative risk classification network can more accurately learn the relationship between the callus growth situation and the postoperative risk, thereby effectively improving the classification accuracy of the postoperative risk classification network.

[0059] When it is necessary to perform an imaging-assisted assessment based on the spinal images of a real-time patient after spinal fusion surgery, the callus growth coefficient corresponding to the spinal images of the real-time patient after spinal fusion surgery is extracted, and the extracted callus growth coefficient is input into the trained postoperative risk classification network. The postoperative risk classification network outputs the postoperative risk grade of the real-time patient, and this postoperative risk grade is provided to the doctor as auxiliary assessment data for reference. Finally, the doctor determines the postoperative risk situation of the real-time patient. It should be emphasized that the postoperative risk grade of the real-time patient output by the postoperative risk classification network is only used as a reference data for auxiliary assessment, which does not represent the final postoperative risk of the patient. The doctor still needs to combine other characteristics of the patient for joint assessment and determine the final postoperative risk situation.

[0060] Based on the same inventive concept, as Figure 2 shown, an embodiment of the present invention further provides an imaging-assisted assessment method for orthopedic spine surgery after operation. The method includes the following steps: Obtain the initial vertebral fusion area in the spinal images of historical patients after spinal fusion surgery; According to the gradient and position distribution of the pixel points in the low-change area of the initial vertebral fusion area, correct the initial vertebral fusion area to obtain the true vertebral fusion area; Determine a plurality of first intervertebral gradient areas and second intervertebral gradient areas in the true vertebral fusion area. According to the position distribution and pixel grayscale values of the first intervertebral gradient areas and the second intervertebral gradient areas, determine the callus growth coefficient of the true vertebral fusion area, where the gradient of the pixel points in the first intervertebral gradient area is higher than the gradient of the pixel points in the second intervertebral gradient area; Use the callus growth coefficients of all historical patients and the postoperative risk grades calibrated by the initial vertebral fusion areas to form a training set, and use the training set to train the postoperative risk classification network to obtain a trained postoperative risk classification network.

[0061] Based on the same inventive concept, an embodiment of the present invention further provides an imaging-assisted assessment device for orthopedic spine surgery after operation. As Figure 4 shown, the device includes: a memory 401, a processor 402, and computer program code 403 stored in the memory 401 and running on the processor 402. When the processor 402 executes the computer program code 403, the device can execute the module implementation steps in any one of the aforementioned imaging-assisted assessment systems for orthopedic spine surgery after operation.

[0062] In the embodiments of the present invention, the device can be divided into functional modules according to the step examples of the modules in the above system. 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 illustrative, only a logical function division, and there can be other division methods in actual implementation.

[0063] Based on the same inventive concept, the embodiments of the present invention also provide a computer program product, which includes: computer program code. When the computer program code runs on a computer, the computer is enabled to execute the module implementation steps in any one of the imaging-assisted evaluation systems for orthopedic spine surgery described above.

[0064] Based on the same inventive concept, the embodiments of the present invention also provide a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer is enabled to execute the module implementation steps in any one of the imaging-assisted evaluation systems for orthopedic spine surgery described above.

[0065] It should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An imaging-assisted evaluation system for orthopedic spine surgery, characterized in that: The system comprises: A region acquisition module for acquiring an initial vertebral fusion region in a spinal image of a historical patient after spinal fusion surgery; A region correction module, used for correcting the initial vertebral fusion region according to the gradient and position distribution of the pixel points in the low-variation region of the initial vertebral fusion region to obtain a real vertebral fusion region; a callus growth coefficient acquisition module, used to determine a plurality of first intervertebral gradient regions and second intervertebral gradient regions in the actual vertebral fusion region, and determine the callus growth coefficient of the actual vertebral fusion region according to the position distribution of the first intervertebral gradient regions and the second intervertebral gradient regions and the grayscale values ​​of the pixels, wherein the gradient of the pixels in the first intervertebral gradient region is higher than the gradient of the pixels in the second intervertebral gradient region; The network training module is used to form a training set using the callus growth coefficients of all historical patients and the postoperative risk classification marked by the initial vertebral fusion area, and to train the postoperative risk classification network using the training set to obtain a trained postoperative risk classification network.

2. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 1, characterized in that: The area correction module comprises: An intervertebral space coefficient acquisition unit, used for determining the intervertebral space coefficient of the low-variation area according to the gradient and position distribution of the pixel points in the low-variation area; The area correction unit is used to screen out the gap areas in all the low-variation areas according to the intervertebral space coefficient, remove the gap areas in the initial vertebral fusion area, and obtain the real vertebral fusion area, wherein the intervertebral space coefficient of the gap area is greater than a set coefficient threshold.

3. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 2, characterized in that: The intervertebral space coefficient acquisition unit comprises: A pixel point cluster determination unit, configured to cluster the pixels in the low-variation area according to the gradients of the pixels in the low-variation area, and divide the low-variation area into a plurality of pixel point clusters; a sequence distance determination unit, configured to determine each pixel point sequence in the pixel point cluster in the vertebra arrangement direction, and a first distance from the pixel point cluster to the initial vertebral fusion region in the vertical direction of the vertebra arrangement; The intervertebral space coefficient determining unit is used to determine the intervertebral space coefficient of the low-variation area according to the number of pixels in the pixel sequence in the pixel cluster, the first distance corresponding to the pixel cluster and the gradient of the pixels in the pixel cluster.

4. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 3, characterized in that: The intervertebral space coefficient determination unit comprises: A first arrangement determining unit, configured to arrange all the pixel clusters in the low-variance area in ascending order of the first distances to obtain a first arrangement; A second arrangement determining unit, configured to determine an average gradient of pixels in the pixel clusters, and arrange all the pixel clusters in the low-variance region in ascending order of the average gradient to obtain a second arrangement; A first arrangement difference determining unit, used to determine the total number of pixel point clusters having arrangement differences between the first arrangement and the second arrangement; A third arrangement determining unit, configured to arrange the number of pixels in the pixel sequence in the pixel cluster in ascending order to obtain a third arrangement; A second arrangement difference determining unit is used to determine the total number of pixel point sequences whose arrangement order of the pixel point sequence corresponding to the third arrangement is different from the original arrangement of the pixel point sequence in the pixel point cluster; The intervertebral space coefficient calculation unit is used to determine the increasing slope of the number of pixels according to the third arrangement, and determine the intervertebral space coefficient of the low-variance area according to the total number of pixel clusters with arrangement differences corresponding to the low-variance area, the increasing slope of the number of pixels and the total number of pixel sequences.

5. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 4, characterized in that: The intervertebral space coefficient calculation unit comprises: A first calculation unit is used to determine an average value of a first product of an increase slope of the number of pixels in all the pixel clusters and a total number of pixel sequences with arrangement differences, to obtain a pixel cluster change index; The second calculation unit determines a second product of the total number of pixel clusters with arrangement differences corresponding to the low-variation area and the pixel cluster change index, and normalizes the second product to obtain the intervertebral space coefficient of the low-variation area.

6. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 2, characterized in that: The area correction module also includes: The low-variation region acquisition unit is used to perform Otsu threshold segmentation on the grayscale values ​​of the pixels in the initial vertebral fusion region to obtain a plurality of first grayscale value regions and second grayscale value regions, and use the second grayscale value regions as the low-variation regions of the initial vertebral fusion region, wherein the grayscale values ​​of the first grayscale value regions are higher than the grayscale values ​​of the second grayscale value regions.

7. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 1, characterized in that: The callus growth coefficient acquisition module includes: an information value acquisition unit, configured to determine a second distance in a direction perpendicular to the vertebral arrangement direction from the first intervertebral gradient region to the real vertebral fusion region, and determine a distance distribution difference information value according to distribution differences of all the second distances; a grayscale difference value acquisition unit, used for determining an adjacent second intervertebral gradient region of the first intervertebral gradient region in a target direction, and determining a grayscale difference value between the first intervertebral gradient region and its adjacent second intervertebral gradient region, wherein the target direction refers to a direction in which the average grayscale value of each pixel point sequence in the first intervertebral gradient region in the vertebral arrangement direction gradually decreases in a direction perpendicular to the vertebral arrangement direction; an adjacent region quantity acquisition unit, configured to determine the quantity of adjacent first intervertebral gradient regions of the adjacent second intervertebral gradient region of the first intervertebral gradient region; The callus growth coefficient determination unit is used to determine the callus growth coefficient of the actual vertebral fusion area according to the distance distribution difference information value, the grayscale difference values ​​corresponding to all the first intervertebral gradient areas, and the number of adjacent first intervertebral gradient areas.

8. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 7, characterized in that: The callus growth coefficient determination unit comprises: a third calculation unit, configured to determine an average value of a third product of the grayscale difference values ​​corresponding to all the first intervertebral gradient regions and the number of adjacent first intervertebral gradient regions, to obtain a callus growth index; The fourth calculation unit is used to determine a fourth product of the distance distribution difference information value and the callus growth index, and normalize the fourth product to obtain the callus growth coefficient of the actual vertebral fusion area.

9. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 7, characterized in that: The information value acquisition unit includes: The information value calculation unit is used to determine the information entropy of the second distance of all the first intervertebral gradient areas, and use the information entropy as the distance distribution difference information value.

10. The imaging-assisted evaluation system for orthopedic spine surgery according to claim 1, characterized in that: The callus growth coefficient acquisition module comprises: The threshold segmentation region acquisition unit is used to perform Otsu threshold segmentation on the gradient of the pixel points in the real vertebral fusion region to obtain a plurality of first intervertebral gradient regions and second intervertebral gradient regions, wherein the gradient of the first intervertebral gradient region is higher than the gradient of the second intervertebral gradient region.

Images