Arthroscope positioning method and system for knee joint minimally invasive surgery

Abstract

The invention discloses an arthroscope positioning method and system for knee joint minimally invasive surgery. The anatomical mark points are recognized and tracked in real time through the operation video stream collected by the arthroscope camera in combination with the computer vision algorithm. The pre-planned ideal bone tunnel position and the anatomical marker in the real-time image are accurately registered, so that the system can dynamically generate surgical navigation information and visually display the position of the entrance point of the bone tunnel. According to the scheme, dependence of a traditional navigation system on a special surgical instrument and a complex calibration process is avoided, the visual field of the arthroscope familiar with a surgeon can be fully utilized, and more visual and convenient surgical guidance is provided, so that the surgical precision and efficiency are remarkably improved, and the risk of surgical wounds and complications is reduced.

Description

Technical Field

[0001] The present invention generally relates to the field of arthroscopic positioning, and more particularly to an arthroscopic positioning method and system for minimally invasive knee surgery. Background Art

[0002] During minimally invasive knee surgery, the surgeon inserts an arthroscope and surgical instruments through a small incision into the knee joint. The arthroscope is equipped with a lens and light source, providing illumination and magnification for viewing internal structures. Therefore, the surgeon relies on the arthroscopic view displayed on a screen above the patient to manipulate surgical tools and perform the procedure. Although arthroscopic surgery effectively reduces the size of the incision, minimally invasive knee surgery using arthroscopic navigation is more challenging than open surgery. This is primarily due to unintuitive hand-eye coordination and the limited range of motion of surgical instruments. Specifically, surgeons face several challenges when performing knee surgery using arthroscopic navigation. First, arthroscopic navigation provides only a partial view, which limits the surgeon's ability to accurately determine the positioning of surgical instruments. Understanding spatial orientation requires extensive clinical experience when relying solely on arthroscopic images. Due to these challenges, a recent report indicated that even experienced surgeons still experience significant lateral discrepancies and errors in nearly 30% of cases during arthroscopic navigation-based knee surgery for ligament reconstruction.

[0003] Improper femoral tunnel positioning during anterior cruciate ligament reconstruction is a major cause of graft failure. Accurate femoral tunnel positioning during arthroscopic surgery remains challenging due to limited visual field and difficulty identifying anatomical landmarks. Therefore, establishing a reliable and reproducible reference system for tunnel positioning is of great clinical significance. In existing research, Hart et al. proposed a "vertex reference system" based on the proximal-distal orientation, providing an important reference for tunnel positioning. However, the literature lacks reliable reference standards for the posterior / anterior orientation, which to some extent limits surgical precision and reconstructive outcomes. Using the CLR as a bony landmark to localize the anteroposterior position of the femoral tunnel during arthroscopic ACLR is an effective method. The CLR can be identified intraoperatively as a "white line" corresponding to the joint capsule, located at the medial and posterior edge of the lateral femoral condyle. After bone dissection with a bone planer, the CLR can be visualized, providing a stable and readily available posterior reference point. This method helps prevent excessive anterior positioning of the femoral tunnel and improves surgical accuracy and stability. Bernard et al., using standard lateral femoral X-rays, found that the center of the femoral insertion has a specific anatomical relationship, referenced by the Blumensaat line (intercondylar line): parallel to the posterior edge of the lateral femoral condyle, its distance to the posterior margin is 24.8% of the anteroposterior diameter of the lateral femoral condyle; vertically, its distance to the top of the intercondylar notch is 28.5% of the height of the intercondylar notch. Based on this finding, intraoperative C-arm X-ray fluoroscopy can be used to accurately locate the ACL bone tunnel. Although this method is considered the "gold standard" for bone tunnel positioning during ACL reconstruction, it has not been widely used in clinical practice due to its complex operational procedures. Summary of the Invention

[0004] To at least address the technical issues described in the background technology section above, the present invention proposes an arthroscopic positioning method and system for minimally invasive knee surgery, which can improve positioning accuracy. In view of this, the present invention provides solutions in the following aspects.

