Artificial intelligence assisted chest wall bony reconstruction system and three-dimensional model pre-cutting method

Abstract

The invention relates to the technical field of medicine, and discloses an artificial intelligence assisted chest wall bony reconstruction system and a three-dimensional model pre-cutting method.The three-dimensional model is pre-cut and measured, and the size and number of prostheses needed for reconstruction are determined; based on the size and the number of prostheses, producing an individualized 3D preformed reconstructed titanium plate; selecting a titanium plate with proper length according to the size and shape of the thoracic wall defect, cutting and shaping by using a corresponding tool, and recovering the natural radian of the thorax as far as possible; in the operation, the reconstruction plate and the fixing plate are combined and fixed through the locking screws, and individualized thoracic wall bony reconstruction is completed. According to the reconstruction system, through combination of different components, embedding and insertion of the clamping grooves and longitudinal and transverse connection of the screws, fixation is firm, and the system can be suitable for reconstruction of almost all types of chest wall defects, so that the clinical curative effect is improved, and complications are reduced.

Description

Technical Field

[0001] This invention relates to the field of medical technology, and in particular to an artificial intelligence-assisted chest wall bony reconstruction system and a three-dimensional model pre-cutting method. Background Technology

[0002] For most malignant tumors of the chest wall, surgical resection is the primary treatment. However, extensive chest wall defects resulting from surgery require reconstruction. The main goals of all chest wall reconstructions are to eliminate dead space, restore chest wall rigidity, maintain respiratory mechanics, protect intrathoracic organs, provide soft tissue coverage, and minimize deformities. During chest wall reconstruction, it is not necessary to reconstruct every single rib.

[0003] Currently, there is no consensus guideline clearly defining the absolute indications for chest wall reconstruction, leading to significant differences in clinical surgical procedures. The key lies in selecting appropriate reconstruction materials based on the tumor location and resection extent, thereby ensuring the rationality and simplicity of the reconstruction surgery. Furthermore, three-dimensional reconstruction technology still has some issues, such as incomplete matching between anatomical structures and preoperative 3D models; in addition, image data quality and reconstruction time are also key factors affecting the effectiveness of three-dimensional reconstruction and require further improvement. Summary of the Invention

[0004] The main objective of this invention is to address the technical problem of the lack of a clear consensus guideline defining the absolute indications for chest wall reconstruction. An artificial intelligence-assisted chest wall bony reconstruction system is provided, comprising: The module for obtaining the resection range of chest wall tumors analyzes and obtains the resection range of chest wall tumors, designs surgical incisions, and calculates the safe resection range, thereby maximizing the removal of tumor tissue and preserving the maximum amount of healthy tissue. By pre-cutting and measuring the 3D model, the required size and number of prostheses for reconstruction are determined; based on the size and number of prostheses, individualized 3D pre-formed reconstruction titanium plates are produced. Select titanium plates of appropriate length according to the size and shape of the chest wall defect, and use appropriate tools to cut and shape them to restore the natural curvature of the chest as much as possible. During the procedure, the reconstruction plate and fixation plate are combined and fixed with locking screws to complete individualized bony reconstruction of the chest wall.

[0005] This invention also relates to a method for pre-cutting a three-dimensional model, comprising the following steps: Step S1: Obtain the 3D surface model to be pre-cut, and create a 2D display canvas corresponding to the current viewpoint in the 3D view window; Step S2: Receive the closed pre-cut line drawn by the user on the two-dimensional display canvas, and map the two-dimensional coordinates of the pre-cut line into a three-dimensional pre-cut line in the three-dimensional surface model space through the inverse projection transformation corresponding to the current viewpoint. Step S3: Based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewpoint direction, calculate at least one cutting plane, which is determined by the spatial normal vector and a point passing through the plane; Step S4: Generate an interactive set of control points on the cutting plane, receive user selection and drag operations on the control points, and update the spatial parameters of the cutting plane in real time according to the changes in the position of the control points. Step S5: Based on the updated cutting plane, traverse the triangular facets of the three-dimensional surface model, determine the relative positional relationship between each triangular facet and the cutting plane, calculate the intersection points of the triangular facets located on both sides of the cutting plane with the cutting plane, and construct the closed cross-sectional profile on the cutting plane from the intersection points. Step S6: Based on the closed cross-sectional profile, perform a surface trimming Boolean operation on the three-dimensional surface model to obtain two independent three-dimensional models.

