Digital simulation evaluation method for minimally invasive repair of tooth defects
Patent Information
- Application Number
- CN202610090972.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-23
- Publication Date
- 2026-10-27
AI Technical Summary
[0004]为了解决现有技术无法在设计阶段预判因材料特性和机床动力学限制引起的加工偏差问题,导致牙齿缺损的数字化模拟评估结果准确性低的技术问题,本发明的目的在于提供一种用于牙齿缺损微创修复的数字化模拟评估方法,所采用的技术方案具体如下:
本发明考虑到预备体制备过程中车针会受到材料特性和机床加工误差的影响,导致切削时车针实际路径和理论路径存在误差,使得对牙齿缺损修复的模拟评估准确性低。本发明首先滤除口内的无关噪声,得到静态牙体表面模型上的理论车针中心路径,能够在数学上等效呈现机床为兼顾加工安全与运动平滑性而产生的让刀与滤波轨迹,将抽象的轨迹偏差转化为临床指征;其次,通过每个路径点形态特征的复杂程度和边缘截面厚度的大小得到路径偏离代价,实现了车针在不同路径下移动时所需代价的准确量化,提高了后续获取的模拟路径的可靠性;然后,通过路径偏离代价对所有路径进行寻优,得到最优的模拟路径,显著提高了模拟路径的物理可解释性和准确性;最后,根据模拟路径和理论路径之间的差异程度,通过计算模拟车针中心路径相对于理论车针中心路径的径向偏移,能够在不进行实体试切的前提下,进行牙齿缺损微创修复的模拟评估,量化出由防崩裂机制和伺服跟随误差引起的微米级加工偏差,从而识别出几何上连续但物理不可加工的边缘形态,提升了评估结果的准确性。
Smart Images

Figure CN122991642A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of minimally invasive dental restoration technology, and more specifically to a digital simulation evaluation method for minimally invasive restoration of tooth defects. Background Technology
[0002] In the field of minimally invasive restoration of tooth defects, porcelain veneers and ultra-thin inlays are widely used due to their minimal material removal. This results in the marginal preparation of restorations often being at the sub-millimeter level, with complex and irregular curves in their shape. Currently, digital manufacturing processes generally utilize CNC machine tools to grind ceramic materials using rotating burs to manufacture such restorations. Design software only performs morphological design based on static geometric data, assuming that the processing equipment can perfectly reproduce any complex geometric contour, while ignoring the physical constraints in actual processing. These physical constraints manifest in two ways: firstly, the ceramic material cannot withstand the cutting stress of the bur at extremely thin edges, making it prone to brittle fracture; secondly, the feed axes of CNC machine tools have physical inertia, making it impossible to track sharp-angled paths with extremely high curvature with zero error at high speeds.
[0003] Existing technologies cannot predict processing deviations caused by material properties and machine tool dynamics limitations during the design phase, resulting in low accuracy of digital simulation assessments of tooth defects. Summary of the Invention
[0004] To address the problem that existing technologies cannot predict processing deviations caused by material properties and machine tool dynamics limitations during the design phase, resulting in low accuracy of digital simulation evaluation results for tooth defects, the present invention aims to provide a digital simulation evaluation method for minimally invasive tooth defect repair. The specific technical solution adopted is as follows: This invention proposes a digital simulation evaluation method for minimally invasive restoration of tooth defects, the method comprising: Obtain the theoretical bur center path on the static tooth surface model; The anatomical features of each path point on the theoretical bur center path are obtained, including the morphological features and edge section thickness of each path point; the path deviation cost of each path point is obtained based on the anatomical features; multiple simulated path points are obtained at each path point along the direction away from the tooth, and the distance difference between each simulated path point and the corresponding path point is obtained. The simulated path points between adjacent path points are matched point by point to obtain simulated path point pairs. The path transfer cost between simulated path point pairs is obtained based on the distance difference and path deviation cost. All paths are optimized based on the path transfer cost to obtain the local optimal path between adjacent path points. All local optimal paths form the simulated needle center path. The assessment of tooth defect restoration is based on the degree of difference between the simulated bur center path and the theoretical bur center path, as well as anatomical features.
[0005] Furthermore, the method for obtaining the theoretical bur center path on the static tooth surface model includes: The preparation body edge curve is obtained based on the geometric features of the tooth surface. The first tangential vector at each point on the preparation body edge curve is obtained. The normal direction on the plane where the first tangential vector and the optimal placement path vector are located is obtained. Each point on the preparation body edge curve is translated along the normal direction to obtain the theoretical bur center path. The optimal placement path vector is the spindle direction during CNC machining.
[0006] Furthermore, the method for obtaining the morphological features and edge cross-sectional thickness of each path point includes: In the path points, obtain the position coordinates of the path point and the two adjacent path points, and use the discrete curvature formula to calculate the curvature of the path point as a morphological feature. At the path point, a ray is emitted toward the preparation body along the reverse extension direction away from the tooth body, and the distance the ray penetrates the preparation body is taken as the edge section thickness.
