Computer-aided plastic surgery aided navigation method and system

Abstract

The invention relates to the technical field of image processing, in particular to a plastic surgery aided navigation method and system based on computer aiding.The method comprises the steps that a skeleton image is collected, suspected pixel points are obtained according to a gray histogram of the skeleton image, edge vectors are constructed according to skeleton edges of the skeleton image, and all center pixel points are obtained; the skeleton roughness and the edge included angle of each center pixel point are obtained, skeleton roughness abnormal factor parameters of each suspected pixel point are calculated, and all abnormal pixel points are obtained in combination with the gray value and the distance between each suspected pixel point and the nearest skeleton edge; all target pixel points are obtained according to the gray value difference between each abnormal pixel point and the surrounding pixel points, the target pixel points are clustered to obtain a plurality of target clusters, and the suspected bone uplift abnormal area is obtained after segmentation. The method aims at solving the problem of wrong segmentation caused by the fact that seed points are arranged at low-density bones, and the accuracy of bone uplift abnormal region segmentation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a computer-aided plastic surgery assistance navigation method and system. Background Art

[0002] Since most plastic surgeries involve craniofacial bones and the central nervous system, the craniofacial changes are diverse, its anatomical structure is complex, the risks involved in the cranio-maxillofacial region are high, the processing difficulty is high, and it is difficult to obtain satisfactory surgical results. With the rapid development and rapid application of computer hardware and artificial intelligence technologies, by using high-performance computers and software analysis algorithms for the real structure of the human body, the diagnostic images obtained before surgery are registered with the surgical field, so as to realize the construction of a three-dimensional model of the plastic surgery site.

[0003] When the machine vision detection technology realizes the precise positioning of the abnormal bone eminence area, the watershed algorithm is commonly used to segment the abnormal bone eminence area in the bone eminence CT image. The selection of the seed points of the watershed has a great influence on the segmentation effect of the abnormal bone surface area. The segmentation effect is the best when the seed points of the watershed are selected inside the abnormal bone surface area; since the abnormal bone surface area often appears as an area with a lower gray value in the CT image compared with the normal area, the watershed algorithm often selects the pixel point with the lowest gray value in the CT image as the seed point of the watershed algorithm. However, due to the different bone densities at different positions in the bone area, the gray value is also lower in the low-density bone area compared with the normal-density bone area, and directly selecting the pixel point with the lowest gray value as the seed point may misplace the seed point in the low-density bone area, thus affecting the segmentation accuracy of the watershed algorithm for the abnormal bone eminence area, and further affecting the accuracy of the subsequent orthopedic surgery assistance navigation technology. Summary of the Invention

[0004] The present invention provides a computer-aided plastic surgery assistance navigation method and system to solve the problem that the existing method of selecting the pixel point with the lowest gray value as the seed point may misplace the seed point in the low-density bone area, thus affecting the segmentation accuracy of the watershed algorithm for the abnormal bone eminence area, and further affecting the accuracy of the subsequent orthopedic surgery assistance navigation technology.

[0005] The computer-aided plastic surgery assistance navigation method and system of the present invention adopt the following technical solutions: An embodiment of the present invention provides a computer-aided plastic surgery assistance navigation method, and the method includes the following steps: Collect bone CT images to obtain bone images; Obtain suspected pixel points according to the gray histogram of the bone image; Construct an edge vector based on the bone edges of the bone image to obtain all the central pixel points of the bone image; construct a thickness line for each central pixel point of the bone image to obtain the first intersection point and the second intersection point of each central pixel point; obtain the bone thickness and edge angle of each central pixel point based on the first intersection point and the second intersection point of each central pixel point; obtain the bone thickness anomaly factor parameter of each suspected pixel point based on the bone thickness and edge angle of the central pixel points on the same thickness line as the suspected pixel point and the gray value of the suspected pixel point; Obtain all abnormal pixel points based on the bone thickness anomaly factor parameter and gray value of each suspected pixel point, and the distance of each suspected pixel point from the nearest bone edge; Obtain all target pixel points based on the gray value difference between each abnormal pixel point and its surrounding pixel points; Cluster the target pixel points to obtain several target clustering clusters; segment the target clustering clusters to obtain the suspected bone bulge abnormal area; the suspected bone bulge abnormal area is used to formulate a surgical assistance navigation method.

