Orthopedic surgery navigation method, device and system, electronic equipment and storage medium

Abstract

The invention provides an orthopedic surgery navigation method, device and system, electronic equipment and a storage medium, and relates to the technical field of computers.The method comprises the steps that a three-dimensional bone model of a diseased bone is generated based on a first image and a second image, and surgery parameters corresponding to the diseased bone are obtained based on the three-dimensional bone model and all bony mark points on the three-dimensional bone model, on the basis of the corresponding relation between all the bony mark points on the three-dimensional bone model and all the bony mark points on the affected bone, the affected bone and the three-dimensional bone model are registered, the mapping relation between the affected bone and the three-dimensional bone model and / or the first image is established, and surgical navigation is conducted on the basis of the surgical parameters corresponding to the affected bone and the mapping relation. According to the orthopedic surgery navigation method, device and system, the electronic equipment and the storage medium provided by the invention, more objective and quantitative preoperative surgery planning can be realized under the condition that the diseased bone of the patient has bone defect and / or abnormal shape, and more accurate diseased bone space positioning can be realized in the orthopedic surgery.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an orthopedic surgery navigation method, device, system, electronic equipment and storage medium. Background Art

[0002] Surgical navigation is a visual image-guided surgical technology developed with the help of computers, precision instruments and image processing technology, based on medical images such as ultrasound, X-rays, CT (computed tomography) and MRI (magnetic resonance imaging). Surgical navigation can achieve visualization and automation of surgical operations by digitizing the patient's lesion tissue in three dimensions and tracking the position of surgical instruments in real time, thereby assisting doctors or robots to complete surgical tasks more quickly, accurately and safely.

[0003] However, in orthopedic surgery, traditional surgical navigation methods in related technologies are difficult to achieve objective and quantitative preoperative surgical planning when the patient's affected bone has bone defects and / or morphological abnormalities, resulting in unstable surgical results, increased surgical risks, and affecting the patient's postoperative recovery and ultimate efficacy.

[0004] In addition, the traditional surgical navigation methods in the related art are difficult to accurately achieve spatial positioning of the patient's affected bone when there is a bone defect or morphological abnormality in the patient's affected bone, which makes it difficult to achieve accurate intraoperative calibration and guidance based on the traditional surgical navigation methods. Therefore, how to achieve more objective and quantitative preoperative surgical planning and more accurate spatial positioning of the affected bone in orthopedic surgery when there is a bone defect and / or morphological abnormality in the patient's affected bone is a technical problem that needs to be solved in this field. Summary of the invention

[0005] The present invention provides an orthopedic surgery navigation method, device, system, electronic device and storage medium, which are used to solve the defects in the prior art that it is difficult to achieve objective and quantitative preoperative surgical planning when the patient's affected bone has bone defects and / or morphological abnormalities, and it is difficult to achieve accurate spatial positioning of the affected bone during orthopedic surgery. The present invention can achieve more objective and quantitative preoperative surgical planning when the patient's affected bone has bone defects and / or morphological abnormalities, and achieve more accurate spatial positioning of the affected bone during orthopedic surgery.

[0006] The present invention provides an orthopedic surgery navigation method, comprising: Generate a three-dimensional bone model of the affected bone based on a first image and a second image, wherein the first image includes a medical image of the affected bone before orthopedic surgery, and the second image includes a medical image of a healthy bone corresponding to the affected bone, wherein the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline; Determine a plurality of bony landmarks on the three-dimensional bone model, and obtain surgical parameters corresponding to the affected bone based on the three-dimensional bone model and each of the bony landmarks on the three-dimensional bone model, wherein the surgical parameters include at least one of an osteotomy position, an osteotomy size, a prosthesis model, a prosthesis size, and a prosthesis installation position; Based on the correspondence between each of the bony landmarks on the three-dimensional bone model and each of the bony landmarks on the affected bone, the affected bone and the three-dimensional bone model are registered to establish a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image; Surgical navigation is performed based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0007] The present invention also provides an orthopedic surgery navigation device, comprising the following modules: A data acquisition module, configured to generate a three-dimensional bone model of an affected bone based on a first image and a second image, wherein the first image includes a medical image of the affected bone before orthopedic surgery, and the second image includes a medical image of a healthy bone corresponding to the affected bone, wherein the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline; a preoperative planning module, configured to determine a plurality of bony landmarks on the three-dimensional bone model, and to obtain surgical parameters corresponding to the affected bone based on the three-dimensional bone model and the bony landmarks on the three-dimensional bone model, wherein the surgical parameters include at least one of an osteotomy position, an osteotomy size, a prosthesis model, a prosthesis size, and a prosthesis installation position; an intraoperative registration module, configured to register the affected bone with the three-dimensional bone model based on the correspondence between each of the bony landmarks on the three-dimensional bone model and each of the bony landmarks on the affected bone, and establish a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image; An intraoperative execution module is used to perform surgical navigation based on surgical parameters corresponding to the affected bone and a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0008] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the orthopedic surgery navigation method described above is implemented.

[0009] The present invention also provides an orthopedic surgery navigation system, comprising: the electronic device as described above and a surgery navigation device; the electronic device is communicatively connected to the surgery navigation device.

[0010] An orthopedic surgery navigation system provided according to the present invention further includes: a human-computer interaction device; the human-computer interaction device is communicatively connected with the electronic device.

[0011] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the orthopedic surgery navigation method described in any one of the above is implemented.

[0012] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the orthopedic surgery navigation method described above is implemented.

[0013] The orthopedic surgery navigation method, device, system, electronic device and storage medium provided by the present invention generate a three-dimensional bone model of the affected bone by combining the preoperative medical image of the affected bone and the medical image of the healthy bone corresponding to the affected bone, so as to more accurately reflect the healthy morphology of the affected bone when the affected bone has bone defects and / or abnormal morphology, and then obtain the surgical parameters corresponding to the affected bone based on the three-dimensional bone model of the affected bone, so that the preoperative surgical planning is more objective and quantitative. In the process of performing orthopedic surgery on the affected bone, by aligning the bony landmarks on the three-dimensional bone model with the actual bony landmarks on the affected bone, an accurate mapping relationship between the affected bone and the three-dimensional bone model can be established, so as to achieve accurate spatial positioning of the affected bone during the surgery. Based on the surgical parameters corresponding to the affected bone and the mapping relationship with the three-dimensional bone model, the condition of the affected bone can be displayed in real time during the surgery, so as to achieve more accurate spatial positioning of the affected bone during the orthopedic surgery, provide intuitive visual guidance for doctors, and help doctors make more informed decisions during the surgery, thereby reducing surgical complications and postoperative recovery time, significantly improving the accuracy and reliability of orthopedic surgery, effectively reducing surgical risks, and achieving more stable surgical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0015] Figure 1 It is a flow chart of the orthopedic surgery navigation method provided by the present invention.

[0016] Figure 2 It is a structural schematic diagram of the orthopedic surgery navigation device provided by the present invention.

[0017] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention.

[0018] Figure 4 This is one of the structural schematic diagrams of the orthopedic surgery navigation system provided by the present invention.

[0019] Figure 5 This is the second structural schematic diagram of the orthopedic surgery navigation system provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0022] In the description of the present application, the terms "first", "second", etc. are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are usually a class, and the number of objects is not limited. For example, the first object can be one or more. In addition, in the description of the present application, "and / or" represents at least one of the connected objects, and the character " / " generally represents that the front and back associated objects are in an "or" relationship.

[0023] It should be noted that orthopedic surgery is a surgical treatment method for skeletal system diseases, covering a variety of surgical types. The general process of orthopedic surgery usually includes preoperative examination, surgical approach selection, surgical operation and postoperative treatment.

[0024] In the related art, before performing orthopedic surgery, doctors can only plan the surgical plan based on subjective experience according to the CT images of the patient's affected bones taken before the operation. When doctors plan the surgical plan based on subjective experience, they are limited to qualitatively analyzing whether the patient's affected bones are suitable for surgery, and it is difficult to achieve quantitative planning of the surgical plan. For example, it is difficult to quantitatively plan the size of the osteotomy, the size of the implanted prosthesis, and the optimal position of the prosthesis. Accordingly, in the process of orthopedic surgery in the related art, doctors can only rely on subjective experience to complete orthopedic surgery due to the lack of objective and quantitative preoperative surgical planning, which leads to unstable surgical results, increased surgical risks, and affects the patient's postoperative recovery and final efficacy.

[0025] In addition, there are muscles, blood vessels, nerves and other tissue structures around the bones, so it is difficult for doctors to accurately judge the entire shape of the affected bones with the naked eye. When the patient's affected bones have bone defects or abnormal morphology, it is also difficult for doctors to accurately judge the extent of the bone defects or abnormal bone morphology with the naked eye, which may lead to errors in surgical operations and affect the accuracy and safety of the operation.

[0026] For example, shoulder replacement surgery is an orthopedic procedure used to treat severe shoulder pain and dysfunction. Shoulder replacement surgery is often performed on patients who experience persistent pain and limited function due to shoulder arthritis, shoulder fractures, rotator cuff injuries, or other conditions that destroy the shoulder joint.