[0007] Furthermore, the method for obtaining the path deviation cost of each path point based on the anatomical features includes: The morphological similarity between the morphological features of each path point and the most prominent morphology is obtained, as well as the thickness comparison features between the edge section thickness of each path point and the thickness threshold. Based on the morphological and thickness comparison features, the path deviation cost is obtained by weighted fusion of acceleration-sensitive weight and cutting force load coefficient. The thickness threshold, acceleration-sensitive weight, and cutting force load coefficient are obtained by experimental calibration.
[0008] Furthermore, the direction away from the tooth body includes: At each path point, calculate the second tangential vector of the path point, and obtain the cross-sectional circle of the path point based on the second tangential vector and the position coordinates of the path point; obtain the corresponding point of the path point on the edge curve of the preparation body, calculate the Euclidean distance between each point on the cross-sectional circle and the corresponding point, and take the direction of the path point pointing to the point on the cross-sectional circle corresponding to the maximum value of the Euclidean distance as the direction away from the tooth body.
[0009] Furthermore, the method for obtaining multiple simulated path points includes: Multiple sampling points are obtained by sampling along the straight line corresponding to each path point in the direction away from the tooth, and these multiple sampling points are used as simulated path points.
[0010] Furthermore, the method for obtaining the distance difference between each simulated path point and its corresponding path point includes: For each path point and its corresponding simulated path point, calculate the Euclidean distance between the simulated path point and the corresponding path point, and use this distance difference as the distance difference.
[0011] Furthermore, the method of performing point-by-point matching of simulated path points between adjacent path points to obtain simulated path point pairs, and obtaining the path transition cost between simulated path point pairs based on distance differences and path deviation costs includes: In two adjacent path points, the simulated path point corresponding to the current step path point is matched with the simulated path point corresponding to the next step path point point one by one to obtain multiple simulated path point pairs; Between two adjacent path points, calculate the squared difference of the distance difference between the current path point and the next path point. Multiply the squared difference by the path deviation cost of the current path point as the motion potential energy term. Calculate the squared distance difference of the current path point as the geometric potential energy term. Sum the geometric potential energy term and the motion potential energy term as the path transfer cost between the simulated path point pairs.
[0012] Furthermore, the method for optimizing all paths to obtain locally optimal paths between adjacent path points includes: Select the starting point of the path from all path points. For each simulated path point pair between the path point and the starting point, traverse all path transition costs on the complete path formed by connecting the simulated path point pairs end to end from the current step path point to the starting point. The simulated path point pair contained in the path with the smallest sum of path transition costs is taken as the local optimal path. The local optimal path satisfies the closure constraint.
[0013] Furthermore, the method for assessing tooth defect restoration based on the degree of difference between the simulated bur center path and the theoretical bur center path, as well as anatomical features, includes: The difference in positional coordinates between the simulated path point on the simulated needle center path and the path point on the theoretical needle center path is calculated. Based on the magnitude of the difference in positional coordinates, the size of the machining gap is determined. Based on the difference between the anatomical features and the preset thresholds in the corresponding dimensions, the causes of the machining gap are analyzed. The size of the machining gap and the causes of the machining gap are used as the evaluation results.
[0014] The present invention has the following beneficial effects: This invention takes into account that the bur is affected by material properties and machine tool processing errors during the preparation process, which leads to errors between the actual path and the theoretical path of the bur during cutting, resulting in low accuracy of simulation evaluation for tooth defect repair. This invention first filters out irrelevant noise within the mouth to obtain the theoretical bur center path on a static tooth surface model. This mathematically represents the deflection and filtering trajectories generated by the machine tool to balance machining safety and smooth movement, transforming abstract trajectory deviations into clinical indicators. Second, the path deviation cost is obtained by analyzing the complexity of the morphological features of each path point and the thickness of the edge cross-section. This accurately quantifies the cost required for the bur to move along different paths, improving the reliability of the subsequently obtained simulated paths. Then, the optimal simulated path is obtained by optimizing all paths using the path deviation cost, significantly improving the physical interpretability and accuracy of the simulated path. Finally, based on the degree of difference between the simulated and theoretical paths, the radial offset of the simulated bur center path relative to the theoretical bur center path is calculated. This allows for the simulated evaluation of minimally invasive tooth defect repair without physical trial cutting, quantifying the micron-level machining deviations caused by anti-fracture mechanisms and servo following errors. This identifies geometrically continuous but physically unmachinable edge morphologies, improving the accuracy of the evaluation results. Attached Figure Description
[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 This is a flowchart of a digital simulation evaluation method for minimally invasive repair of tooth defects, provided as an embodiment of the present invention. Detailed Implementation
[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a digital simulation evaluation method for minimally invasive repair of tooth defects proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for a digital simulation evaluation method for minimally invasive repair of tooth defects provided by the present invention.
[0020] Please see Figure 1 The diagram illustrates a flowchart of a digital simulation evaluation method for minimally invasive repair of tooth defects according to an embodiment of the present invention. The method includes: Step S101: Obtain the theoretical bur center path on the static tooth surface model.