[0006] Further, the obtaining of the suspected pixel points according to the gray histogram of the bone image includes the following specific steps: Use the GMM Gaussian mixture model to fit the number of pixel points at each gray level in the gray histogram of the bone image to obtain the gray distribution fitting curve of the bone image, and use the AMPD peak detection algorithm to extract the gray distribution fitting curve of the bone image to obtain the peaks and valleys of the bone image; Take the gray level corresponding to the valley of the bone image as the segmentation threshold of the bone image, and record all the pixel points with gray values less than the segmentation threshold in the bone image as suspected pixel points.

[0007] Further, the obtaining of all the central pixel points of the bone image by constructing an edge vector based on the bone edges of the bone image includes the following specific steps: Use the Canny edge detection algorithm to obtain the edge curve of the bone image, input all the edge curves of the bone image into the Hough line detection algorithm to obtain several detected lines, select the two longest detected lines among them as the edge pixel lines of the bone image, and each edge pixel line of the bone image corresponds to a bone edge; construct two edge vectors according to the two edge pixel lines of the bone image, and the specific construction method of the edge vector is: record the direction from the left side of the maxillofacial region to the right side of the maxillofacial region of the edge pixel line as the vector direction of the edge vector, and record the length of the edge pixel line as the modulus value of the edge vector; Add the two edge vectors to obtain the superimposed vector of the bone image, and record all the pixel points passed by the superimposed vector of the bone image in the bone image as the central pixel points of the bone image.

[0008] Further, constructing a thickness line based on each central pixel point of the bone image to obtain a first intersection point and a second intersection point of each central pixel point includes the following specific steps: Taking the normal direction of the superimposed vector of the bone image as the line, denoted as the thickness line. The thickness line intersects two bone edges of the bone image at two points after passing through each central pixel point, which are respectively denoted as the first intersection point and the second intersection point of each central pixel point.

[0009] Further, obtaining the bone thickness and edge angle of each central pixel point based on the first intersection point and the second intersection point of each central pixel point includes the following specific steps: Denoting the distance between the first intersection point and the second intersection point of each central pixel point as the bone thickness of the central pixel point; denoting the included angle between the gradient vectors of the first intersection point and the second intersection point of each central pixel point as the edge angle of each central pixel point.

[0010] Further, obtaining the bone thickness anomaly factor parameter of each suspected pixel point based on the bone thickness and edge angle of the central pixel points on the same thickness line as the suspected pixel point, and the gray value of the suspected pixel point. The corresponding specific formula is as follows: In the formula, represents the bone thickness anomaly factor parameter of the k-th suspected pixel point in the bone image; represents the edge angle of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image; represents the gray value of the k-th suspected pixel point in the bone image; represents the bone thickness of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image; represents the logarithmic function with the natural constant as the base; is the linear normalization function, is the absolute value function.

[0011] Further, obtaining all abnormal pixel points based on the bone thickness anomaly factor parameter and gray value of each suspected pixel point, and the distance of each suspected pixel point from the nearest bone edge includes the following specific steps: Obtaining the anomaly degree value of each suspected pixel point in the bone image according to the bone thickness anomaly factor parameter of each suspected pixel point in the bone image and the gray value of each suspected pixel point; Among them, the calculation method of the anomaly degree value of the k-th suspected pixel point in the bone image is: In the formula, represents the abnormality degree value of the k-th suspected pixel point in the bone image, represents the bone thickness abnormality factor parameter of the k-th suspected pixel point in the bone image; represents the gray value of the k-th suspected pixel point in the bone image; represents the distance from the k-th suspected pixel point in the bone image to the nearest bone edge; is a linear normalization function, is an absolute value function, is an exponential function with the natural constant as the base; A preset suspected threshold. When the abnormality degree value of the k-th suspected pixel point is greater than the preset suspected threshold, the k-th suspected pixel point is marked as an abnormal pixel point.