[0006] A second aspect of the present invention provides a three-dimensional model pre-cutting system, comprising: The model acquisition and display module is used to acquire the three-dimensional surface model to be pre-cut, establish a two-dimensional display canvas corresponding to the three-dimensional surface model in the three-dimensional view window according to the current viewing angle, and maintain the projection mapping relationship between the two-dimensional display canvas and the three-dimensional model space. The pre-cutting line acquisition and mapping module is used to receive closed pre-cutting lines drawn by the user on the two-dimensional display canvas, and to calculate the two-dimensional coordinates of the closed pre-cutting lines in reverse through the projection mapping relationship into three-dimensional pre-cutting lines in the three-dimensional surface model space. The cutting plane generation module is used to calculate at least one cutting plane based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewing angle direction. The cutting plane is determined by the spatial normal vector and a point passing through the plane. The cutting plane interactive adjustment module is used to generate a set of control points on the cutting plane, receive user selection and drag operations of the control points, and update the spatial parameters of the cutting plane according to the changes in the position of the control points. The cross-sectional profile calculation module is used to traverse the triangular facets of the three-dimensional surface model based on the updated cutting plane, determine the relative positional relationship between each triangular facet and the cutting plane, calculate the intersection points of the triangular facets located on both sides of the cutting plane, and construct a closed cross-sectional profile located on the cutting plane from the intersection points. The surface trimming module is used to perform a surface trimming Boolean operation on the three-dimensional surface model based on the closed cross-sectional contour, dividing the three-dimensional surface model into two independent three-dimensional models.

[0007] The present invention has the following beneficial effects: Compared with the prior art, the present invention has at least the following beneficial effects: (1) AI provides a complete individualized surgical plan based on the case; (2) Both the fixation plate and the reconstruction plate are made of titanium metal, which is lightweight, well tolerated by patients, and does not cause a strong foreign body sensation. It will not affect the follow-up examination after surgery. Moreover, most of the mature products widely used in clinical practice (sternal and rib fixation titanium plates) are made of this material. (3) The titanium plate and locking screws are used to fix the bone defect ends very securely. The fixed steel plate can move with the chest wall up and down with the remaining ribs or sternum, which has little impact on the postoperative lung ventilation function of the patient. (4) Personalized surgery: the defect can be cut according to its length and shaped according to its curvature and extent. Attached Figure Description

[0008] Figure 1 The software interface diagram for drawing pre-cut lines in this invention is shown.

[0009] Figure 2 A planar software interface diagram is created for this invention.

[0010] Figure 3 This is a diagram of the software interface for selecting curved surfaces according to the present invention.

[0011] Figure 4 This is a diagram of the software interface for surface trimming in this invention.

[0012] Figure 5 This is a diagram of the software interface for distance measurement on a two-dimensional plane according to the present invention.

[0013] Figure 6 This is a software interface diagram for angle measurement in a two-dimensional plane according to the present invention.

[0014] Figure 7 This is a diagram of the software interface for distance measurement of the 3D model of this invention.

[0015] Figure 8 This is a diagram of the software interface for measuring the angles of the 3D model of this invention.

[0016] Figure 9 This is a diagram of the software interface for volume measurement of the three-dimensional model of this invention.

[0017] Figure 10 A schematic diagram of the new prefabricated titanium chest wall for reconstruction.

[0018] Figure 11 This is a schematic diagram of a Type I fixed plate structure.

[0019] Figure 12 This is a schematic diagram of a Type II fixed plate structure.

[0020] Figure 13 This is a schematic diagram of the reconstruction slab structure.

[0021] Figure 14 This is a schematic diagram of the locking screw structure.

[0022] Figure 15 This is a schematic diagram for small-scale rib resection (with fixation plates and reconstruction plates used for transverse and longitudinal fixation).

[0023] Figure 16 This is a schematic diagram for sternal resection and reconstruction (the tail end of the reconstruction plate can be directly fixed to the sternum or used in conjunction with a fixation plate).

[0024] Figure 17 This is a schematic diagram of a structure used for the resection and reconstruction of multiple ribs.