[0005] A first aspect of the present invention provides an arthroscopic positioning method for minimally invasive knee surgery, comprising: reconstructing a three-dimensional surface model of the distal femur based on a distal femoral image; cutting the three-dimensional surface model of the distal femur to obtain the lateral condyle; extracting a characteristic curve of the distal lateral condyle of the femur based on anatomical constraints, wherein the characteristic curve is a continuous curve formed by connecting characteristic points in a high curvature area; establishing a Bernard & Hertel grid on the three-dimensional model of the lateral condyle: by identifying the bony contour of the top of the intercondylar notch, a smooth continuous line is drawn between the front and rear boundary points to obtain the Blumensaat line, the top line of the intercondylar notch; using the Blumensaat line as the main reference, establishing two measurement directions: a depth line parallel to the Blumensaat line, the depth line running from the posterior edge to the anterior edge of the lateral condyle, and a height line perpendicular to the Blumensaat line, the height line running from the top of the intercondylar notch to the lowest point of the lateral condyle; based on the continuous curve, using a feature point registration method, the Bernard & Hertel grid is superimposed and displayed on the intraoperative arthroscopic image.

[0006] In one embodiment, reconstructing the three-dimensional surface model of the distal femur based on the distal femur image includes: using image segmentation technology to separate the femur from surrounding tissue; reconstructing the two-dimensional slice data obtained by segmentation into a three-dimensional model; and smoothing and repairing the generated three-dimensional model to eliminate noise and irregularities.

[0007] In one embodiment, cutting the three-dimensional surface model of the distal femur to obtain the lateral condyle includes: identifying key anatomical landmarks of the distal femur, including the medial condyle, the lateral condyle, and the intercondylar fossa located therebetween, which is a significantly depressed area; selecting two reference points located on the centerline of the anterior and posterior edges of the intercondylar fossa; defining a sagittal plane through these two reference points, wherein the sagittal plane is a longitudinal section, perpendicular to the coronal plane and the transverse plane, and should be perpendicular to the cross-section of the distal femur, while separating the medial condyle and the lateral condyle along the centerline of the intercondylar fossa.

[0008] In one embodiment, the anatomically constrained extraction of the characteristic curve of the lateral condyle of the distal femur includes: calculating the principal curvature information of the mesh vertices through a discrete differential geometry method; given a three-dimensional mesh model M = (V, E, F), for any vertex v_i∈V, its one-ring neighborhood is denoted as N(v_i), and the principal curvatures κ_1 and κ_2 of vi_i are obtained based on the curvature tensor estimation method of discrete differential geometry; the main directions corresponding to the principal curvatures are d_1 and d_2; thereby, the average curvature H(v_i) and Gaussian curvature K(v_i) can be obtained: H(v_i) = (κ_1+κ_2) / 2, K(v_i) = κ_1·κ_2; identifying the feature points of the high curvature area by setting a threshold τ, P_f = {v_i||K(v_i)|>τ}; and connecting the feature points into a continuous curve.

[0009] In one embodiment, the Bernard & Hertel grid is established on the three-dimensional model of the lateral condyle, including: in the standard lateral position, that is, when ensuring that the medial and lateral condyles completely overlap, on the sagittal plane reconstructed image, by identifying the bony contour of the top of the intercondylar fossa, a smooth continuous line is drawn between the anterior and posterior boundary points to obtain the Blumensaat line; with the Blumensaat line as the main reference, two key measurement directions are established: a depth line parallel to the Blumensaat line, the depth line runs from the posterior edge to the anterior edge of the lateral condyle, and a height line perpendicular to the Blumensaat line, the height line runs from the top of the intercondylar fossa to the lowest point of the lateral condyle.

[0010] In one embodiment, the Bernard & Hertel grid is superimposed and displayed on the intraoperative arthroscopic image based on the continuous curve using a feature point registration method, including: enhancing the arthroscopic image and extracting the joint capsule line; and establishing a spatial correspondence between the joint capsule line and the characteristic curve through a feature point registration method.

[0011] A second aspect of the present invention provides an arthroscopic positioning system for minimally invasive knee surgery, utilizing any of the above-mentioned arthroscopic positioning methods for minimally invasive knee surgery.