[0012] Furthermore, the step of obtaining all target pixel points according to the gray value difference between each abnormal pixel point and its surrounding pixel points includes the following specific steps: Construct a neighborhood range centered on each abnormal pixel point. According to the abnormality degree value of each abnormal pixel point in the bone image and the difference between each abnormal pixel point and the gray values of the surrounding pixel points within the neighborhood range, obtain the target degree value of each abnormal pixel point in the bone image; Among them, the calculation method of the target degree value of the n-th abnormal pixel point is: In the formula, represents the target degree value of the n-th abnormal pixel point; represents the gray value of the n-th abnormal pixel point; represents the gray value of the m-th pixel point within the neighborhood range of the n-th abnormal pixel point; represents the distance between the n-th abnormal pixel point and the m-th pixel point within the neighborhood range; represents the maximum side length of the neighborhood range; is the abnormality degree value of the n-th abnormal pixel point in the bone image; is a linear normalization function, is an absolute value function; A preset abnormal threshold. When the target degree value of the n-th abnormal pixel point is greater than the abnormal threshold, the n-th abnormal pixel point is marked as a target pixel point.

[0013] Furthermore, the step of clustering the target pixel points to obtain several target clustering clusters; and segmenting the target clustering clusters to obtain the suspected bone bulge abnormal area includes the following specific steps: Taking the grayscale value and pixel position of the target pixel as input, applying the DBSCAN density clustering algorithm to the target pixel to obtain several target clustering clusters, calculating the mean value of the target degree values of all target pixels in each target clustering cluster, and selecting the target clustering cluster with the highest mean value of the target degree value as the high-probability bone eminence abnormal area; Using the watershed algorithm to select the suspected bone eminence abnormal area, selecting the pixel with the lowest grayscale value in the high-probability bone eminence abnormal area as the watershed seed point, and performing watershed segmentation on the bone image according to the selected watershed seed point. The area of the segmentation result is recorded as the suspected bone eminence abnormal area.

[0014] The present invention also proposes a computer-aided plastic surgery assistance navigation system, including a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned computer-aided plastic surgery assistance navigation method.

[0015] The beneficial effects of the technical solution of the present invention are as follows: collecting a bone image, obtaining suspected pixels according to the grayscale histogram of the bone image, constructing an edge vector based on the bone edge of the bone image to obtain all central pixels, constructing a thickness line to obtain the first intersection point and the second intersection point of each central pixel, obtaining the bone thickness and edge angle of each central pixel, performing targeted analysis on the abnormal area in combination with the characteristics of the bone eminence abnormal area, calculating the bone thickness abnormal factor parameter of each suspected pixel, obtaining all abnormal pixels in combination with the grayscale value and the distance of each suspected pixel from the nearest bone edge, by considering the positive proportional relationship between bone density and grayscale value, improving the accuracy of selecting the watershed segmentation seed point, obtaining all target pixels according to the grayscale value difference between each abnormal pixel and its surrounding pixels, and further improving the accuracy of selecting the watershed seed point by analyzing the positional relationship of each suspected pixel inside the bone image, clustering the target pixels to obtain several target clustering clusters, and segmenting the suspected bone eminence abnormal area, improving the accuracy of subsequent orthopedic surgery assistance navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions 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, without creative efforts, other drawings can be obtained according to these drawings.

[0017] Figure 1 It is a flowchart of the steps of the computer-aided plastic surgery assistance navigation method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and effects of the computer-aided plastic surgery assistance navigation method and system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0020] The following specifically describes the specific solutions of the computer-aided plastic surgery assistance navigation method and system provided by the present invention with reference to the accompanying drawings.

[0021] Please refer to Figure 1 , which shows a flowchart of the steps of a computer-aided plastic surgery assistance navigation method provided by an embodiment of the present invention. The method includes the following steps: Step S001: Collect bone CT images to obtain bone images.

[0022] The purpose of this embodiment is to extract the abnormal bone prominence regions in the bone CT images. Therefore, it is first necessary to use a CT device to collect bone CT images.

[0023] Furthermore, since the bone prominence is located inside the bone, and the collected bone CT images are interfered by muscles and blank black backgrounds, it is necessary to input the obtained bone CT images into the trained neural network to obtain the bone regions in the bone CT images; the neural network used in this embodiment is ResNET, and the acquisition method of the data set for training this neural network is as follows: collect a large number of bone CT images, artificially mark the bone regions in each bone CT image with bounding boxes, and record the marking results as the labels of each bone CT image; obtain the training data set by marking all bone CT images, and use this training data set to train this neural network. The loss function used during the training process is the mean square error loss function; the specific training process is well-known content of the neural network, and the specific training process is not described in detail in this embodiment.