[0027] Artificial shoulder replacement includes hemi-shoulder replacement, total shoulder replacement and reverse total shoulder replacement. Among them, hemi-shoulder replacement is mainly used for arthritis involving the humeral head and osteonecrosis not involving the labrum, and severe proximal humeral fractures. Total shoulder replacement is mainly used for osteoarthritis, inflammatory arthritis, osteonecrosis involving the labrum and postmenopausal joint degenerative diseases. Reverse total shoulder replacement is mainly used for patients with osteoarthritis and complex humeral fractures. It is different from total shoulder and hemi-shoulder replacement in that the indications are mainly accompanied by severe rotator cuff tear or loss but the biceps function is still complete. Because of its good patient recovery effect and lower revision rate than total shoulder replacement, it is currently the most widely used shoulder replacement. Shoulder replacement surgery removes the damaged joint surface (osteotomy) and uses an artificial prosthesis to implant the remaining joint to replace the damaged joint, thereby reducing pain and restoring the joint's movement function. The general process of shoulder surgery includes preoperative examination (such as medical imaging acquisition), surgical approach, dislocated humeral head, osteotomy and medullary cavity expansion, prosthesis installation, rotator cuff repair and postoperative evaluation. Among them, the steps of osteotomy and prosthesis installation are the most critical. Factors such as the amount of osteotomy, the angle of the osteotomy plane, the location of the prosthesis installation hole, and the fit of the prosthesis implant surface directly affect the success or failure of the operation. If the prosthesis is improperly positioned or the implantation technique is poor, it may cause postoperative pain, joint instability, dislocation or prosthesis loosening. How to complete osteotomy and prosthesis installation with high quality has become the core demand of doctors in clinical practice. If a personalized surgical plan can be made before surgery, a suitable prosthesis can be selected, and joint mobility can be simulated, navigation technology can be used during osteotomy to help doctors complete the preoperative planning plan, and the accuracy of osteotomy and prosthesis placement can be improved, postoperative complications can be reduced, thereby improving the quality of surgery. In addition, preoperative planning can also predict the accuracy and stability of prosthesis placement, adjust the range of motion of the joint, and plan postoperative rehabilitation, thereby improving surgical expectations.

[0028] However, in traditional shoulder replacement surgery, doctors can only roughly plan the surgical plan in a 2D plane based on preoperative medical images and experience. The plan can only be limited to qualitative analysis of whether the patient's joint condition is suitable for surgery, and cannot be quantitatively analyzed, such as the size of the prosthesis, the direction of prosthesis placement, or the amount of osteotomy. Therefore, it is difficult for doctors to formulate personalized surgical plans, and it is even more difficult to simulate the impact of factors that affect the postoperative effect, such as postoperative joint range of motion. During the intraoperative operation, due to the lack of quantitative preoperative planning, doctors can only make judgments based on experience within the current limited time. For example, when encountering a glenoid with bone defects or abnormal morphology, the glenoid morphology may be misjudged during the operation, resulting in incorrect placement of the base; when encountering a patient with a proximal humeral fracture, there is no obvious bone alignment mark, and it is difficult to determine the height level of the humeral stem and the anteroposterior tilt angle during the operation, which affects the doctor's judgment. The lack of precise prosthesis positioning technology may cause the postoperative prosthesis position to deviate from the optimal state, thereby affecting the recovery of joint function.

[0029] Surgical navigation can achieve visualization and automation of surgical operations through three-dimensional digitization of patient lesion tissue, real-time tracking of the position of surgical instruments, and thus assist doctors or robots to complete surgical tasks more quickly, accurately, and safely.

[0030] However, when the affected bone has bone defects or morphological abnormalities, medical images may not be able to accurately reflect the actual condition of the affected bone, which makes it difficult for traditional surgical navigation methods to achieve objective and quantitative preoperative surgical planning when the patient's affected bone has bone defects and / or morphological abnormalities.

[0031] In addition, bone defects and / or morphological abnormalities in the patient's affected bone may cause changes in the bone's landmark structures, making it difficult for traditional surgical navigation methods to accurately achieve spatial positioning of the affected bone when the patient's affected bone has bone defects or morphological abnormalities, which in turn makes it difficult to achieve accurate intraoperative calibration and guidance based on traditional surgical navigation methods.

[0032] For example, during reverse shoulder replacement surgery, screws are required to install the base. However, since the glenoid bone is too thin, if the affected bone cannot be accurately calibrated and guided during the operation, the screws may penetrate the glenoid, resulting in surgical failure.

[0033] In this regard, the present invention provides an orthopedic surgery navigation method. The orthopedic surgery navigation method provided by the present invention can solve the technical problem that the traditional surgical navigation method is difficult to achieve objective and quantitative preoperative surgical planning and accurate spatial positioning of the affected bone when the patient's affected bone has bone defects and / or morphological abnormalities, which leads to the difficulty of accurate intraoperative calibration and guidance based on the traditional surgical navigation method, so as to improve the stability of the surgical effect and reduce the surgical risk and postoperative complications.

[0034] Among them, the preoperative planning part includes completing the reconstruction of the three-dimensional bone model based on the collected preoperative medical images of the patient. The doctor can plan the surgical steps in detail before the operation, including the selection, position and angle adjustment of the scapula and humeral prostheses. At the same time, the joint range of motion can be simulated to predict postoperative joint function, reduce intraoperative decision-making time, and improve the accuracy of the operation.

[0035] The intraoperative registration part includes establishing the correspondence between the patient's affected bone and the three-dimensional bone model through the registration algorithm, achieving accurate matching between the patient's affected bone and the three-dimensional bone model constructed before surgery, and providing accurate reference for intraoperative navigation.

[0036] The intraoperative navigation part tracks the position of surgical tools and displays them in real time to ensure that the doctor has a clear understanding of the progress of the operation. It calculates the distance and angle deviation between the surgical tools and the target position to provide the doctor with accurate navigation information. It provides visual feedback to help the doctor adjust the position and direction of the surgical tools to ensure that the operation is carried out accurately according to the preoperative plan.

[0037] Combine the following Figure 1 The orthopedic surgery navigation method provided by the present invention is described.

[0038] Figure 1 is a flow chart of the orthopedic surgery navigation method provided by the present invention, such as Figure 1As shown, the method includes the following: Step 101, generating a three-dimensional bone model of the affected bone based on a first image and a second image, wherein the first image includes a medical image of the affected bone before orthopedic surgery, and the second image includes a medical image of a healthy bone corresponding to the affected bone, and the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline.

[0039] It should be noted that the embodiment of the present invention is performed by an orthopedic surgery navigation device. The orthopedic surgery navigation device can be configured in an electronic device such as a computer or a server.

[0040] Specifically, the orthopedic surgery navigation method provided by the present invention can provide surgical navigation for doctors performing the orthopedic surgery before and during the orthopedic surgery on the patient's affected bone, thereby improving the accuracy, safety and efficiency of the orthopedic surgery.

[0041] It is understandable that in order to maintain the balance and stability of the human body, the human skeleton is symmetrically distributed with the spine as the midline. Normally, most of the bones in the skull, trunk and limbs appear in pairs, such as the scapula, humerus, ulna, radius, femur, patella, tibia and fibula distributed on the left and right sides of the spine.

[0042] In the embodiment of the present invention, the affected bone and the healthy bone are a pair of human bones symmetrically distributed with the spine as the midline. The affected bone is the bone on the surgical side that needs orthopedic surgery, and the healthy bone is the bone on the healthy side that corresponds to the affected bone and does not need orthopedic surgery. The surgical side is the side of the body that needs orthopedic surgery, and the healthy side is the side of the body that does not need orthopedic surgery.

[0043] In the case where the patient's affected bone has bone defects and / or abnormal morphology, it is difficult to completely and accurately generate a three-dimensional bone model of the affected bone based solely on the image of the affected bone. Therefore, in the embodiment of the present invention, the image of the healthy bone of the patient's healthy side is used for mapping, so that a complete and accurate three-dimensional bone model and bony landmarks of the affected bone can be generated, thereby improving the accuracy of preoperative surgical planning.

[0044] It is understandable that the patient and the affected bone of the patient in the embodiment of the present invention can be determined based on actual needs. The patient and the affected bone of the patient in the embodiment of the present invention are not specifically limited.

[0045] Optionally, the orthopedic surgery performed on the patient's affected bone may be a shoulder replacement surgery. The patient's affected bone may be a scapula, a humerus or a clavicle.

[0046] Before performing orthopedic surgery on the patient's affected bone, a medical image of the patient's affected bone is obtained as a first image, and a medical image of a healthy bone corresponding to the affected bone is obtained as a second image.

[0047] It should be noted that the medical images of the affected bone and the healthy bone in the embodiment of the present invention may include but are not limited to X-ray images, CT images, MRI images and other medical images of the affected bone and the healthy bone.