[0025] Figure 18 This is a schematic diagram for single rib resection and reconstruction.

[0026] Figure 19 A schematic diagram for resection and reconstruction of a single symmetrical rib.

[0027] The attached figures are labeled as follows: 1-1 Trapezoidal groove 1-2 Locking holes 1-3 Threaded hole 1 1-4 Folding Groove 2-1 Threaded hole 2 3-2 Cross groove. Detailed Implementation

[0028] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 The first embodiment of the three-dimensional model pre-cutting method in this invention includes: Step S1: Obtain the 3D surface model to be pre-cut, and create a 2D display canvas corresponding to the current viewpoint in the 3D view window; Step S2: Receive the closed pre-cut line drawn by the user on the two-dimensional display canvas, and map the two-dimensional coordinates of the pre-cut line into a three-dimensional pre-cut line in the three-dimensional surface model space through the inverse projection transformation corresponding to the current viewpoint. Step S3: Based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewpoint direction, calculate at least one cutting plane, which is determined by the spatial normal vector and a point passing through the plane; Step S4: Generate an interactive set of control points on the cutting plane, receive user selection and drag operations on the control points, and update the spatial parameters of the cutting plane in real time according to the changes in the position of the control points. Step S5: Based on the updated cutting plane, traverse the triangular facets of the three-dimensional surface model, determine the relative positional relationship between each triangular facet and the cutting plane, calculate the intersection points of the triangular facets located on both sides of the cutting plane with the cutting plane, and construct the closed cross-sectional profile on the cutting plane from the intersection points. Step S6: Based on the closed cross-sectional contour, perform a surface trimming Boolean operation on the three-dimensional surface model to obtain two independent three-dimensional models.

[0030] By drawing closed pre-tangent lines on a two-dimensional display canvas and establishing an inverse projection transformation relationship between two-dimensional coordinates and three-dimensional model space in conjunction with the current viewing perspective, the closed pre-tangent lines are mapped to three-dimensional pre-tangent lines. Therefore, the cutting boundary corresponding to the position in the model space can be determined without directly manipulating the surface of the three-dimensional model.

[0031] Since the cutting plane is calculated based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewing angle direction, and the cutting plane is spatially defined by the normal vector and the points passing through the plane, the process of generating the cutting plane can maintain consistency with the pre-cutting lines drawn by the user.

[0032] Because interactive control points are set on the cutting plane, and the spatial parameters of the cutting plane are updated in real time according to the user's selection and dragging of the control points, the position and orientation of the cutting plane can be continuously adjusted during the interaction.

[0033] Because the triangular facets of the 3D surface model are re-recalculated after the cutting plane parameters change, and a closed cross-sectional profile is constructed based on the relative positional relationship between the plane and the triangular facets, the cutting boundary can be updated synchronously with the adjustment of the cutting plane.

[0034] Since the three-dimensional surface model is divided into two independent parts by performing a surface trimming Boolean operation on the closed cross-sectional contour, the segmented model structure can be obtained in one trimming process.

[0035] Specifically: The three-dimensional surface model to be pre-cut is obtained from the model data source. The three-dimensional surface model is composed of multiple triangular facets and is stored in the model space in the form of three-dimensional coordinates. In the 3D view window, the 3D surface model is set to visible, and the viewing angle parameters, including viewpoint position, viewing direction and projection method, are determined according to the current user operation. Based on the observation perspective parameters, a two-dimensional display canvas corresponding to the three-dimensional view is established on the display terminal to receive the user's screen interaction input. There is a one-to-one projection mapping relationship between the two-dimensional display canvas and the three-dimensional model space.

[0036] The system receives manual drawing input from the user on the two-dimensional display canvas. Obtain a set of two-dimensional screen coordinate points arranged in chronological order, and construct a closed pre-tangent line based on the two-dimensional coordinate points; Based on the projection mapping relationship established in step S1, the inverse projection transformation calculation is performed on each two-dimensional coordinate point in the closed pre-tangent line to map it to the corresponding three-dimensional coordinate point in the three-dimensional model space, thereby forming a three-dimensional pre-tangent line located on the surface of the three-dimensional surface model or in its adjacent space.