[0024] Denoise each bone region using the Gaussian filtering algorithm, and the denoising result of each bone region is a bone image. It should be noted that the Gaussian filtering algorithm is a well-known existing technology, and it is not described in detail in this embodiment.

[0025] Thus, bone images are obtained.

[0026] Step S002: Obtain suspected pixel points according to the grayscale histogram of the bone image.

[0027] It should be noted that the bone image includes abnormal bone protrusion areas with lower grayscale values, normal bone areas with lower grayscale values ​​and low bone density, and normal bone areas with higher grayscale values ​​and high bone density. The grayscale values ​​of the normal areas with high bone density are larger than those of the abnormal bone protrusion areas and the normal bone areas with low bone density. The grayscale histogram of the bone image shows a more obvious bimodal nature, so the normal areas with high bone density can be screened out based on the characteristics of the grayscale histogram.

[0028] Specifically, this embodiment takes any one bone image as an example for analysis, and the analysis methods of other bone images are the same; the grayscale histogram of the bone image is obtained, and the number of pixels of each gray level in the grayscale histogram of the bone image is fitted using the GMM Gaussian mixture model to obtain the grayscale distribution fitting curve of the bone image, and the AMPD peak detection algorithm is used to extract the grayscale distribution fitting curve of the bone image to obtain the peaks and troughs of the bone image. It should be noted that the GMM Gaussian mixture model and the AMPD peak detection algorithm described in this embodiment are both existing well-known technologies, and this embodiment will not be described in detail.

[0029] Furthermore, since the grayscale histogram of the bone image presents a bimodal property, the pixels in the smaller peak range correspond to abnormal bone protrusion areas with lower grayscale values ​​or normal areas with low bone density. Therefore, this embodiment uses the grayscale level corresponding to the trough of the bone image as the segmentation threshold of the bone image, and records all pixels in the bone image with grayscale values ​​less than the segmentation threshold as suspected pixels.

[0030] At this point, all the suspected pixels in the bone image are obtained.

[0031] Step S003: Obtain the bone thickness abnormality factor parameter of each suspected pixel point according to the gray value and gradient direction of each suspected pixel point in the bone image.

[0032] It should be noted that in order to further determine whether the suspected pixel position is in the abnormal bone protrusion area or the normal area of ​​low bone density, it is necessary to analyze the pixel grayscale and bone characteristics. The bone density is different at different bone thickness and bone positions, and there is a proportional regular relationship between bone density and grayscale value, and the grayscale value of the pixel in the abnormal bone protrusion area will destroy this proportional regular relationship; the smaller the direction angle of the gradient vector of the edge pixel of the bone image, the smaller the angle formed by the straight line segment where the edge pixel of the bone image is located, and the smaller the bone thickness is in the middle part of the bone. Therefore, this embodiment obtains the bone thickness abnormality factor parameter of the suspected pixel through the relationship between the suspected pixel and the bone thickness of the suspected pixel position, and the grayscale change relationship of the bone position.

[0033] Specifically, the Canny edge detection algorithm is used to obtain the edge curves of the bone image. All the edge curves of the bone image are input into the Hough line detection algorithm to obtain several detected lines. The two longest detected lines are selected as the edge pixel lines of the bone image, and each edge pixel line of the bone image corresponds to a bone edge. Two edge vectors are constructed based on the two edge pixel lines of the bone image. The specific construction method of the edge vector is as follows: the direction from the left side of the maxillofacial region to the right side of the maxillofacial region of the edge pixel line is denoted as the vector direction of the edge vector, and the length of the edge pixel line is denoted as the modulus value of the edge vector. It should be noted that the Hough line detection described in this embodiment is a well-known prior art, and will not be elaborated too much in this embodiment.

[0034] Further, the two edge vectors are added to obtain the superimposed vector of the bone image. All the pixel points passed by the superimposed vector of the bone image in the bone image are denoted as the central pixel points of the bone image; the line made in the normal direction of the superimposed vector of the bone image is denoted as the thickness line. The thickness line intersects the two bone edges of the bone image at two points after passing through each central pixel point, which are respectively denoted as the first intersection point and the second intersection point of each central pixel point. The distance between the first intersection point and the second intersection point of each central pixel point is denoted as the bone thickness at the central pixel point; the included angle between the gradient vectors of the first intersection point and the second intersection point of each central pixel point is denoted as the edge included angle of each central pixel point.