[0048] It is understandable that medical images such as X-ray images, CT images, and MRI images follow standardized patient positions, fixed imaging equipment, image reconstruction and positioning methods, and standardized reports and markings, so the position of the affected bone in the above medical images is relatively fixed. In the embodiment of the present invention, the horizontal direction in the above medical images can be determined as the X-axis direction, and the vertical direction in the above medical images can be determined as the Y-axis direction.

[0049] In the embodiments of the present invention, medical images of the patient's affected bone and the healthy bone corresponding to the affected bone can be obtained in a variety of ways before orthopedic surgery is performed on the patient's affected bone. For example, the medical images of the patient's affected bone and the healthy bone corresponding to the affected bone can be obtained by data query; or, the medical images of the patient's affected bone and the healthy bone corresponding to the affected bone can be obtained based on user input. The specific method of obtaining the medical images of the patient's affected bone and the healthy bone corresponding to the affected bone is not limited in the embodiments of the present invention.

[0050] After obtaining medical images of the patient's affected bone and the healthy bone corresponding to the affected bone, the medical image of the affected bone can be determined as the first image, and the medical image of the healthy bone corresponding to the affected bone can be determined as the second image. Then, based on the first image and the second image, a three-dimensional bone model of the affected bone can be generated through data calculation, mathematical statistics or deep learning and other technologies.

[0051] As an optional embodiment, generating a three-dimensional bone model of the affected bone based on the first image and the second image includes: constructing a first three-dimensional point cloud model based on the first image, and constructing a second three-dimensional point cloud model based on the second image.

[0052] Specifically, after acquiring the first image and the second image, numerical calculation, mathematical statistics, deep learning and the like can be used to construct a first three-dimensional point cloud model based on the first image and a second three-dimensional point cloud model based on the second image.

[0053] As an optional embodiment, constructing a first three-dimensional point cloud model based on the first image includes: when it is determined that the image quality of the first image meets a preset standard, performing data preprocessing on the first image to obtain the first image after data preprocessing.

[0054] Based on the distribution of bone fragments of the affected bone, the first image after data preprocessing is segmented to obtain a plurality of sub-images corresponding to the first image, each of which includes only one bone fragment of the affected bone.

[0055] The sub-image of the largest bone block of the affected bone is determined as the target sub-image, and then based on the target sub-image, a three-dimensional point cloud model of the largest bone block of the affected bone is constructed as the first three-dimensional point cloud model.

[0056] It should be noted that, since it is difficult to generate a three-dimensional point cloud model based on the first image and the second image when the image quality of the first image and the second image is poor, in the embodiment of the present invention, a preset standard can be set based on prior knowledge and / or actual conditions, and the first image and the second image can be checked based on the preset standard to determine whether the image quality of the first image and the second image can meet the requirements of generating a three-dimensional point cloud model.

[0057] When it is determined that the image quality of the first image and the second image meets the preset standard, it can be determined that the image quality of the first image and the second image can meet the requirements of generating a three-dimensional point cloud model, and then the first image can be preprocessed to further improve the image quality of the first image. The data preprocessing may include image denoising, contrast enhancement, and image sharpening.

[0058] It should be noted that, considering the possibility of bone fracture in the affected bone, after data preprocessing of the first image, the first image may be segmented to obtain a plurality of sub-images corresponding to the first image, wherein each sub-image corresponding to the first image includes only one bone fragment of the affected bone.

[0059] After acquiring the sub-images corresponding to the first image, the sub-image including the largest bone block in the affected bone can be determined as the target sub-image, and then a three-dimensional point cloud model of the largest bone block in the affected bone can be generated based on the target sub-image as the first three-dimensional point cloud model.

[0060] It should be noted that in the embodiment of the present invention, the Marching Cubes (MC) algorithm can be used to generate a three-dimensional point cloud model of the largest bone block in the above-mentioned affected bone based on the above-mentioned target sub-image as the first three-dimensional point cloud model. In the embodiment of the present invention, the Marching Cubes algorithm can also be used to generate a three-dimensional point cloud model of the healthy bone corresponding to the above-mentioned affected bone based on the second image as the second three-dimensional point cloud model.

[0061] Among them, the marching cube algorithm is a very mature algorithm for extracting isosurfaces. The basic idea of ​​the algorithm is to process the cubes (voxels) in the data field one by one, separate the cubes that intersect with the isosurface, and use interpolation to calculate the intersection of the isosurface and the edge of the cube. According to the relative position of each vertex of the cube and the isosurface, the intersection of the isosurface and the edge of the cube is connected in a certain way to generate an isosurface, which is an approximate representation of the isosurface in the cube. This is because the marching cube algorithm has a basic assumption: the data field along the edge of the hexahedron changes continuously. In other words, if the two vertices of an edge are respectively greater than or less than the value of the isosurface, then there is one and only one point on the edge that is the intersection of this edge and the isosurface.

[0062] The embodiment of the present invention can ensure that the clarity and contrast of the image meet the preset standards for subsequent analysis or modeling by preprocessing the first image, which helps to reduce analysis errors caused by poor image quality. By determining the sub-image containing the largest bone block of the affected bone as the target sub-image and constructing a three-dimensional point cloud model based on the sub-image, it can fully take into account the possible bone fracture of the affected bone and provide a more accurate data basis for generating a three-dimensional bone model of the affected bone.

[0063] The centroids of the first three-dimensional point cloud model and the second three-dimensional point cloud model are extracted, and based on the centroids of the first three-dimensional point cloud model and the second three-dimensional point cloud model, the second three-dimensional point cloud model is mirror-reversed along the horizontal direction to obtain a third three-dimensional point cloud model.

[0064] Specifically, in the embodiments of the present invention, Represents the first 3D point cloud model, using Represents a second 3D point cloud model.

[0065] Construct the first 3D point cloud model And the second 3D point cloud model Afterwards, the first 3D point cloud model can be extracted The centroid And the second 3D point cloud model The centroid .

[0066] The position information of the centroid of the 3D point cloud model can be calculated using the following formula: in, Represents the location information of the centroid of the three-dimensional point cloud model; Represents the first The location information of each point; Indicates the number of points in the 3D point cloud model; Represents the first The quality of each point, in the embodiment of the present invention The value of is 1; Represents a positive integer greater than zero.

[0067] It should be noted that in the embodiment of the present invention, the position information of any point in the three-dimensional point cloud model can be represented by a coordinate value, and the position information of the centroid of the three-dimensional point cloud model can also be represented by a coordinate value.

[0068] Get the first 3D point cloud model The centroid And the second 3D point cloud model The centroid Afterwards, based on the second 3D point cloud model The centroid , the second 3D point cloud model The coordinates of each point in the image are flipped horizontally to obtain a third 3D point cloud model. .

[0069] Third 3D point cloud model Middle The location information of each point can be expressed by the following formula: in, Represents the second 3D point cloud model Middle The location information of each point, Represents the second 3D point cloud model The centroid location information; Represents the second 3D point cloud model Middle The position information of the points after mirror flipping in the horizontal direction, that is, the third 3D point cloud model Middle The location information of each point; , Represents the second 3D point cloud model The number of midpoints.

[0070] The first 3D point cloud model and the third 3D point cloud model are aligned based on the centroid of the first 3D point cloud model and the centroid of the third 3D point cloud model.

[0071] Specifically, obtain the third 3D point cloud model Afterwards, you can follow Translate the above third 3D point cloud model , complete the third 3D point cloud model And the first 3D point cloud model alignment.

[0072] A feature descriptor of each point in the third three-dimensional point cloud model is obtained, and then based on the feature descriptor of each point in the third three-dimensional point cloud model, a consistent initial registration algorithm is used to perform point cloud coarse registration on the third three-dimensional point cloud model and the first three-dimensional point cloud model to obtain a fourth three-dimensional point cloud model.

[0073] It should be noted that the SPFH (Simplified Point Feature Histograms) feature vector is a feature representation method used to describe the local geometric relationship of each point in the point cloud. The dimension of the SPFH feature vector is 33. The SPFH feature is generated by calculating the geometric relationship between the point and the points in its neighborhood. The SPFH feature vector consists of three angle histograms: the angle between the normal vector of the point and its neighboring point, the angle between the normal vector of the point and the line connecting the two points, and the angle between the normal vector of the neighboring point and the line connecting the two points.

[0074] FPFH (Fast Point Feature Histograms) feature vector is a feature descriptor widely used in point cloud processing.

[0075] For the third 3D point cloud model The points, the third 3D point cloud model can be obtained by numerical calculation The The SPFH feature vector of points. Among them, Represents a positive integer greater than zero.

[0076] Get the third 3D point cloud model The After the SPFH feature vectors of the points, the third 3D point cloud model can be obtained by weighted accumulation The The SPFH features of the points in the neighborhood of points are used to obtain the third 3D point cloud model. The The FPFH feature vector of each point is calculated as follows: in, Represents the third 3D point cloud model The FPFH feature vector of points; Represents the third 3D point cloud model The SPFH feature vector of points; Represents the third 3D point cloud model Middle Point Neighborhood points; Represents the third 3D point cloud model Middle Point Neighborhood points The SPFH feature vector of Represents the third 3D point cloud model The Point and Neighborhood points The distance between represents a positive integer greater than zero; Represents the third 3D point cloud model Middle The total number of neighboring points of a point.