[0012] The present invention utilizes visual markers attached to the patient's anatomical structure, combined with Bernard & Hertel grid positioning technology, to provide surgeons with real-time, accurate bone tract positioning guidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0014] Figure 1 The present invention illustrates an arthroscopic positioning method for minimally invasive knee joint surgery according to an embodiment of the present invention;

[0015] Figure 2 1. The present invention illustrates reconstruction of a distal femur three-dimensional surface model from a CT image according to an embodiment of the present invention;

[0016] Figure 3 1. A distal femur model is shown using a sagittal plane to cut according to an embodiment of the present invention;

[0017] Figure 4 3D mesh feature curve extraction based on principal curvature analysis according to an embodiment of the present invention is illustrated;

[0018] Figure 5 1 is a diagram showing a characteristic curve of extracting the lateral condyle based on anatomical constraints according to an embodiment of the present invention;

[0019] Figure 6 FIG. 1 is a diagram illustrating establishing a Bernard & Hertel grid with the Blumensaat line as a reference according to an embodiment of the present invention;

[0020] Figure 7 1 is a diagram illustrating matching an arthroscopic image and a preoperative CT image based on a reference joint capsule line according to an embodiment of the present invention;

[0021] Figure 8 FIG2 is a diagram illustrating surgical navigation for achieving bone tract positioning by superimposing an arthroscopic image on a Bernard & Hertel grid in anterior cruciate ligament reconstruction surgery according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described 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 those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0023] It should be understood that the terms "first," "second," "third," and "fourth," etc. in the claims, description, and drawings of the present invention are used to distinguish different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the description and claims of the present invention indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0024] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the present invention. As used in the specification and claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should further be understood that the term "and / or" as used in the specification and claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0025] As used in this specification and claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0026] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0027] A first aspect of the present invention provides an arthroscopic positioning method for minimally invasive knee joint surgery. Figure 1 Schematic diagram of an arthroscopic positioning method for minimally invasive knee surgery according to an embodiment of the present invention. The method for cutting a three-dimensional model of a distal femur of the present invention comprises:

[0028] Step S100: reconstructing a three-dimensional surface model of the distal femur;

[0029] Step S200: cutting the distal femoral model to obtain the lateral condyle;

[0030] Step S300: extracting the characteristic curve of the lateral condyle of the distal femur based on anatomical constraints;

[0031] Step S400: creating a Bernard & Hertel mesh on the lateral condyle three-dimensional model;

[0032] Step S500: Overlaying and displaying the Bernard & Hertel grid on the intraoperative arthroscopic image.

[0033] In a preferred embodiment of the present invention, the specific implementation of the above step S100: reconstructing the distal femur three-dimensional surface model can be referred to Figure 2 .

[0034] Figure 2The present invention illustrates the reconstruction of a three-dimensional surface model of the distal femur from a CT image according to an embodiment of the present invention. The method proposed in the present invention is suitable for analyzing a three-dimensional surface model of the distal femur obtained using a CT, MRI, digitization, or laser scanning system. Without loss of generality, this implementation example uses a CT image as an example to reconstruct a three-dimensional surface model of the distal femur. The CT image is preprocessed to improve image quality and contrast. The femur is separated from surrounding tissues using image segmentation technology. Common methods include threshold-based segmentation, region growing, active contour models (such as Snake or Level Set methods), and deep learning-based segmentation (such as U-Net, etc.). The two-dimensional slice data obtained by segmentation is reconstructed into a three-dimensional model. The Marching Cubes algorithm or other volume rendering techniques are usually used to generate a three-dimensional surface model. The generated three-dimensional model is smoothed and repaired to eliminate noise and irregularities. Techniques such as mesh simplification and smoothing filtering can be used.

[0035] In a preferred embodiment of the present invention, the above step S200: cutting the distal femoral model to obtain the lateral condyle can be referred to in detail. Figure 3 .

[0036] like Figure 3 As shown in the figure, first identify the key anatomical landmarks of the distal femur: the medial and lateral condyles, and the intercondylar notch between them, a prominent concave area. Two reference points are selected at the anterior and posterior edges of the intercondylar notch, located on its centerline. A sagittal plane is defined through these two reference points. This sagittal plane is a longitudinal section perpendicular to the coronal and transverse planes and should be perpendicular to the cross-section of the distal femur. It also separates the medial and lateral condyles along the centerline of the intercondylar notch.

[0037] In a preferred embodiment of the present invention, the above step S300: extracting the characteristic curve of the lateral condyle of the distal femur based on anatomical constraints. The specific implementation method can be referred to Figure 4 .