[0035] Further, in this embodiment, the bone thickness anomaly factor parameter of each suspected pixel point is obtained according to the bone thickness and edge included angle of the central pixel point on the same thickness line as the suspected pixel point, and the gray value of the suspected pixel point.

[0036] Specifically, the calculation method of the bone thickness anomaly factor parameter of the k-th suspected pixel point in the bone image is as follows: In the formula, represents the bone thickness anomaly factor parameter of the k-th suspected pixel point in the bone image; represents the edge included angle of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image; represents the gray value of the k-th suspected pixel point in the bone image; represents the bone thickness of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image; represents the logarithmic function with the natural constant as the base; is a linear normalization function that normalizes the data value to the interval [0, 1]; is the absolute value function.

[0037] Take the arctangent of the included angle, which represents the product of the distance and the arctangent value. The larger the product, the larger the edge included angle of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image, and the larger the bone thickness at this location; is the ratio of the bone thickness parameter corresponding to the k-th suspected pixel point in the bone image to the gray value of the k-th suspected pixel point. The closer the ratio is to 1, the more proportional the gray value of the k-th suspected pixel point is to the bone thickness, and the smaller the probability that the k-th suspected pixel point is in the abnormal area of bone prominence.

[0038] Similarly, according to the relationship between the bone thickness and the gray value change of the bone position for each suspected pixel point in the bone image, the bone thickness anomaly factor parameter of each suspected pixel point is obtained.

[0039] Thus far, the bone thickness anomaly factor parameters of each suspected pixel point have been obtained.

[0040] Step S004: According to the bone thickness anomaly factor parameter of each suspected pixel point in the bone image and the gray value of each suspected pixel point, obtain the anomaly degree value of each suspected pixel point in the bone image; obtain the abnormal pixel points in the bone image according to the anomaly degree value of each suspected pixel point in the bone image.

[0041] It should be noted that since there is more tissue inside the bone than at the bone edge, the bone density at the bone edge is relatively higher than that inside the bone. When there is no bone prominence inside the bone image, the closer the suspected pixel point is to the bone edge, the higher the bone density and the larger the gray value. Due to the existence of bone prominence, the suspected pixel points in the abnormal area of bone prominence disrupt this gray regularity. For example, at a location relatively close to the bone edge, the gray value of the suspected pixel point in the abnormal area of bone prominence is darker. Therefore, in this embodiment, the anomaly degree value of the suspected pixel point is obtained according to the gray value of the suspected pixel point, the distance from the suspected pixel point to the bone edge, and the bone thickness anomaly factor of the suspected pixel point.

[0042] Specifically, the calculation method of the anomaly degree value of the k-th suspected pixel point in the bone image is as follows: In the formula, represents the anomaly degree value of the k-th suspected pixel point in the bone image, represents the bone thickness anomaly factor parameter of the k-th suspected pixel point in the bone image; represents the gray value of the k-th suspected pixel point in the bone image; represents the distance from the k-th suspected pixel point in the bone image to the nearest bone edge; is a linear normalization function that normalizes the data value to the interval [0,1], is the absolute value function, is the exponential function with the natural constant as the base.

[0043] The closer it is to 1, the less the k-th suspected pixel point satisfies the gray-scale change rule that the closer to the bone edge, the greater the gray-scale value, and the greater the possibility that it is a pixel point in the abnormal area of bone bulge; As a confidence parameter, the larger this value, the higher the credibility of the abnormality degree of the obtained pixel point.

[0044] Furthermore, a preset suspected threshold is set. In this embodiment, the suspected threshold is for description. In other embodiments, it can be set to other values, which are not limited in this embodiment; when the abnormality degree value of the k-th suspected pixel point satisfies , the k-th suspected pixel point is marked as an abnormal pixel point; when the abnormality degree value of the k-th suspected pixel point satisfies , the k-th suspected pixel point is not processed.