[0077] Get the third 3D point cloud model The FPFH feature vector of the point Afterwards, the third 3D point cloud model can be The FPFH feature vector of the point Determined as a third 3D point cloud model The The feature descriptor of a point.

[0078] Get the third 3D point cloud model The After the feature descriptors of the points are obtained, the embodiment of the present invention can use the Sample Consensus Initial Alignment (SAC-IA) algorithm to align the third 3D point cloud model. And the first 3D point cloud model The point cloud coarse registration includes the following steps: Select sampling points, Represents a positive integer greater than zero.

[0079] It should be noted that in order to ensure that the third 3D point cloud model The sampling points selected from the third three-dimensional point cloud model have different FPFH features. The distance between any two sampling points selected from the .

[0080] For the third 3D point cloud model The sampling points can be based on the third 3D point cloud model The The feature descriptor of the sampling points , the first 3D point cloud model 3D point cloud model The The sampling points have one or more points with similar FPFH features, which are determined as the third 3D point cloud model. The The similarity points corresponding to the sampling points.

[0081] It should be noted that in the first 3D point cloud model The FPFH feature of any point in the third 3D point cloud model The The feature descriptor of the sampling points When the Euclidean distance between them is less than the preset value, the first three-dimensional point cloud model can be determined. Any of the above points in the third 3D point cloud model The The sampling points have similar FPFH characteristics.

[0082] In the third 3D point cloud model The When the number of similar points corresponding to the sampling points is one, the third 3D point cloud model can be The The similar points corresponding to the sampling points are determined as the third 3D point cloud model The The associated points corresponding to the sampling points; the third 3D point cloud model The When the number of similar points corresponding to the sampling points is multiple, the third 3D point cloud model can be The Any one of the similar points corresponding to the sampling points is determined as the third 3D point cloud model The The associated points corresponding to the sampling points.

[0083] Determine the third 3D point cloud model The After the associated points corresponding to the sampling points, the third 3D point cloud model can be calculated The Sampling points and the third 3D point cloud model The The rigid body transformation matrix between the associated points corresponding to the sampling points can then be used to obtain the third 3D point cloud model by solving the distance error and function. The Sampling points and the third 3D point cloud model The The matching degree between the associated points corresponding to the sampling points. The Sampling points and the third 3D point cloud model The The higher the matching degree between the associated points corresponding to the sampling points, the better the 3D point cloud model The Sampling points and the third 3D point cloud model The The higher the matching degree between the associated points corresponding to the sampling points.

[0084] The distance error and function can be expressed using the Huber penalty function, denoted as ,in: in, represents a predefined threshold; Represents the third 3D point cloud model The Sampling points and the third 3D point cloud model The The distance difference between the diameters of the associated points corresponding to the sampling points.

[0085] Through the third 3D point cloud model And the first 3D point cloud model By coarse registration of the point cloud, we can find a set of optimal transformations that minimize the above Huber penalty function. Among them, the optimal transformation It can be solved by a gradient-based optimization algorithm.

[0086] The third 3D point cloud model According to the optimal transformation By transforming, we can obtain the fourth 3D point cloud model .

[0087] The iterative closest point algorithm and singular value decomposition are used to perform precise point cloud registration on the fourth three-dimensional point cloud model and the first three-dimensional point cloud model to obtain a three-dimensional bone model of the affected bone.

[0088] Specifically, obtain the fourth three-dimensional point cloud model After that, the iterative closest point (ICP) algorithm can be used to calculate the fourth 3D point cloud model. And the first 3D point cloud model Perform precise point cloud registration. The specific steps include: using initial transformation The fourth 3D point cloud model All points are transformed into the first 3D point cloud model Next, define the following energy function: in, represents the energy function value; Represents the fourth 3D point cloud model The After the point is transformed, The nearest point in represents the first transformation parameter; represents the second transformation parameter; Represents the fourth 3D point cloud model The points.

[0089] It should be noted that the optimization target of the energy function in the embodiment of the present invention is the energy function value Minimum.

[0090] Use the Singular Value Decomposition (SVD) algorithm to obtain the first transformation parameters and the second transformation parameter The optimal solution of Among them, the first transformation parameter in the embodiment of the present invention is and the second transformation parameter The optimal solution is The first transformation parameter is the smallest when The value of and the second transformation parameter Based on the first transformation parameter and the second transformation parameter The optimal solution of The specific method is: transform the upper left corner 3x3 matrix to the first transformation parameter The optimal solution of , the 3x1 on the right is the second transformation parameter The optimal solution is to fill the bottom row with 0, 0, 0, 1, and get the transformation .

[0091] The fourth 3D point cloud model According to the optimal transformation By performing the transformation, a three-dimensional bone model of the affected bone can be obtained.

[0092] Second 3D point cloud model To the first 3D point cloud model Transformation .

[0093] In the embodiment of the present invention, after constructing a first three-dimensional point cloud model based on the first image and a second three-dimensional point cloud model based on the second image, the centroid of the first three-dimensional point cloud model and the second three-dimensional point cloud model are extracted, and the second three-dimensional point cloud model is mirror-reversed in the horizontal direction based on the centroid to obtain a third three-dimensional point cloud model, so that symmetric processing of the second three-dimensional point cloud model can be achieved, and the first three-dimensional point cloud model and the third three-dimensional point cloud model are roughly registered by using a consistent initial registration algorithm to obtain a fourth three-dimensional point cloud model, so that the initial corresponding relationship between the first three-dimensional point cloud model and the third three-dimensional point cloud model can be quickly found, which helps to narrow the calculation range of subsequent precise registration and improve the registration efficiency. By iteratively optimizing the transformation relationship between the first three-dimensional point cloud model and the fourth three-dimensional point cloud model, the error between the first three-dimensional point cloud model and the fourth three-dimensional point cloud model can be gradually reduced, and finally accurate registration between the first three-dimensional point cloud model and the second three-dimensional point cloud model is achieved to obtain a three-dimensional bone model of the affected bone, which can provide a more accurate data basis for subsequent surgical plan planning for the affected bone and intraoperative surgical navigation.

[0094] Step 102: determine multiple bony landmarks on the three-dimensional bone model, and obtain surgical parameters corresponding to the affected bone based on the three-dimensional bone model and the bony landmarks on the three-dimensional bone model. The surgical parameters include at least one of osteotomy position, osteotomy size, prosthesis model, prosthesis size, and prosthesis installation position.

[0095] Specifically, after obtaining the three-dimensional bone model of the affected bone, multiple bone landmarks can be determined on the three-dimensional bone model of the affected bone. Among them, bone landmarks refer to bones in certain parts of the human body, which often form obvious protrusions or depressions and are often used for positioning in clinical applications, becoming bone landmarks; in the embodiment of the present invention, the bone landmarks can be determined on the three-dimensional bone model of the affected bone by using an interactive picking method.

[0096] As an optional embodiment, multiple bony landmark points are determined on the three-dimensional bone model, including: determining several bony landmark points at the proximal end of the three-dimensional bone model, and determining several bony landmark points at the distal end of the three-dimensional bone model, the proximal end being the end close to the incision, and the distal end being the end away from the incision.

[0097] It should be noted that the multiple bone landmarks on the three-dimensional bone model of the affected bone in the embodiment of the present invention are evenly distributed at one end of the affected bone close to the incision and the other end away from the incision.

[0098] It should be noted that the incision refers to the skin and tissue structure cut open in orthopedic surgery to expose the surgical site and perform surgical operations. The specific location of the incision can be determined based on the first image.

[0099] Based on the three-dimensional bone model of the affected bone and the bony landmarks on the three-dimensional bone model of the affected bone, the surgical parameters corresponding to the affected bone can be obtained. For example, based on the three-dimensional bone model of the affected bone and the bony landmarks on the three-dimensional bone model of the affected bone, the prosthesis model can be automatically selected, and the initial planning posture of the prosthesis can be set according to the patient's personalized characteristics. Based on the three-dimensional bone model of the affected bone and the bony landmarks on the three-dimensional bone model of the affected bone, the offset between the three-dimensional bone model of the affected bone and the first three-dimensional point cloud model can also be obtained, and the prosthesis planning can be adjusted according to the offset.

[0100] Specifically, based on the three-dimensional bone model of the affected bone and the bony landmarks on the three-dimensional bone model of the affected bone, the surgical parameters corresponding to the affected bone are obtained, and then the specific steps for obtaining the initial prosthesis planning may include: first, according to the coordinates of the bony landmarks, parameter values ​​such as bone surface width, height, bone surface center position, anteversion angle, upward inclination angle, neck-shaft angle, etc. can be calculated; secondly, the bone surface width, height, etc. are used to find the prosthesis with the closest size in the prosthesis library according to the matching principle as the initial prosthesis; thirdly, the prosthesis is placed at the center position of the bone surface calculated above; finally, according to the different requirements of each part, the prosthesis is spatially rotated according to the anteversion angle, upward inclination angle, etc. to obtain the initial planned posture.