[0038] Figure 4 The figure shows the extraction of three-dimensional mesh feature curves based on principal curvature analysis according to an embodiment of the present invention. The left figure shows the curvature calculation results, the middle figure shows the identification of feature points in high curvature areas, and the right figure shows the connection of feature points into a continuous curve. For the three-dimensional mesh model of the distal femur, feature curve extraction is performed based on principal curvature analysis. First, the principal curvature information of the mesh vertices is calculated using the discrete differential geometry method. Given a three-dimensional mesh model M = (V, E, F), for any vertex v i ∈V, its one-ring neighborhood is denoted as N(v i ), the curvature tensor estimation method based on discrete differential geometry is used to obtain v iThe main curvatures κ1, κ2; the main directions corresponding to the main curvatures are d1, d2; thus, the average curvature H(v i ) and Gaussian curvature K(v i )=κ1·κ2. By setting the threshold τ to identify the feature points of the high curvature area, P f ={v i ||K(v i )|>τ}. Based on the recognition of feature points, an improved region growing algorithm is used, combined with the three criteria of spatial adjacency, curvature similarity and directional continuity, to connect the feature points into a continuous curve: Based on the recognition of feature points, the process of curve connection using the improved region growing algorithm is as follows: First, the feature points are spatially sorted and filtered, the local curvature and main direction are calculated, and a KD tree index structure is established to accelerate the spatial search. The core of the algorithm is to start from the seed point, search for candidate connection points in its neighborhood, calculate the comprehensive connection probability based on the three criteria of spatial adjacency, curvature similarity and directional continuity, and select the point with the highest probability for connection. In the connection process, the minimum connection probability threshold is set, the maximum search radius is limited (2-3 times the average spacing of feature points), the smoothness of the curve growth direction is controlled, and bifurcations and intersections are properly handled. Finally, by removing abnormally short curves, merging similar parallel curves, smoothing nodes, and optimizing endpoint connections and other post-processing steps, the generated curve is ensured to have good geometric continuity. Such as Figure 4 As shown in the figure, the Gaussian curvature of the mesh vertices is first calculated; then the feature points in the high curvature area are identified; finally, the feature points are connected into a continuous curve.

[0039] Figure 5 Figure 2 shows the characteristic curves of the lateral condyle extracted based on anatomical constraints according to an embodiment of the present invention. The lateral condyle of the distal femur is located on the lateral portion of the distal femur, separated from the medial condyle by the intercondylar notch. It is connected to the patellar articular surface anteriorly and forms the posterolateral condyle surface posteriorly. Lateral condyle region Ω L ={p∈R 3 | <p-c,e i >∈[l i ,u i ], i = 1, 2, 3}, where c is the anatomical center of the lateral condyle, [l i ,u i ] is the anatomical boundary in all directions, e i is the unit vector of the anatomical coordinate axis. Figure 5 Shown is the result of extracting the characteristic curve of the lateral condyle based on anatomical constraints.

[0040] In a preferred embodiment of the present invention, the above step S400: establishing a Bernard & Hertel mesh in the lateral condyle three-dimensional model, the specific implementation of which can be referred to Figure 6 .

[0041] Figure 6 FIG. 1 is a diagram showing how to establish a Bernard & Hertel grid using the Blumensaat line as a reference according to an embodiment of the present invention. Figure 6 As shown in the figure, in the standard lateral position (ensuring complete overlap of the medial and lateral condyles), on the sagittal reconstructed image, by identifying the bony contour of the top of the intercondylar notch, a smooth continuous line is drawn between the anterior and posterior boundary points to obtain the Blumensaat line. With the Blumensaat line (top line of the intercondylar notch) as the main reference, two key measurement directions are established: the deep and shallow diameter line parallel to the Blumensaat line (from the posterior edge to the anterior edge of the lateral condyle) and the height line perpendicular to the Blumensaat line (from the top of the intercondylar notch to the lowest point of the lateral condyle). In this coordinate system, the center of the femoral insertion of the ACL is located at 24.8% (t value) from the posterior edge and 28.5% (h value) from the top. As shown in the figure, the deep and shallow diameter line parallel to the Blumensaat line (from the posterior edge to the anterior edge of the lateral condyle) and the height line perpendicular to the Blumensaat line (from the top of the intercondylar notch to the lowest point of the lateral condyle) are drawn between the anterior and posterior boundary points. As shown in the figure, the deep and shallow diameter line parallel to the Blumensaat line (from the posterior edge to the anterior edge of the lateral condyle) and the height line perpendicular to the Blumensaat line (from the top of the intercondylar notch to the lowest point of the lateral condyle) are drawn between the posterior edge and the ... Figure 6 As shown, the Bernard & Hertel grid was constructed on the lateral condyle using the Blumensaat line as a reference. The circle marks the center of the femoral insertion of the ACL. This method, through standardized anatomical positioning, provides a precise reference for bone tract positioning during ACL reconstruction surgery and has important clinical guidance.