[0045] Similarly, according to the gray-scale value of each suspected pixel point and the distance from the suspected pixel point to the bone edge, combined with the bone thickness abnormality factor of each suspected pixel point, the abnormality degree value of the suspected pixel point is obtained, and the suspected threshold is used to judge the abnormality degree of each suspected pixel point to obtain all abnormal pixel points of the bone image.

[0046] So far, the abnormality degree value of each suspected pixel point has been obtained, and all abnormal pixel points in the bone image are obtained according to the abnormality degree value of each suspected pixel point in the bone image.

[0047] Step S005: According to the difference between the abnormality degree value of each abnormal pixel point in the bone image and the gray-scale value of the surrounding pixel points, obtain the target degree value of each abnormal pixel point in the bone image; according to the target degree value of each abnormal pixel point, obtain the target pixel points of the bone image.

[0048] It should be noted that the gray-scale of the non-bone bulge abnormal area in the bone image is more uniform than that of the bone bulge abnormal area, and in the bone bulge, due to the presence of some connective tissues and blood vessels, the gray-scale performance inside the bone bulge in the CT image is more chaotic and non-uniform compared with the normal bone area. Therefore, in this embodiment, the target degree value is determined according to the gray-scale uniformity characteristics of the abnormal pixel point in its different neighborhoods, and then the target pixel points are obtained according to the target degree value.

[0049] Specifically, the calculation method of the target degree value of the n-th abnormal pixel point is: In the formula, represents the target degree value of the n-th abnormal pixel point; represents the gray value of the nth abnormal pixel point; represents the gray value of the mth pixel point within the neighborhood range of the nth abnormal pixel point; represents the distance between the nth abnormal pixel point and the mth pixel point within the neighborhood range; represents the maximum side length of the neighborhood range. In this embodiment, the preset selection range is described by taking the pixel point area with a radius of 5 centered on the abnormal pixel point as an example. In other embodiments, it can be set to other values, which are not limited in this embodiment; is the abnormality degree value of the nth abnormal pixel point in the bone image, is a linear normalization function that normalizes the data value to within the interval, is the absolute value function.

[0050] Subtract from and take the absolute value. The larger this absolute value is, the greater the difference in pixels within the current neighborhood range, and the worse the uniformity of the current neighborhood; Take the reciprocal of , that is, the closer the currently traversed neighborhood pixel point is to the center point, the higher the credibility of the result obtained by taking the absolute value of the difference between the gray values of the two pixels; is the neighborhood gray uniformity, and the gray uniformity features within different neighborhood ranges are superimposed; is used as the confidence parameter, that is, the larger this value is, the higher the credibility of the target degree value of the obtained abnormal pixel point.

[0051] Furthermore, preset the abnormal threshold , which is described as the suspected threshold in this embodiment. In other embodiments, it can be set to other values, which are not limited in this embodiment; when the target degree value of the nth abnormal pixel point satisfies , mark the nth abnormal pixel point as the target pixel point; when the target degree value of the nth abnormal pixel point satisfies , do not process the nth abnormal pixel point.

[0052] Similarly, based on the difference between the gray value of each abnormal pixel point and the gray values of the surrounding pixel points, combined with the abnormality degree value of each abnormal pixel point, use the suspected threshold to judge the target degree value of each abnormal pixel point to obtain all the target pixel points in the bone region.

[0053] Thus far, the target degree value of each abnormal pixel point has been obtained, and all the target pixel points in the bone image have been obtained based on the target degree value of each abnormal pixel point.

[0054] Step S006: Cluster the target pixel points in the bone image to obtain several target clustering clusters; obtain the high-probability bone eminence abnormal area according to the target degree values of all target pixel points in the target clustering clusters; use the watershed algorithm for the high-probability bone eminence abnormal area to obtain the suspected bone eminence abnormal area; the suspected bone eminence abnormal area is used to formulate a suitable surgical assistance navigation method.

[0055] Specifically, taking the gray value and pixel position of the target pixel points as inputs, use the DBSCAN density clustering algorithm for the target pixel points to obtain several target clustering clusters, calculate the mean value of the target degree values of all target pixel points in each target clustering cluster, and select the target clustering cluster with the highest mean value of the target degree values as the high-probability bone eminence abnormal area. It should be noted that DBSCAN density clustering is a well-known existing technology, and this embodiment will not elaborate too much.