[0101] It should be noted that, when the orthopedic surgery on the affected bone is a shoulder replacement surgery, based on the three-dimensional bone model of the affected bone and the bone landmarks on the three-dimensional bone model of the affected bone, planning schemes for three different surgical procedures, namely, normal shoulder, half shoulder, and reverse shoulder, can be obtained. The offset of the surgical side humerus relative to the preoperative side and relative to the contralateral side can also be provided, and the prosthesis planning can be adjusted according to the offset.

[0102] Step 103: Based on the correspondence between the bony landmarks on the three-dimensional bone model and the bony landmarks on the affected bone, the affected bone and the three-dimensional bone model are registered to establish a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0103] Specifically, after obtaining the three-dimensional bone model of the affected bone and determining a plurality of bony landmarks on the three-dimensional bone model of the affected bone, the three-dimensional bone model of the affected bone and the bony landmarks on the three-dimensional bone model may be displayed in the human-computer interaction device.

[0104] During the process of the doctor performing orthopedic surgery on the affected bone, the doctor can use a probe or other equipment to determine the bone landmarks on the affected bone based on the three-dimensional bone model of the affected bone, the bone landmarks of the three-dimensional bone model, and the prompt information displayed in the human-computer interaction device, thereby establishing a correspondence between the bone landmarks on the three-dimensional bone model of the affected bone and the bone landmarks on the affected bone.

[0105] For example, when the bony landmarks include the medial epicondyle and the lateral epicondyle at the distal end of the humerus, the probe can be used to determine the above-mentioned bony landmarks on the affected bone without piercing the patient's epidermis.

[0106] After establishing the correspondence between the bony landmark points on the three-dimensional bone model of the affected bone and the bony landmark points on the affected bone, the affected bone and the three-dimensional bone model of the affected bone can be registered based on the correspondence between the bony landmark points on the three-dimensional bone model of the affected bone and the bony landmark points on the affected bone through numerical calculation, mathematical statistics or deep learning and other technologies to establish an imaging relationship between the affected bone and the three-dimensional bone model of the affected bone and / or the first image.

[0107] As an optional embodiment, based on the correspondence between each bony landmark point on the three-dimensional bone model and each bony landmark point on the affected bone, the affected bone and the three-dimensional bone model are registered, and a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image is established, including: establishing a first registration coordinate system corresponding to the affected bone based on the positional relationship between each bony landmark point on the affected bone, and establishing a second registration coordinate system corresponding to the three-dimensional bone model based on the positional relationship between each bony landmark point on the three-dimensional bone model.

[0108] It should be noted that in the related art, when registering two three-dimensional models, a point pair registration algorithm is used, that is, based on 4-6 points with corresponding relationships on the two three-dimensional models, a point pair registration algorithm is used to calculate the registration matrix. However, for the registration of the three-dimensional bone model of the affected bone and the affected bone, since the determination of the bony landmarks on the affected bone is easily affected by factors such as the patient's soft tissue and cartilage, the bony landmarks determined on the affected bone are not very accurate. At the same time, due to the small surgical area that can be operated on the affected bone, the error in registering the three-dimensional bone model of the affected bone and the affected bone based on the above registration algorithm is relatively large.

[0109] In this regard, an embodiment of the present invention provides an improved registration method. The improved registration method in the embodiment of the present invention performs two registrations on the affected bone and the three-dimensional bone model of the affected bone by establishing a registration coordinate system.

[0110] Specifically, in an embodiment of the present invention, a first registration coordinate system corresponding to the affected bone can be established by mathematical methods based on the positional relationship between the bony landmark points on the affected bone, and a second registration coordinate system corresponding to the three-dimensional bone model can be established by mathematical methods based on the positional relationship between the bony landmark points on the three-dimensional bone model.

[0111] It can be understood that the method of establishing the first registration coordinate system corresponding to the affected bone and the method of establishing the second registration coordinate system corresponding to the three-dimensional bone model in the embodiment of the present invention are the same.

[0112] As an optional embodiment, the bony landmark points on the three-dimensional bone model include two bony landmark points located at the inner and outer sides of the proximal end of the three-dimensional bone model and two bony landmark points located at the inner and outer sides of the distal end of the three-dimensional bone model; the bony landmark points on the affected bone include two bony landmark points located at the inner and outer sides of the proximal end of the affected bone and two bony landmark points located at the inner and outer sides of the distal end of the affected bone.

[0113] Based on the positional relationship between the bony landmarks on the affected bone, a first registration coordinate system corresponding to the affected bone is established, and based on the positional relationship between the bony landmarks on the three-dimensional bone model, a second registration coordinate system corresponding to the three-dimensional bone model is established, including: taking the direction from the bony landmark located on the inner side of the proximal end of the affected bone to the bony landmark located on the outer side of the proximal end of the affected bone as the X-axis direction of the first registration coordinate system, taking the midpoint of the line connecting the two bony landmarks located on the inner and outer sides of the distal end of the affected bone as the origin of the first registration coordinate system, and taking the upward direction of the plane passing through the origin of the first registration coordinate system and perpendicular to the origin of the first registration coordinate system and the two bony landmarks located on the inner and outer sides of the proximal end of the affected bone as the Y-axis of the first registration coordinate system to obtain the first registration coordinate system.

[0114] The direction from the bony landmark located on the inner side of the proximal end of the three-dimensional bone model to the bony landmark located on the outer side of the proximal end of the three-dimensional bone model is taken as the X-axis direction of the second registration coordinate system, the midpoint of the line connecting the two bony landmarks located on the inner and outer sides of the distal end of the three-dimensional bone model is taken as the origin of the second registration coordinate system, and the upward direction of the plane passing through the origin of the second registration coordinate system and perpendicular to the origin of the second registration coordinate system and the two bony landmarks located on the inner and outer sides of the proximal end of the three-dimensional bone model is taken as the Y-axis of the second registration coordinate system to obtain the second registration coordinate system.

[0115] Specifically, taking the humerus as an example, the bony landmarks on the affected bone include bony landmark 1 and bony landmark 2 located on both sides of the proximal end of the humerus, bony landmark A located on the medial epicondyle of the humerus, and bony landmark B located on the lateral epicondyle of the humerus.

[0116] The bony landmarks on the three-dimensional bone model of the affected bone include bony landmark 1 and bony landmark 2 located on both sides of the proximal end of the three-dimensional bone model of the humerus, as well as bony landmark A located at the medial epicondyle of the three-dimensional bone model of the humerus and bony landmark B located at the lateral epicondyle of the three-dimensional bone model of the humerus.

[0117] The vector from the bony landmark point A located on the medial epicondyle of the humerus to the bony landmark point B located on the lateral epicondyle of the humerus The X-axis of the first registration coordinate system is taken as the midpoint of the line connecting the bony landmark point 1 and the bony landmark point 2 located on both sides of the proximal end of the humerus as the origin O of the first registration coordinate system. The plane AOB is determined by the origin O of the first registration coordinate system and the bony landmark point A located at the medial epicondyle of the humerus pointing to the bony landmark point B located at the lateral epicondyle of the humerus. The direction passing through the origin O of the first registration coordinate system and perpendicular to the upward direction of the plane AOB is determined as the Y-axis of the first registration coordinate system, and the first registration coordinate system can be obtained.

[0118] Similarly, the vector from the bony landmark point A located at the medial epicondyle of the 3D bone model of the humerus to the bony landmark point B located at the lateral epicondyle of the humerus is The X-axis of the second registration coordinate system is taken as the midpoint of the line connecting the bony landmark point 1 and the bony landmark point 2 located on both sides of the proximal end of the three-dimensional bone model of the humerus as the origin O of the second registration coordinate system. The plane AOB is determined by the origin O of the second registration coordinate system and the bony landmark point A located at the medial epicondyle of the three-dimensional bone model of the humerus pointing to the bony landmark point B located at the lateral epicondyle of the humerus. The direction passing through the origin O of the second registration coordinate system and perpendicular to the upward direction of the plane AOB is determined as the Y-axis of the second registration coordinate system, and the second registration coordinate system can be obtained.

[0119] A representation matrix of the first registration coordinate system and a representation matrix of the second registration coordinate system are established, and a transformation matrix between the representation matrix of the first registration coordinate system and the representation matrix of the second registration coordinate system is calculated as a target change matrix.

[0120] Specifically, the representation matrix of the first registration coordinate system is established by mathematical methods: and the representation matrix of the second registration coordinate system Afterwards, the representation matrix of the first registration coordinate system can be calculated and the representation matrix of the second registration coordinate system The change matrix between , the specific calculation formula is as follows: Based on the target change matrix, each bony landmark point on the affected bone is spatially transformed to obtain a first spatial point set, and based on each bony landmark point on the affected bone, a normal distribution transformation algorithm is used to calculate a first transformation matrix.