[0042] In a preferred embodiment of the present invention, the above step S500: according to the reference curve, the Bernard & Hertel grid is superimposed on the arthroscopic image during the operation. The specific implementation method can be referred to Figure 7 .

[0043] In arthroscopic surgery, the use of video capture cards provides an efficient solution for exporting and recording real-time images. Specifically, you must first select a video capture card that is compatible with the output interface of the arthroscopic system and supports the target resolution and frame rate, and connect the video output of the arthroscopic system to the input port of the capture card through an adapter cable, and connect the capture card to a computer or other display device. Subsequently, install the capture card driver and related software to ensure that the device can correctly identify and connect to the video signal source. In the acquisition software, you need to further select the corresponding video input source and configure the resolution, frame rate, and encoding format to optimize image quality and real-time performance. In addition, the recording function can be enabled as needed to save surgical images. Through the above process, real-time monitoring and high-quality export of surgical images can be achieved, providing reliable support for intraoperative navigation and postoperative analysis.

[0044] Figure 7 1 is a diagram showing how to match arthroscopic images and preoperative CT images based on a reference joint capsule line according to an embodiment of the present invention. Figure 7As shown, the arthroscopic image is first enhanced and the joint capsule line is extracted. Simultaneously, the CT image is reconstructed 3D to obtain the lateral condyle characteristic curve described in step 3. A feature point registration method, including anatomical landmark marking, SIFT / SURF feature extraction, and the ICP algorithm, is then used to establish a spatial correspondence between the two modal images. The process is as follows: A professional doctor first marks the corresponding anatomical landmarks on the arthroscopic image and the CT 3D model to establish an initial coarse registration transformation matrix. The SIFT / SURF algorithm is then used to extract local features from the two modal images: scale-invariant feature points and their 128-dimensional descriptors are extracted from the arthroscopic image, and corresponding features are extracted from the CT projection image. Feature point correspondence is established using a nearest neighbor ratio matching strategy, and RANSAC is used to eliminate incorrect matching points to improve the reliability of feature matching. Finally, the ICP algorithm is applied for precise registration: using the feature point matching results as the initial value, the root mean square distance between the corresponding point sets is minimized through iterative optimization. The kd tree is used to accelerate the nearest neighbor search, the convergence threshold is set to 0.01 mm, and the maximum number of iterations is 100. Finally, an accurate spatial transformation matrix is ​​obtained to achieve accurate spatial correspondence between the two modal images.

[0045] Figure 8 FIG2 is a diagram illustrating surgical navigation for achieving bone tract positioning by superimposing an arthroscopic image on a Bernard & Hertel grid in anterior cruciate ligament reconstruction surgery according to an embodiment of the present invention. Figure 8 The above picture shows the original arthroscopic image during ACLR surgery. Figure 8 The figure below shows the arthroscopic image superimposed with the Bernard & Hertel grid.

[0046] like Figure 8 The figure shows the final effect of surgical navigation for bone tunnel positioning by superimposing the Bernard & Hertel grid on the arthroscopic image. The system accurately superimposes the Bernard & Hertel standard 4×4 grid, which has undergone perspective transformation and spatial correction, on the real-time arthroscopic video image. The grid is presented in a semi-transparent manner, ensuring clear visibility of the original anatomical structure while providing an accurate spatial positioning reference. The precise correspondence between key anatomical landmarks such as the posterior edge of the lateral femoral condyle and the Blumensaat line and the grid can be seen in the image. The system updates the grid position in real time through a dynamic tracking algorithm to ensure that it always remains correctly aligned with the anatomical structure. The ideal bone tunnel entry point is displayed as a prominent marker on the grid. This intuitive visualization solution enables surgeons to accurately grasp the position of the bone tunnel within the familiar arthroscopic field of view, effectively improving the accuracy and safety of surgical operations.

[0047] The second aspect of the present invention discloses an arthroscopic positioning system for minimally invasive knee joint surgery, utilizing the arthroscopic positioning method for minimally invasive knee joint surgery disclosed above.

[0048] Although this specification has shown and described a plurality of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will appreciate that many modifications, variations, and alternatives can be made without departing from the spirit and scope of the present invention. It should be understood that in practicing the present invention, various alternatives to the embodiments of the present invention described herein may be employed. The appended claims are intended to define the scope of protection of the present invention and therefore cover modular compositions, equivalents, or alternatives within the scope of these claims.