[0037] Based on the spatial distribution characteristics of the three-dimensional pre-cutting lines and the viewing direction of the current three-dimensional view, a cutting plane for model segmentation is calculated. The cutting plane is determined in the following way: based on the spatial positions of at least some points in the three-dimensional pre-cutting lines and combined with the normal vector related to the viewing direction, a plane normal vector is calculated; at the same time, a point located on the three-dimensional pre-cutting lines is selected as a point passing through the plane, thereby uniquely determining the cutting plane. The cutting plane is visualized in a 3D view, and control points for interactive adjustment are generated on the cutting plane.

[0038] Receive user selection and drag operations on control points on the cutting plane, and convert the operations into changes in the position of the control points in three-dimensional space; Based on the changes in the control point positions, the spatial parameters of the cutting plane are updated in real time, including the direction of the plane's normal vector and the position of the points passing through the plane, so that the cutting plane undergoes corresponding displacement, rotation, or deformation in the three-dimensional model space. After the cutting plane parameters are updated, the system synchronously updates the display status of the cutting plane in the 3D view.

[0039] Based on the currently updated cutting plane, each triangular facet in the three-dimensional surface model is traversed and calculated one by one. For each triangular facet, determine the spatial positional relationship of its vertices relative to the cutting plane; when the vertices of the triangular facet are located on different sides of the cutting plane, calculate the intersection point of the cutting plane and the edge of the triangular facet. Connect all the intersections that meet the conditions according to their spatial adjacency to construct one or more closed cross-sectional profiles located on the cutting plane. These closed cross-sectional profiles are used to define the cutting boundaries of the model.

[0040] Based on the closed cross-sectional profile, perform a surface trimming Boolean operation on the three-dimensional surface model; Based on the cutting plane and the closed cross-sectional contour, the three-dimensional surface model is divided into two independent three-dimensional models along the cutting plane, and corresponding surface model data structures are generated for each part. After the segmentation is completed, the two parts of the 3D model are displayed independently in the 3D view to complete the pre-slicing operation.

[0041] The pre-cutting tool is mainly used to manually draw the pre-cut surface, and then deform the pre-cut surface to divide the model into left and right parts with curved surfaces, which helps with surgical planning.

[0042] The process of surface trimming can be roughly summarized as: drawing pre-cutting lines - creating a plane - selecting the surface - completing the trimming.

[0043] The core principle of drawing pre-cut lines and creating planes is as follows: First, click on the 2D canvas coordinates on the screen. Through a 3D-to-2D projection transformation, the system calculates the specific position of the pre-cut line in the 3D model space. Based on the shape of the pre-cut line (e.g., a straight line or a closed curve) and the current view orientation, the system automatically calculates and generates one (or more) spatial cutting planes. This plane is typically precisely defined by a normal vector and a point passing through the plane.

[0044] The system then iterates through all the triangular faces of the model, determining their positional relationship to the cutting plane. If the three vertices of a triangle are located on opposite sides of the cutting plane, the intersection points of the plane and the triangle's sides are calculated. Connecting all intersection points forms one or more closed contour lines on the cutting plane. The area enclosed by these contour lines is the new cross-section created after the model is cut.

[0045] The core principle of point-based surface cutting is that, in arbitrary path cutting, the points on the path generated by "point-based cutting" are the control points. Users can directly drag these points to modify the shape of the cutting path. The system will then re-execute the above calculation process based on the new path and immediately update the cutting results. This provides extremely high flexibility, allowing doctors to perform "minimally invasive" simulated cuts.

[0046] Surface trimming is expressed using implicit functions. By selecting and dragging its control points, the parameters of the implicit functions can be directly changed. This changes the distribution of function values ​​in real time, which in turn changes the shape and size of the trimming body, thus allowing precise control over which parts of the model are cut off and which are preserved.

[0047] The cutting principle is based on Boolean operations, simulating the effect after cutting, dividing the model into "parts to be removed" and "parts to be retained", thus splitting the curved surface cutting model into two parts.

[0048] See the present invention Figure 1-4 The operation in the software of this invention is as follows: [1] Select the face model to be cut in the data manager in the upper left corner of the 3D view window to make the face model visible, and adjust the view of the face model to a suitable position.

[0049] [2] Click the surface cutting tool icon, and in the pop-up function box, click the "Draw Pre-cut Line" button. Draw a closed curve on the surface model to be cut as needed. For example... Figure 1 As shown.