[0121] It should be noted that the representation matrix of the first registration coordinate system is The 3x3 matrix in the upper left corner is the direction vector of the X-axis, Y-axis and Z-axis in the first registration coordinate system. The 3x1 matrix to the right of the above 3x3 matrix is ​​the origin coordinate of the first registration coordinate system. The 1x3 below the above 3x3 matrix and 3x1 matrix is ​​{0, 0, 0, 1}.

[0122] The representation matrix of the second registration coordinate system The 3x3 matrix in the upper left corner is the direction vector of the X-axis, Y-axis and Z-axis in the second registration coordinate system. The 3x1 matrix to the right of the above 3x3 matrix is ​​the origin coordinate of the second registration coordinate system. The 1x3 below the above 3x3 matrix and 3x1 matrix is ​​{0, 0, 0, 1}. Specifically, in the embodiments of the present invention, Represents the bony landmark point cloud on the affected bone, using A point cloud of bony landmarks on a 3D bone model representing the affected bone.

[0123] Based on the target change matrix Point cloud of bony landmarks on the affected bone By performing spatial transformation, the bony landmark point cloud on the affected bone can be Transform to the second registration coordinate system to obtain the first spatial point set .

[0124] Bone landmark point cloud based on the 3D bone model of the affected bone , the first change matrix can be calculated using the Normal Distribution Transform (NDT) algorithm The specific calculation steps include: Divide into multiple grids and calculate the mean in each grid. The calculation formula is as follows: in, represents the mean value in the grid; Indicates the number of bony landmarks in the mesh; Indicates the first The location information of the bony landmarks.

[0125] Calculate the covariance of each grid using the following formula: in, Indicates transpose.

[0126] Constructing a normal distribution , its probability distribution density It can be expressed as: Use Initial Transformation Transform the first space point set For all points in , establish the following evaluation function : in, Represents the first spatial point set Middle points; express According to the initial transformation The bony landmark point cloud mapped onto the 3D bone model after transformation The mean value of the grid after ; Represents the first spatial point set Middle Points According to the initial transformation The transformed value.

[0127] According to Newton optimization method, the above evaluation function is optimized, that is, to find the evaluation function The optimal transformation parameter ,The key step of optimization is to solve the Jacobian matrix and the Hessian matrix.

[0128] Repeat the above steps to optimize the evaluation function until the convergence condition is met, and obtain the first change matrix .

[0129] The first spatial point set is spatially transformed based on the first transformation matrix to obtain a second spatial point set.

[0130] Specifically, obtain the first change matrix Afterwards, based on the first change matrix For the first set of spatial points By performing spatial transformation, we can obtain the second spatial point set .

[0131] The iterative closest point algorithm and singular value decomposition are used to align the second spatial point set and the bony landmarks on the three-dimensional bone model to obtain a first alignment matrix for describing the mapping relationship between the affected bone and the three-dimensional bone model.

[0132] Specifically, the second spatial point set is obtained Afterwards, the iterative closest point algorithm can be used to calculate the second spatial point set. Bone landmark point cloud on the 3D bone model of the affected bone Perform precise point cloud registration. The specific steps include: The second space point set All points in the image are transformed into the bone landmark point cloud on the 3D bone model of the affected bone. Next, define the following energy function: in, represents the energy function value; Represents the second space point set Middle points to bony landmark point cloud the closest point to the middle; represents the first transformation parameter; represents the second transformation parameter; Represents the second space point set The points.

[0133] It should be noted that the optimization target of the energy function in the embodiment of the present invention is the energy function value Minimum.

[0134] Use the Singular Value Decomposition (SVD) algorithm to obtain the first transformation parameters and the second transformation parameter The optimal solution of Among them, the first transformation parameter in the embodiment of the present invention is and the second transformation parameter The optimal solution is The first transformation parameter is the smallest when The value of and the second transformation parameter Based on the first transformation parameter and the second transformation parameter The optimal solution of The specific method is: transform the upper left corner 3x3 matrix to the first transformation parameter The optimal solution of , the 3x1 on the right is the second transformation parameter The optimal solution is to fill the bottom row with 0, 0, 0, 1 to get the second change matrix .

[0135] The first registration matrix used to describe the mapping relationship between the affected bone and the three-dimensional bone model .

[0136] Based on the mapping relationship between the three-dimensional bone model and the first image and the first registration matrix, a second registration matrix for describing the mapping relationship between the affected bone and the first image is obtained.

[0137] Specifically, obtain the first registration matrix Then, based on the mapping relationship between the three-dimensional bone model of the affected bone and the first image and the first registration matrix , through numerical calculation, a second registration matrix for describing the mapping relationship between the affected bone and the first image is obtained.

[0138] In an embodiment of the present invention, a first registration coordinate system corresponding to the affected bone is established based on the positional relationship between the bony landmarks on the affected bone, and a second registration coordinate system corresponding to the three-dimensional bone model is established based on the positional relationship between the bony landmarks on the three-dimensional bone model. Then, a representation matrix of the first registration coordinate system and a representation matrix of the second registration coordinate system are established, and a transformation matrix between the representation matrix of the first registration coordinate system and the representation matrix of the second registration coordinate system is calculated as a target change matrix. Based on the target change matrix, a spatial transformation is performed on the bony landmarks on the affected bone to obtain a first spatial point set. Based on the bony landmarks on the affected bone, a first transformation matrix is ​​calculated using a normal distribution transformation algorithm, and the first spatial point set is spatially transformed based on the first transformation matrix. The first registration matrix for describing the mapping relationship between the affected bone and the three-dimensional bone model is obtained by transforming between the two points, and the second spatial point set is obtained. The iterative closest point algorithm and singular value decomposition are used to align the second spatial point set and the bony landmarks on the three-dimensional bone model, and the first registration matrix for describing the mapping relationship between the affected bone and the three-dimensional bone model is obtained. The registration coordinate system corresponding to the affected bone and the three-dimensional bone model is established, and the transformation matrix between them is calculated, so as to more accurately describe the spatial relationship between the affected bone and the three-dimensional bone model, overcome the limitations of the traditional point-based registration algorithm, reduce the registration error caused by inaccurate bony landmarks on the affected bone or limited surgical area, and combine the normal distribution transformation algorithm and the iterative closest point algorithm for spatial transformation and registration, which significantly improves the registration accuracy and robustness between the affected bone and the three-dimensional bone model.

[0139] Step 104: Perform surgical navigation based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0140] Specifically, after obtaining the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the image of the affected bone, the orthopedic surgical operation on the affected bone can be completed using a surgical navigation device based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the image of the affected bone.

[0141] When the doctor holds the surgical navigation device to perform orthopedic surgery on the affected bone, the human-computer interaction device can display the real-time image of the affected bone during the operation and the spatial position and posture of the surgical navigation device. It can also calculate the distance, angle and other deviations between the surgical navigation device and any position on the affected bone to provide visual feedback.

[0142] When grinding of the affected bone is required, in order to enhance the visual feedback during the grinding process, the 3D bone model of the affected bone can be colored according to the corresponding surgical parameters of the affected bone, and red, white, and green are used to represent over-grinded, properly ground, and under-ground areas. During the grinding of the affected bone, the coloring information in the 3D bone model of the affected bone can be updated in real time based on the actual grinding condition of the affected bone and the position of the grinding tool to reflect the grinding condition of the affected bone in real time and ensure that the grinding area is completed.

[0143] When it is necessary to grind the center nail, fixed nail or non-fixed nail implanted in the affected bone, the center nail, fixed nail and non-fixed nail can be tracked and displayed in real time, and the three-dimensional bone model distance of the center nail penetrating the affected bone can be calculated to assist in adjusting the direction of the center nail to ensure that the center nail does not pass through the bone surface of the affected bone. At the same time, the direction of the fixed nail and non-fixed nail can be adjusted according to the patient's bone physiological structure to achieve the most stable base fixation effect.

[0144] When a base needs to be installed for the affected bone, the real-time position and posture of the grinding tool and the installation tool can be obtained through the tracer on the affected bone side and the tracer on the surgical navigation device to calculate the distance angle deviation of the grinding reamer and the installation tool and the depth from the target position.

[0145] When the affected bone needs to be reconstructed, different types of osteotomy guides can be adapted, including standard guides, personalized guides, etc., and during the osteotomy process, the humeral tracer and the tracer installed on the guide can be used to calculate and display in real time the error between the guide plane and the planned osteotomy plane.

[0146] During the installation of the affected bone prosthesis, the position of the affected bone and the prosthesis can be obtained in real time through the tracers installed on the prosthesis and the affected bone, and compared with the planned position, and the deviation between the axis of the affected bone and the prosthesis and the anteversion angle and the plan can be provided in real time.

[0147] After the prosthesis is installed, in reverse shoulder replacement, a probe can be used to click on the liner, and the installation position and posture of the humeral stem can be used to further confirm whether the liner is installed in place. In normal shoulder and hemi-shoulder replacement, the tracer on the ball head installer can be used to determine whether the ball head on the humeral stem is installed in place.