[0056] Furthermore, use the watershed algorithm to select the suspected bone eminence abnormal area. Select the pixel point with the lowest gray value in the high-probability bone eminence abnormal area as the watershed seed point, perform watershed segmentation on the bone image according to the selected watershed seed point, and record the area of the segmentation result as the suspected bone eminence abnormal area. The suspected bone eminence abnormal area is used to formulate a suitable surgical assistance navigation method. It should be noted that the watershed algorithm is a well-known existing technology, and this embodiment will not elaborate too much.

[0057] In a second aspect, the present invention also provides a computer-aided plastic surgery assistance navigation system, including a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned computer-aided plastic surgery assistance navigation method.

[0058] It should be noted that the model used in this embodiment only represents a negative correlation relationship and restricts the output result of the model to be within the interval, where is the input of this model. In specific implementation, it can be replaced with other models with the same purpose. This embodiment only takes the model as an example for description and does not make specific limitations.

[0059] So far, the present invention is completed.

[0060] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A computer-aided plastic surgery assisted navigation method, characterized in that, The method includes the following steps: Collect bone CT images to obtain bone images; Obtain suspected pixel points according to the gray histogram of the bone image; Construct edge vectors based on the bone edges of the bone image to obtain all central pixel points of the bone image; construct thickness lines according to each central pixel point of the bone image to obtain the first intersection point and the second intersection point of each central pixel point; obtain the bone thickness and edge angle of each central pixel point according to the first intersection point and the second intersection point of each central pixel point; obtain the bone thickness anomaly factor parameter of each suspected pixel point according to the bone thickness and edge angle of the central pixel points on the same thickness line as the suspected pixel point, and the gray value of the suspected pixel point; Obtain all abnormal pixel points according to the bone thickness anomaly factor parameter and gray value of each suspected pixel point, and the distance of each suspected pixel point from the nearest bone edge; Obtain all target pixel points according to the gray value difference between each abnormal pixel point and its surrounding pixel points; Cluster the target pixel points to obtain several target cluster clusters; segment the target cluster clusters to obtain suspected bone bulge abnormal regions; the suspected bone bulge abnormal regions are used to formulate a surgical assistance navigation method; The step of obtaining suspected pixel points according to the gray histogram of the bone image includes the following specific steps: Use the GMM Gaussian mixture model to fit the number of pixel points at each gray level in the gray histogram of the bone image to obtain the gray distribution fitting curve of the bone image, and use the AMPD peak detection algorithm to extract the gray distribution fitting curve of the bone image to obtain the peaks and valleys of the bone image; Take the gray level corresponding to the valley of the bone image as the segmentation threshold of the bone image, and record all pixel points in the bone image with gray values less than the segmentation threshold as suspected pixel points; The step of constructing edge vectors based on the bone edges of the bone image to obtain all central pixel points of the bone image includes the following specific steps: Use the Canny edge detection algorithm to obtain the edge curve of the bone image, input all edge curves of the bone image into the Hough line detection algorithm to obtain several detected lines, select the two longest detected lines as the edge pixel lines of the bone image, and each edge pixel line of the bone image corresponds to a bone edge; construct two edge vectors according to the two edge pixel lines of the bone image. The specific construction method of the edge vector is: record the direction from the left side of the maxillofacial surface to the right side of the maxillofacial surface of the edge pixel line as the vector direction of the edge vector, and record the length of the edge pixel line as the modulus value of the edge vector; Add the two edge vectors to obtain the superimposed vector of the bone image, and record all pixel points passed by the superimposed vector of the bone image in the bone image as the central pixel points of the bone image; The step of constructing thickness lines according to each central pixel point of the bone image to obtain the first intersection point and the second intersection point of each central pixel point includes the following specific steps: Take the normal direction of the superimposed vector of the bone image as a straight line and record it as the thickness line. The thickness line intersects the two bone edges of the bone image at two points after passing through each central pixel point, which are respectively recorded as the first intersection point and the second intersection point of each central pixel point.

2. The computer-aided plastic surgery assistance navigation method according to claim 1, wherein Obtaining the bone thickness and edge angle of each central pixel point based on the first intersection point and the second intersection point of each central pixel point includes the following specific steps: Denote the distance between the first intersection point and the second intersection point of each central pixel point as the bone thickness of the central pixel point; denote the included angle between the gradient vectors of the first intersection point and the second intersection point of each central pixel point as the edge angle of each central pixel point.