[0050] [3] In the pop-up function box, click the "Create Plane" button. The system will automatically generate a plane based on the drawn closed curve, dividing the plane model into two parts. On the created plane, click the white dot to select it; the white dot will change color to green. At this point, you can drag the green dot to adjust the size and orientation of the plane. For example... Figure 2 As shown.

[0051] [4] After point adjustment is complete, click the "Select Surface" button in the pop-up function box. The system will then switch from point adjustment to surface adjustment. Select the center of the surface area to be adjusted, slide the mouse to adjust the size of the surface, and then drag to adjust the surface. For example... Figure 3 As shown [5] In the pop-up function box, click the "Surface Trimming" button. The system will execute the surface trimming function. After trimming, the surface model will be divided into two parts, displayed in different colors. The volume and percentage of the two parts after pre-trimming will be displayed in the upper left corner of the 3D view. Figure 4 As shown [6] If you are not satisfied with the surface trimming result, you can click the "Select Surface" button again, and the system will go back to step 3. Adjust the surface to a satisfactory position and perform pre-cutting.

[0052] In addition, the software of this invention also includes five measurement tools. These are: ① Distance Measurement: Used for distance measurement within a two-dimensional plane image. Select a cross-section of the two-dimensional plane image, click the measurement tool, and then click on two points within that cross-section to obtain the distance information between the two points. See the image below. The principle is to obtain the world coordinates of the two points and calculate the distance between them using a distance calculation function. Figure 5 As shown.

[0053] ② Angle Measurement: Used for measuring angles within a two-dimensional plane image. Select a cross-section of the two-dimensional plane image, click the angle measurement tool, and then click three points within that cross-section to form an angle. The angle information of this angle can then be obtained. See the image below. The principle is to obtain the world coordinates of the three points, calculate the cosine value of the angle, and finally use the inverse cosine function to calculate the angle. Figure 6 As shown.

[0054] ③ Distance: Used for distance measurement on 3D models. Select the 3D space interface, and click at least two points on the 3D model to measure the distance information of the line connecting these points. The distance information of the line projected onto the 2D image will then be displayed. The principle is to obtain the 3D world coordinates of two points and apply the 3D Euclidean distance formula to obtain the true spatial distance between the two points. For example... Figure 7 As shown.

[0055] ④ Angle: Used for angle measurement on 3D models. Select the 3D space interface, and click at least three points on the 3D model to measure the angle formed by connecting these points. The angle is then displayed on a 2D image as the projection of this angle onto the 2D image. The principle is to define a fixed angle point B and two adjacent points A and C. The system converts these three points into spatial coordinates and generates vectors BA and BC. The angle is then obtained based on the dot product formula. For example... Figure 8 As shown.

[0056] ⑤ Volume: Used to measure the volume of a 3D model. Click on the 3D model you want to measure, and its volume will be displayed in the upper left corner of the 3D interface. The principle is that any closed triangular mesh can be decomposed into a series of tetrahedrons with the origin as their common vertex. By calculating the sum of the directed volumes of these tetrahedrons, the volume of the entire mesh can be obtained.

[0057] Volume calculation: The algorithm iterates through every triangle in the mesh. For each triangle, it forms a tetrahedron with the origin O(0, 0, 0) in 3D space. Then, the directed volume of this tetrahedron is calculated using a formula.

[0058] Surface area calculation: Surface area calculation is relatively straightforward. The algorithm also iterates through all triangles, using the formula to calculate the area of ​​each triangle. For example... Figure 9 As shown.

[0059] This invention also relates to an artificial intelligence-assisted bony reconstruction system for the chest wall. Based on the above pre-cutting and measurement, a novel prefabricated titanium chest wall can be constructed, as shown in the results. Figure 10 The specific process includes: (1) AI learning is used to help determine the extent of chest wall tumor resection, and surgical incision design and safe resection range calculation can be performed to maximize the removal of tumor tissue and preserve healthy tissue to the maximum extent. (2) Based on the resection range, AI simulates three-dimensional reconstruction of the defect to determine the size and number of prostheses required for reconstruction. (3) Individualized 3D pre-formed reconstruction titanium plates are directly produced by digital machine tools to meet the reconstruction needs of various defects. The appropriate length of titanium plate is selected according to the size and shape of the chest wall defect, and the appropriate tools are used for cutting and shaping to restore the natural curvature of the thoracic cavity as much as possible. (4) During the operation, individualized bony reconstruction of the chest wall is completed by simply inserting tissue and fixing with locking screws.