[0148] The embodiment of the present invention generates a three-dimensional bone model of the affected bone by combining the preoperative medical image of the affected bone and the medical image of the healthy bone corresponding to the affected bone. When the affected bone has bone defects and / or abnormal morphology, the healthy morphology of the affected bone can be more accurately reflected, and the surgical parameters corresponding to the affected bone can be obtained based on the three-dimensional bone model of the affected bone, so that the preoperative surgical planning is more objective and quantitative. In the process of orthopedic surgery on the affected bone, by aligning the bony landmarks on the three-dimensional bone model with the actual bony landmarks on the affected bone, an accurate mapping relationship between the affected bone and the three-dimensional bone model can be established, and accurate spatial positioning of the affected bone during surgery can be achieved. Based on the surgical parameters corresponding to the affected bone and the mapping relationship with the three-dimensional bone model, the condition of the affected bone can be displayed in real time during the surgery, and more accurate spatial positioning of the affected bone can be achieved during orthopedic surgery. Intuitive visual guidance can be provided for doctors, which helps doctors make more informed decisions during surgery, thereby reducing surgical complications and postoperative recovery time, significantly improving the accuracy and reliability of orthopedic surgery, effectively reducing surgical risks, and achieving more stable surgical results.

[0149] Figure 2 This is a schematic diagram of the structure of the orthopedic surgery navigation device provided by the present invention. Figure 2 The orthopedic surgery navigation device provided by the present invention is described. The orthopedic surgery navigation device described below and the orthopedic surgery navigation method provided by the present invention described above can be referred to each other. Figure 2 As shown, the device includes: a data acquisition module 201, a preoperative planning module 202, an intraoperative registration module 203 and an intraoperative execution module 204.

[0150] The data acquisition module 201 is used to generate a three-dimensional bone model of the affected bone based on a first image and a second image, wherein the first image includes a medical image of the affected bone before orthopedic surgery, and the second image includes a medical image of a healthy bone corresponding to the affected bone, wherein the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline.

[0151] The preoperative planning module 202 is used to determine multiple bony landmarks on the three-dimensional bone model, and obtain surgical parameters corresponding to the affected bone based on the three-dimensional bone model and the bony landmarks on the three-dimensional bone model. The surgical parameters include at least one of osteotomy position, osteotomy size, prosthesis model, prosthesis size and prosthesis installation position.

[0152] The intraoperative registration module 203 is used to register the affected bone with the three-dimensional bone model based on the correspondence between each bony landmark point on the three-dimensional bone model and each bony landmark point on the affected bone, and establish a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0153] The intraoperative execution module 204 is used to perform surgical navigation based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0154] Specifically, the data acquisition module 201, the preoperative planning module 202, the intraoperative registration module 203 and the intraoperative execution module 204 are electrically connected.

[0155] The orthopedic surgery navigation device in the embodiment of the present invention generates a three-dimensional bone model of the affected bone by combining the preoperative medical image of the affected bone and the medical image of the healthy bone corresponding to the affected bone. When the affected bone has bone defects and / or abnormal morphology, the healthy morphology of the affected bone can be more accurately reflected, and the surgical parameters corresponding to the affected bone can be obtained based on the three-dimensional bone model of the affected bone, so that the preoperative surgical planning is more objective and quantitative. During the orthopedic surgery on the affected bone, by registering the bony landmarks on the three-dimensional bone model with the actual bony landmarks on the affected bone, an accurate mapping relationship between the affected bone and the three-dimensional bone model can be established, and accurate spatial positioning of the affected bone during the surgery can be achieved. Based on the surgical parameters corresponding to the affected bone and the mapping relationship with the three-dimensional bone model, the condition of the affected bone can be displayed in real time during the surgery, and more accurate spatial positioning of the affected bone can be achieved during the orthopedic surgery. It can provide intuitive visual guidance for doctors, help doctors make more informed decisions during the surgery, thereby reducing surgical complications and postoperative recovery time, significantly improving the accuracy and reliability of orthopedic surgery, effectively reducing surgical risks, and achieving more stable surgical results.

[0156] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3As shown, the electronic device 1100 may include: a processor 310 , a communications interface 320 , a memory 330 and a communication bus 340 , wherein the processor 310 , the communications interface 320 , and the memory 330 communicate with each other via the communication bus 340 . The processor 310 can call the logic instructions in the memory 330 to execute the orthopedic surgery navigation method, which includes: generating a three-dimensional bone model of the affected bone based on the first image and the second image, the first image includes a medical image of the affected bone before the orthopedic surgery, and the second image includes a medical image of the healthy bone corresponding to the affected bone, the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline; determining multiple bony landmarks on the three-dimensional bone model, and obtaining surgical parameters corresponding to the affected bone based on the three-dimensional bone model and each bony landmark on the three-dimensional bone model, the surgical parameters including at least one of osteotomy position, osteotomy size, prosthesis model, prosthesis size and prosthesis installation position; based on the correspondence between each bony landmark on the three-dimensional bone model and each bony landmark on the affected bone, aligning the affected bone and the three-dimensional bone model, and establishing a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image; performing surgical navigation based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0157] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0158] Figure 4 This is one of the structural schematic diagrams of the orthopedic surgery navigation system provided by the present invention. Figure 5 This is the second structural diagram of the orthopedic surgery navigation system provided by the present invention. Figure 4 and Figure 5 As shown, the orthopedic surgery navigation system 1000 includes: the electronic device 1100 as described above and a surgery navigation device 2000; the electronic device is electrically connected to the surgery navigation device.

[0159] As an optional embodiment, the orthopedic surgery navigation system also includes a human-computer interaction device 1200 .

[0160] The human-computer interaction device 1200 is in communication connection with the electronic device 1100 .

[0161] The surgical navigation device 2000 includes a navigation camera 2100 , a patient tracer 2200 , a probe 2300 , and a navigation tool 2400 .

[0162] Specifically, the electronic device can communicate with the surgical navigation device to obtain the actual spatial position of the affected bone and the surgical tools.

[0163] The electronic device 1100 is respectively connected to the human-machine interaction device 1200 and the navigation camera 2100 for communication, receives information transmitted by the human-machine interaction device 1200 and the navigation camera 2100 , and sends relevant information or instructions to the human-machine interaction device 1200 and the navigation camera 2100 .

[0164] The navigation camera 2100 receives signals from the patient tracer 2200, the probe 2300, and the navigation tool 2400, and determines the relative spatial position relationship between the surgical navigation device and the affected bone in the same spatial coordinates. On the premise that the spatial position relationship between the affected bone and the surgical navigation device is determined, the navigation camera 2100 receives the signal from the probe 2300 and completes the collection of the bony landmarks of the affected bone.

[0165] The orthopedic surgery navigation system provided by the present invention includes an electronic device for executing the orthopedic surgery navigation method provided by the present invention, a surgical navigation device and a human-computer interaction device, which can improve the accuracy, stability and safety of the surgery, while reducing the operation time, reducing the surgical risk, and improving the surgical efficiency. It can complete the personalized surgical plan of the affected bone before the operation based on the preoperative medical image of the affected bone and the medical image of the healthy bone corresponding to the affected bone, and can use the healthy side data to map and generate the affected side data to deal with cases of fractures or severe bone defects. At the same time, it can provide the offset of the humerus on the surgical side relative to the preoperative side and the opposite side, adjust the prosthesis planning according to the offset, use an infrared camera for intraoperative navigation, and provide high-precision real-time navigation while reducing the impact of the environment on the navigation camera. Real-time tool tracking and visual feedback can effectively improve the accuracy and efficiency of prosthesis implantation.

[0166] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the orthopedic surgery navigation method provided by the above methods, the method comprising: generating a three-dimensional bone model of the affected bone based on a first image and a second image, the first image including a medical image of the affected bone before orthopedic surgery, the second image including a medical image of a healthy bone corresponding to the affected bone, the affected bone and the healthy bone corresponding to the affected bone being a pair of human bones symmetrically distributed with the spine as the midline; in the three-dimensional A plurality of bony landmarks are determined on the bone model, and surgical parameters corresponding to the affected bone are obtained based on the three-dimensional bone model and the bony landmarks on the three-dimensional bone model, the surgical parameters including at least one of osteotomy position, osteotomy size, prosthesis model, prosthesis size and prosthesis installation position; based on the correspondence between the bony landmarks on the three-dimensional bone model and the bony landmarks on the affected bone, the affected bone and the three-dimensional bone model are registered, and a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image is established; surgical navigation is performed based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0167] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented by a processor to execute the orthopedic surgery navigation method provided by the above methods when executed, the method comprising: generating a three-dimensional bone model of an affected bone based on a first image and a second image, the first image comprising a medical image of the affected bone before orthopedic surgery, the second image comprising a medical image of a healthy bone corresponding to the affected bone, the affected bone and the healthy bone corresponding to the affected bone being a pair of human bones symmetrically distributed with the spine as the midline; determining a plurality of bony landmarks on the three-dimensional bone model, and obtaining surgical parameters corresponding to the affected bone based on the three-dimensional bone model and the bony landmarks on the three-dimensional bone model, the surgical parameters comprising at least one of an osteotomy position, an osteotomy size, a prosthesis model, a prosthesis size, and a prosthesis installation position; registering the affected bone and the three-dimensional bone model based on the correspondence between the bony landmarks on the three-dimensional bone model and the bony landmarks on the affected bone, and establishing a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image; performing surgical navigation based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