Claims

1. An arthroscopic positioning method for minimally invasive knee surgery, characterized in that: include: reconstructing a three-dimensional surface model of the distal femur according to the distal femur image; cutting the distal femoral three-dimensional surface model to obtain the lateral condyle; Extraction of a characteristic curve of the lateral condyle of the distal femur based on anatomical constraints, wherein the characteristic curve is a continuous curve formed by connecting characteristic points in a high curvature area; A Bernard & Hertel mesh was constructed on the 3D model of the lateral condyle. The bony contour of the intercondylar notch was identified and a smooth, continuous line was drawn between the anterior and posterior boundary points to obtain the Blumensaat line, the intercondylar notch top line. Using the Blumensaat line as the primary reference, two measurement directions were established: a deep and shallow line parallel to the Blumensaat line, running from the posterior to the anterior edge of the lateral condyle, and a height line perpendicular to the Blumensaat line, running from the intercondylar notch top to the lowest point of the lateral condyle. According to the continuous curve, the Bernard & Hertel grid is superimposed and displayed on the intraoperative arthroscopic image using a feature point registration method.

2. The arthroscopic positioning method for minimally invasive knee surgery according to claim 1, characterized in that: The method of reconstructing a three-dimensional surface model of the distal femur according to the distal femur image comprises: Image segmentation technology is used to separate the femur from surrounding tissue; the two-dimensional slice data obtained by segmentation are reconstructed into a three-dimensional model; and the generated three-dimensional model is smoothed and repaired to eliminate noise and irregularities.

3. The arthroscopic positioning method for minimally invasive knee surgery according to claim 1, characterized in that: Cutting the distal femoral three-dimensional surface model to obtain the lateral condyle includes: Identify key anatomical landmarks of the distal femur, including the medial and lateral condyles and the intercondylar notch, a prominent depression located between them. At the anterior and posterior edges of the intercondylar notch, select two reference points located on its centerline; A sagittal plane is defined by these two reference points. The sagittal plane is a longitudinal section perpendicular to the coronal and transverse planes and should be perpendicular to the cross section of the distal femur while dividing the medial and lateral condyles along the centerline of the intercondylar notch.

4. The arthroscopic positioning method for minimally invasive knee surgery according to claim 3, characterized in that: The extraction of the characteristic curve of the lateral condyle of the distal femur based on anatomical constraints includes: Calculate the principal curvature information of mesh vertices through discrete differential geometry methods; Given a three-dimensional mesh model M = (V, E, F), for any vertex v_i∈V, its one-ring neighborhood is denoted as N(v_i). The principal curvatures κ_1 and κ_2 of v_i are obtained based on the curvature tensor estimation method of discrete differential geometry; the principal directions corresponding to the principal curvatures are d_1 and d_2. From this, the mean curvature H(v_i) and Gaussian curvature K(v_i) can be obtained: H(v_i) = (κ_1 + κ_2) / 2, K(v_i) = κ_1·κ_2. The feature points of the high curvature area are identified by setting a threshold τ, P_f = {v_i||K(v_i)|>τ}. Connect feature points into a continuous curve.

5. The arthroscopic positioning method for minimally invasive knee surgery according to claim 4, characterized in that: The Bernard & Hertel mesh is established in the lateral condyle three-dimensional model, including: In the standard lateral position, that is, when the medial and lateral condyles are completely overlapped, the bony contour of the top of the intercondylar notch is identified on the sagittal reconstructed image, and a smooth continuous line is drawn between the anterior and posterior boundary points to obtain the Blumensaat line; Using the Blumensaat line as the primary reference, two key measurement directions were established: a depth line parallel to the Blumensaat line, which runs from the posterior edge to the anterior edge of the lateral condyle, and a height line perpendicular to the Blumensaat line, which runs from the top of the intercondylar notch to the lowest point of the lateral condyle.

6. The arthroscopic positioning method for minimally invasive knee surgery according to claim 5, characterized in that: The method of superimposing and displaying the Bernard & Hertel grid on the intraoperative arthroscopic image using a feature point registration method based on the continuous curve includes: Enhance the arthroscopic image and extract the joint capsule line; By using a feature point registration method, the joint capsule line and the feature curve are used to establish a spatial correspondence between the two modality images.

7. An arthroscopic positioning system for minimally invasive knee surgery, characterized in that: The method is performed using the arthroscopic positioning method for minimally invasive knee surgery as described in any one of claims 1 to 6.

Images