[0060] Digital machine tool production of individualized reconstructed titanium plates The dimensions and quantity of the reconstructed titanium plates have been determined, and digital machine tool production of the reconstructed titanium plates is now possible. See the attached document for the specific structure. Figures 11-14 The structure is described below: (1) Fixing plate: There are two types, Type I and Type II. The difference between Type I and Type II is that Type II adds a fracture groove structure. ① It has a trapezoidal groove into which the reconstruction plate can be embedded. The trapezoidal groove has threaded holes, and the reconstruction plate is fixed by the cooperation of screws and threaded holes. The sides of the trapezoidal groove are arched, allowing the reconstruction plate to slide within the groove before fixation, or to be adjusted at a certain angle according to the resection location. ② It has a boss with locking holes, which are also threaded holes, to connect the fixation plate to the remaining ribs. The fixation plate and reconstruction plate are cut according to the chest wall resection area to fit the resection area. (The Type II fixation plate has a fracture groove on the boss, making it easy to cut).

[0061] (2) Reconstruction plate: ① It has several threaded holes that can be inserted into the trapezoidal groove of the fixing plate. ② It can be bent to simulate the shape of the ribs, and can also be trimmed according to the chest wall reconstruction of different resection ranges.

[0062] (3) Locking screws: ① The screw head has a cross groove at the upper end, which can be used with surgical instruments; ② The screw head has a double-ended tapered thread, which is convenient for use with fixation plates and reconstruction plates, and for fixing them to the ribs or sternum.

[0063] 4. Customized combination of reconstruction plate and fixing plate During the surgery, the reconstruction plate and fixation plate are combined and secured with locking screws to complete individualized bony reconstruction of the chest wall. Based on the size and quantity of titanium plates obtained in step 2, the reconstruction plate and fixation plate are selected. If necessary, the titanium plates can be cut to the required length using specialized tools; if shaping is required, sharp bends, reverse bends, or bending operations on the implant at the threaded holes should be avoided. Scratches or marks on the implant should also be avoided. Lateral reconstruction requires fixation to the two rib remnants, while longitudinal reconstruction requires connection between the sternal remnants. The combination method is shown in the appendix. Figures 15-19 This combination is for reference only, and the combination method can be changed according to the actual surgical resection range.

[0064] The three-dimensional model pre-slicing method in the embodiments of the present invention has been described above. The three-dimensional model pre-slicing system in the embodiments of the present invention is described below: The model acquisition and display module is used to acquire the three-dimensional surface model to be pre-cut, establish a two-dimensional display canvas corresponding to the three-dimensional surface model in the three-dimensional view window according to the current viewing angle, and maintain the projection mapping relationship between the two-dimensional display canvas and the three-dimensional model space. The pre-cutting line acquisition and mapping module is used to receive closed pre-cutting lines drawn by the user on the two-dimensional display canvas, and to calculate the two-dimensional coordinates of the closed pre-cutting lines in reverse through the projection mapping relationship into three-dimensional pre-cutting lines in the three-dimensional surface model space. The cutting plane generation module is used to calculate at least one cutting plane based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewing angle direction. The cutting plane is determined by the spatial normal vector and a point passing through the plane. The cutting plane interactive adjustment module is used to generate a set of control points on the cutting plane, receive user selection and drag operations of the control points, and update the spatial parameters of the cutting plane according to the changes in the position of the control points. The cross-sectional profile calculation module is used to traverse the triangular facets of the three-dimensional surface model based on the updated cutting plane, determine the relative positional relationship between each triangular facet and the cutting plane, calculate the intersection points of the triangular facets located on both sides of the cutting plane, and construct a closed cross-sectional profile located on the cutting plane from the intersection points. The surface trimming module is used to perform a surface trimming Boolean operation on the three-dimensional surface model based on the closed cross-sectional contour, dividing the three-dimensional surface model into two independent three-dimensional models.