3. The computer-aided plastic surgery assisted navigation method according to claim 1, wherein Obtaining the bone thickness anomaly factor parameter of each suspected pixel point according to the bone thickness and edge angle of the central pixel points on the same thickness line as the suspected pixel point, and the gray value of the suspected pixel point. The corresponding specific formula is as follows: In the formula, represents the bone thickness anomaly factor parameter of the k-th suspected pixel point in the bone image; represents the edge angle of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image; represents the gray value of the k-th suspected pixel point in the bone image; represents the bone thickness of the central pixel point on the same thickness line as the k-th suspected pixel point in the bone image; represents the logarithmic function with the natural constant as the base; is a linear normalization function, is an absolute value function.

4. The computer-aided plastic surgery assisted navigation method according to claim 1, characterized in that, Obtaining all abnormal pixel points according to the bone thickness anomaly factor parameter and gray value of each suspected pixel point, and the distance of each suspected pixel point from the nearest bone edge includes the following specific steps: Obtain the abnormality degree value of each suspected pixel point in the bone image according to the bone thickness anomaly factor parameter of each suspected pixel point in the bone image and the gray value of each suspected pixel point; Among them, the calculation method of the abnormality degree value of the k-th suspected pixel point in the bone image is: In the formula, represents the abnormality degree value of the k-th suspected pixel point in the bone image, represents the bone thickness abnormality factor parameter of the k-th suspected pixel point in the bone image; represents the gray value of the k-th suspected pixel point in the bone image; represents the distance from the k-th suspected pixel point in the bone image to the nearest bone edge; is a linear normalization function, is an absolute value function, is an exponential function with the natural constant as the base; Preset a suspected threshold. When the abnormality degree value of the k-th suspected pixel point is greater than the preset suspected threshold, mark the k-th suspected pixel point as an abnormal pixel point.

5. The computer-aided plastic surgery assistance navigation method according to claim 4, characterized in that, Obtaining all target pixel points according to the gray value difference between each abnormal pixel point and its surrounding pixel points includes the following specific steps: Construct a neighborhood range centered on each abnormal pixel point. According to the abnormality degree value of each abnormal pixel point in the bone image and the difference between the gray value of each abnormal pixel point and the gray values of the surrounding pixel points within the neighborhood range, obtain the target degree value of each abnormal pixel point in the bone image; Among them, the calculation method of the target degree value of the n-th abnormal pixel point is: Wherein, represents the target degree value of the nth abnormal pixel point; represents the gray value of the nth abnormal pixel point; represents the gray value of the mth pixel point within the neighborhood range of the nth abnormal pixel point; represents the distance between the nth abnormal pixel point and the mth pixel point within the neighborhood range; represents the maximum side length of the neighborhood range; is the abnormal degree value of the nth abnormal pixel point in the bone image; is the linear normalization function, is the absolute value function; Preset an abnormal threshold. When the target degree value of the n-th abnormal pixel point is greater than the abnormal threshold, mark the n-th abnormal pixel point as a target pixel point.

6. The computer-aided plastic surgery assisted navigation method according to claim 1, wherein Clustering the target pixel points to obtain a number of target clustering clusters; segmenting the target clustering clusters to obtain a suspected bone bulge abnormal area includes the following specific steps: Taking the gray value and pixel position of the target pixel points as inputs, use the DBSCAN density clustering algorithm for the target pixel points to obtain a number of target clustering clusters, calculate the mean value of the target degree values of all target pixel points in each target clustering cluster, and select the target clustering cluster with the highest mean value of the target degree values as the high-probability bone bulge abnormal area; Use the watershed algorithm to select the suspected bone bulge abnormal area. Select the pixel point with the lowest gray value in the high-probability bone bulge abnormal area as the watershed seed point, and perform watershed segmentation on the bone image according to the selected watershed seed point. The area of the segmentation result is recorded as the suspected bone bulge abnormal area.

7. A computer-aided plastic surgery assisted navigation system, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of the computer-aided plastic surgery assistance navigation method according to any one of claims 1-6.

Images