[0168] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0169] Through the description of the above implementation methods, technicians in this field can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0170] Finally, 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 aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An orthopedic surgery navigation method, characterized in that: include: Generate a three-dimensional bone model of the affected bone based on a first image and a second image, wherein the first image includes a medical image of the affected bone before orthopedic surgery, and the second image includes a medical image of a healthy bone corresponding to the affected bone, wherein the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline; Determine a plurality of bony landmarks on the three-dimensional bone model, and obtain surgical parameters corresponding to the affected bone based on the three-dimensional bone model and each of the bony landmarks on the three-dimensional bone model, wherein the surgical parameters include at least one of an osteotomy position, an osteotomy size, a prosthesis model, a prosthesis size, and a prosthesis installation position; Based on the correspondence between each of the bony landmarks on the three-dimensional bone model and each of the bony landmarks on the affected bone, the affected bone and the three-dimensional bone model are registered to establish a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image; Surgical navigation is performed based on the surgical parameters corresponding to the affected bone and the mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

2. The orthopedic surgery navigation method according to claim 1, characterized in that: The method of generating a three-dimensional bone model of the affected bone based on the first image and the second image includes: Based on the first image, construct a first three-dimensional point cloud model, and based on the second image, construct a second three-dimensional point cloud model; Extracting the centroids of the first three-dimensional point cloud model and the second three-dimensional point cloud model, and based on the centroids of the first three-dimensional point cloud model and the second three-dimensional point cloud model, performing a mirror inversion on the second three-dimensional point cloud model in a horizontal direction to obtain a third three-dimensional point cloud model; Aligning the first three-dimensional point cloud model and the third three-dimensional point cloud model based on the centroid of the first three-dimensional point cloud model and the centroid of the third three-dimensional point cloud model; Acquire a feature descriptor of each point in the third three-dimensional point cloud model, and then based on the feature descriptor of each point in the third three-dimensional point cloud model, use a consistent initial registration algorithm to perform point cloud coarse registration on the third three-dimensional point cloud model and the first three-dimensional point cloud model to obtain a fourth three-dimensional point cloud model; The fourth three-dimensional point cloud model and the first three-dimensional point cloud model are precisely aligned with each other using an iterative closest point algorithm and a singular value decomposition to obtain a three-dimensional bone model of the affected bone.

3. The orthopedic surgery navigation method according to claim 1, characterized in that: The registering the affected bone with the three-dimensional bone model based on the correspondence between each of the bony landmarks on the three-dimensional bone model and each of the bony landmarks on the affected bone, and establishing a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image, comprises: Based on the positional relationship between the bony landmarks on the affected bone, a first registration coordinate system corresponding to the affected bone is established; based on the positional relationship between the bony landmarks on the three-dimensional bone model, a second registration coordinate system corresponding to the three-dimensional bone model is established; Establishing a representation matrix of the first registration coordinate system and a representation matrix of the second registration coordinate system, and calculating a transformation matrix between the representation matrix of the first registration coordinate system and the representation matrix of the second registration coordinate system as a target change matrix; Performing spatial transformation on each of the bony landmarks on the affected bone based on the target change matrix to obtain a first spatial point set, and calculating a first transformation matrix based on each of the bony landmarks on the affected bone using a normal distribution transformation algorithm; Performing spatial transformation on the first spatial point set based on the first transformation matrix to obtain a second spatial point set; Using an iterative closest point algorithm and singular value decomposition, the second spatial point set and each of the bony landmarks on the three-dimensional bone model are registered to obtain a first registration matrix for describing a mapping relationship between the affected bone and the three-dimensional bone model; Based on the mapping relationship between the three-dimensional bone model and the first image and the first registration matrix, a second registration matrix for describing the mapping relationship between the affected bone and the first image is acquired.

4. The orthopedic surgery navigation method according to claim 3, characterized in that: Each of the bony landmarks on the affected bone includes two bony landmarks located at the inner side and the outer side of the proximal end of the affected bone, and two bony landmarks located at the inner side and the outer side of the distal end of the affected bone; The step of establishing a first registration coordinate system corresponding to the affected bone based on the positional relationship between the bony landmarks on the affected bone comprises: The first registration coordinate system is obtained by taking the direction from the bony landmark located on the inner side of the proximal end of the affected bone to the bony landmark located on the outer side of the proximal end of the affected bone as the X-axis direction of the first registration coordinate system, taking the midpoint of the line connecting the two bony landmarks located on the inner side and the outer side of the distal end of the affected bone as the origin of the first registration coordinate system, and taking the upward direction of the plane passing through the origin of the first registration coordinate system and perpendicular to the origin of the first registration coordinate system and the two bony landmarks located on the inner side and the outer side of the proximal end of the affected bone as the Y-axis of the first registration coordinate system; Each of the bony landmarks on the three-dimensional bone model includes two bony landmarks located at the inner side and the outer side of the proximal end of the three-dimensional bone model, and two bony landmarks located at the inner side and the outer side of the distal end of the three-dimensional bone model; Based on the positional relationship between the bony landmarks on the three-dimensional bone model, a second registration coordinate system corresponding to the three-dimensional bone model is established, including: The second registration coordinate system is obtained by taking the direction from the bony landmark located on the inner side of the proximal end of the three-dimensional bone model to the bony landmark located on the outer side of the proximal end of the three-dimensional bone model as the X-axis direction of the second registration coordinate system, taking the midpoint of the line connecting the two bony landmarks located on the inner side and the outer side of the distal end of the three-dimensional bone model as the origin of the second registration coordinate system, and taking the upward direction of the plane passing through the origin of the second registration coordinate system and perpendicular to the origin of the second registration coordinate system and the two bony landmarks located on the inner side and the outer side of the proximal end of the three-dimensional bone model as the Y-axis of the second registration coordinate system.

5. The orthopedic surgery navigation method according to claim 2, characterized in that: The step of constructing a first three-dimensional point cloud model based on the first image includes: When it is determined that the image quality of the first image meets a preset standard, performing data preprocessing on the first image to obtain a first image after data preprocessing; Based on the distribution of the bone fragments of the affected bone, the first image after data preprocessing is subjected to image segmentation to obtain a plurality of sub-images corresponding to the first image, each of the sub-images including only one bone fragment of the affected bone; The sub-image of the largest bone block of the affected bone is determined as the target sub-image, and then based on the target sub-image, a three-dimensional point cloud model of the largest bone block of the affected bone is constructed as the first three-dimensional point cloud model.

6. The orthopedic surgery navigation method according to any one of claims 1 to 5, characterized in that: Determining a plurality of bony landmarks on the three-dimensional bone model includes: A plurality of bony landmark points are determined at the proximal end of the three-dimensional bone model, and a plurality of bony landmark points are determined at the distal end of the three-dimensional bone model, wherein the proximal end is the end close to the incision, and the distal end is the end away from the incision.

7. An orthopedic surgery navigation device, characterized in that: include: A data acquisition module, configured to generate a three-dimensional bone model of an affected bone based on a first image and a second image, wherein the first image includes a medical image of the affected bone before orthopedic surgery, and the second image includes a medical image of a healthy bone corresponding to the affected bone, wherein the affected bone and the healthy bone corresponding to the affected bone are a pair of human bones symmetrically distributed with the spine as the midline; a preoperative planning module, configured to determine a plurality of bony landmarks on the three-dimensional bone model, and to obtain surgical parameters corresponding to the affected bone based on the three-dimensional bone model and the bony landmarks on the three-dimensional bone model, wherein the surgical parameters include at least one of an osteotomy position, an osteotomy size, a prosthesis model, a prosthesis size, and a prosthesis installation position; an intraoperative registration module, configured to register the affected bone with the three-dimensional bone model based on the correspondence between each of the bony landmarks on the three-dimensional bone model and each of the bony landmarks on the affected bone, and establish a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image; An intraoperative execution module is used to perform surgical navigation based on surgical parameters corresponding to the affected bone and a mapping relationship between the affected bone and the three-dimensional bone model and / or the first image.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the orthopedic surgery navigation method according to any one of claims 1 to 6 is implemented.

9. An orthopedic surgery navigation system, characterized in that: include: The electronic device and surgical navigation device as claimed in claim 8; The electronic device is communicatively connected to the surgical navigation device.

10. The orthopedic surgery navigation system according to claim 9, characterized in that: Also includes: A human-computer interaction device; the human-computer interaction device is communicatively connected with the electronic device; The surgical navigation device includes a navigation camera, a patient tracer, a probe and a navigation tool.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the orthopedic surgery navigation method according to any one of claims 1 to 6 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the orthopedic surgery navigation method according to any one of claims 1 to 6 is implemented.

Images