[0065] This method transforms labeled skeletal images into 3D models through 3D reconstruction. The inspiration comes from the idea of ​​"reconstructing the prototype of an object by stacking a bunch of parallel thin sheets." It first precisely identifies and delineates the boundaries of the bones on each 2D CT or MRI slice image (image segmentation). Then, based on the spatial relationship of these contour lines, the computer automatically "skins" adjacent contour layers with countless triangular facets, ultimately forming a complete, closed 3D surface model. This method generates the "surface" of the model, resulting in relatively small file sizes and high processing efficiency, making it ideal for 3D printing and surgical planning.

[0066] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An artificial intelligence-assisted bony reconstruction system for the chest wall, characterized in that, The system includes: The module for obtaining the resection range of chest wall tumors analyzes and obtains the resection range of chest wall tumors, designs surgical incisions, and calculates the safe resection range, thereby maximizing the removal of tumor tissue and preserving the maximum amount of healthy tissue. By pre-cutting and measuring the 3D model, the required size and number of prostheses for reconstruction are determined; based on the size and number of prostheses, individualized 3D pre-formed reconstruction titanium plates are produced. Select titanium plates of appropriate length according to the size and shape of the chest wall defect, and use appropriate tools to cut and shape them to restore the natural curvature of the chest as much as possible. During the procedure, the reconstruction plate and fixation plate are combined and fixed with locking screws to complete individualized bony reconstruction of the chest wall.

2. The system according to claim 1, characterized in that, The individualized 3D preformed reconstructed titanium plate includes: A fixing plate, one end of which is provided with a trapezoidal groove for connecting with a reconstruction plate, the side of which is arched and a first threaded hole is provided on the trapezoidal groove; the other end of the fixing plate is provided with a boss for connecting with residual bone, and a second threaded hole is provided on the boss. A reconstruction plate, the end of which can be embedded in the trapezoidal groove of the fixing plate, and the reconstruction plate is provided with a plurality of third threaded holes; A locking screw, the head of which has a double-ended tapered thread for screwing into the first threaded hole, the second threaded hole, or the third threaded hole; The reconstruction plate is bendable, and both the reconstruction plate and the fixing plate are cut.

3. A method for pre-cutting a three-dimensional model, characterized in that, Includes the following steps: Step S1: Obtain the 3D surface model to be pre-cut, and create a 2D display canvas corresponding to the current viewpoint in the 3D view window; Step S2: Receive the closed pre-cut line drawn by the user on the two-dimensional display canvas, and map the two-dimensional coordinates of the pre-cut line into a three-dimensional pre-cut line in the three-dimensional surface model space through the inverse projection transformation corresponding to the current viewpoint. Step S3: Based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewpoint direction, calculate at least one cutting plane, which is determined by the spatial normal vector and a point passing through the plane; Step S4: Generate an interactive set of control points on the cutting plane, receive user selection and drag operations on the control points, and update the spatial parameters of the cutting plane in real time according to the changes in the position of the control points. Step S5: Based on the updated cutting plane, traverse the triangular facets of the three-dimensional surface model, determine the relative positional relationship between each triangular facet and the cutting plane, calculate the intersection points of the triangular facets located on both sides of the cutting plane with the cutting plane, and construct the closed cross-sectional profile on the cutting plane from the intersection points. Step S6: Based on the closed cross-sectional contour, perform a surface trimming Boolean operation on the three-dimensional surface model to obtain two independent three-dimensional models.

4. The method according to claim 3, characterized in that, Step S1 specifically includes: The three-dimensional surface model to be pre-cut is obtained from the model data source. The three-dimensional surface model is composed of multiple triangular facets and is stored in the model space in the form of three-dimensional coordinates. In the 3D view window, the 3D surface model is set to visible, and the viewing angle parameters, including viewpoint position, viewing direction and projection method, are determined according to the current user operation. Based on the observation perspective parameters, a two-dimensional display canvas corresponding to the three-dimensional view is established on the display terminal to receive the user's screen interaction input. There is a one-to-one projection mapping relationship between the two-dimensional display canvas and the three-dimensional model space.

5. The method according to claim 3, characterized in that, Step S2 specifically includes: The system receives manual drawing input from the user on the two-dimensional display canvas. Obtain a set of two-dimensional screen coordinate points arranged in chronological order, and construct a closed pre-tangent line based on the two-dimensional coordinate points; Based on the projection mapping relationship established in step S1, the inverse projection transformation calculation is performed on each two-dimensional coordinate point in the closed pre-tangent line to map it to the corresponding three-dimensional coordinate point in the three-dimensional model space, thereby forming a three-dimensional pre-tangent line located on the surface of the three-dimensional surface model or in its adjacent space.

6. The method according to claim 3, characterized in that, Step S3 specifically includes: Based on the spatial distribution characteristics of the three-dimensional pre-cutting lines and the viewing direction of the current three-dimensional view, a cutting plane for model segmentation is calculated. The cutting plane is determined in the following way: based on the spatial positions of at least some points in the three-dimensional pre-cutting lines and combined with the normal vector related to the viewing direction, a plane normal vector is calculated; at the same time, a point located on the three-dimensional pre-cutting lines is selected as a point passing through the plane, thereby uniquely determining the cutting plane. The cutting plane is visualized in a 3D view, and control points for interactive adjustment are generated on the cutting plane.

7. The method according to claim 3, characterized in that, Step S4 specifically includes: Receive user selection and drag operations on control points on the cutting plane, and convert the operations into changes in the position of the control points in three-dimensional space; Based on the changes in the control point positions, the spatial parameters of the cutting plane are updated in real time, including the direction of the plane's normal vector and the position of the points passing through the plane, so that the cutting plane undergoes corresponding displacement, rotation, or deformation in the three-dimensional model space. After the cutting plane parameters are updated, the system synchronously updates the display status of the cutting plane in the 3D view.

8. The method according to claim 3, characterized in that, Step S5 specifically includes: Based on the currently updated cutting plane, each triangular facet in the three-dimensional surface model is traversed and calculated one by one. For each triangular facet, determine the spatial positional relationship of its vertices relative to the cutting plane; when the vertices of the triangular facet are located on different sides of the cutting plane, calculate the intersection point of the cutting plane and the edge of the triangular facet. Connect all the intersections that meet the conditions according to their spatial adjacency to construct one or more closed cross-sectional profiles located on the cutting plane. These closed cross-sectional profiles are used to define the cutting boundaries of the model.

9. The method according to claim 3, characterized in that, Step S6 specifically includes: Based on the closed cross-sectional profile, perform a surface trimming Boolean operation on the three-dimensional surface model; Based on the cutting plane and the closed cross-sectional contour, the three-dimensional surface model is divided into two independent three-dimensional models along the cutting plane, and corresponding surface model data structures are generated for each part. After the segmentation is completed, the two parts of the 3D model are displayed independently in the 3D view to complete the pre-slicing operation.

10. A three-dimensional model pre-cutting system, characterized in that, The system includes: The model acquisition and display module is used to acquire the three-dimensional surface model to be pre-cut, establish a two-dimensional display canvas corresponding to the three-dimensional surface model in the three-dimensional view window according to the current viewing angle, and maintain the projection mapping relationship between the two-dimensional display canvas and the three-dimensional model space. The pre-cutting line acquisition and mapping module is used to receive closed pre-cutting lines drawn by the user on the two-dimensional display canvas, and to calculate the two-dimensional coordinates of the closed pre-cutting lines in reverse through the projection mapping relationship into three-dimensional pre-cutting lines in the three-dimensional surface model space. The cutting plane generation module is used to calculate at least one cutting plane based on the spatial distribution of the three-dimensional pre-cutting lines and the current viewing angle direction. The cutting plane is determined by the spatial normal vector and a point passing through the plane. The cutting plane interactive adjustment module is used to generate a set of control points on the cutting plane, receive user selection and drag operations of the control points, and update the spatial parameters of the cutting plane according to the changes in the position of the control points. The cross-sectional profile calculation module is used to traverse the triangular facets of the three-dimensional surface model based on the updated cutting plane, determine the relative positional relationship between each triangular facet and the cutting plane, calculate the intersection points of the triangular facets located on both sides of the cutting plane, and construct a closed cross-sectional profile located on the cutting plane from the intersection points. The surface trimming module is used to perform a surface trimming Boolean operation on the three-dimensional surface model based on the closed cross-sectional contour, dividing the three-dimensional surface model into two independent three-dimensional models.

Images