[0021] In reality, scanning teeth inside a patient's mouth is affected by factors such as soft tissue peristalsis, saliva flow, and slight vibrations of the operator's handheld device. The original tooth model data appears as a dynamic point cloud with timestamps. Directly using single-frame data for micron-level edge fitting introduces random errors. Therefore, it is necessary to reconstruct the dynamic data in the temporal domain into a unique static tooth surface model in the spatial domain. Obtaining the theoretical bur center path based on the static tooth surface model can filter out dynamic interference and improve the reliability of the obtained theoretical bur center path.
[0022] The theoretical bur center path is the path the bur follows when cutting the restoration. Minimally invasive restorations of tooth defects require cutting ceramic restorations to create a shape that fits the patient's tooth surface. The restoration is then bonded tightly to the tooth using adhesive, thus restoring the tooth defect. However, CNC machine tools have machining deviations. Generating the theoretical bur center path directly on the tooth surface model can cause the bur center to enter the tooth, resulting in a poor fit of the restoration. Therefore, the cutting path cannot be generated based on the tooth surface. Instead, it needs to be derived by extending a certain distance outward from the static tooth surface model to obtain the theoretical bur center path. This theoretical bur center path represents the zero-error trajectory the bur center should follow under ideal physical conditions, assuming no machine tool inertia, no material brittleness, and no tool deformation. This lays the foundation for subsequent analysis of material properties and machining deviations caused by machine tool dynamics limitations.
[0023] As an example, in a specific implementation of this invention, the method for obtaining a static tooth surface model is as follows: First, the system connects to the real-time data stream of an intraoral scanner, and sets a probe gaze speed threshold. and a sampling time window The system calculates the moving speed of the scanning probe relative to the scanned object in real time. When the moving speed is lower than the probe's gaze speed threshold, it determines that the scanning is in a gaze state on edge features and triggers local multi-frame acquisition to obtain a point cloud frame sequence within a continuous time period. ,in The number of frames is determined; when the movement speed is lower than the probe gaze speed threshold, the scanning probe is determined to be in high-speed movement mode, and multi-frame acquisition is not triggered. Next, the system selects the first frame from the sequence. A local reference coordinate system is established, and the Rigid Iterative Closest Point (Rigid-ICP) algorithm is used to integrate the remaining frames. Registered to this local reference coordinate system; construct a high-resolution voxel mesh that encloses the scanned region ( ), map all registered point clouds to this mesh; for each voxel in the mesh The system counts the set of point clouds that fall within the voxel. And calculate the spatial location covariance matrix of the set. : ,in, Let be the geometric center of the point cloud within this voxel. This represents the number of point clouds. Then, for... Eigenvalue decomposition yields three eigenvalues. Calculate the location variance characteristics Because soft tissue and saliva exhibit significant displacement fluctuations over time, while the tooth surface remains relatively stationary, their positional variance characteristics differ significantly. Therefore, the system sets a rigidity stability threshold of [value missing]. If voxels If the value exceeds this threshold, it is considered non-rigid noise and is removed; if it is less than or equal to the threshold, the voxel is retained. Finally, the retained steady-state voxel set is meshed using the Poisson Surface Reconstruction algorithm to generate a time-invariant triangular mesh surface as the static tooth surface model. It should be noted that the rigid iterative nearest-point algorithm and the Poisson surface reconstruction algorithm are well-known techniques in the art and will not be elaborated upon or limited here.
[0024] Step S102: Obtain the anatomical features of each path point on the theoretical bur center path, the anatomical features including the morphological features and edge section thickness of each path point; obtain the path deviation cost of each path point based on the anatomical features; obtain multiple simulated path points along the direction away from the tooth at each path point, and obtain the distance difference between each simulated path point and the corresponding path point.
[0025] To consider both the material properties of the preform and the deviations during machine tool processing when conducting simulation evaluations, both need to be quantified. The edge section thickness represents the structural strength of the preform against cutting forces, while the morphological characteristics are positively correlated with the deviations during machine tool processing. That is, the more complex the shape and the more complex the cutting path, the greater the deviation during machine tool processing. Through these two quantified anatomical features, the material properties and the magnitude of machine tool processing deviations can be intuitively displayed, improving the accuracy of subsequent calculations.
[0026] Since the theoretical needle center path obtained in step S101 represents the zero-error trajectory that the needle center should run under ideal physical conditions assuming no inertia of the machine tool, no brittleness of the material, and no deformation of the tool, the actual needle center will not move along the theoretical path when considering the influence of two factors: the material properties of the pre-body and the deviation during machine tool processing. Therefore, it is necessary to obtain a simulated path that deviates from the theoretical path.
[0027] Each path point has different anatomical features, and these different features affect the cost that the bur needs to overcome to deviate from the theoretical path and move towards the simulated path. On the one hand, the more complex the morphological features, the more work is required for the bur to move, and the greater the cost of deviating from the simulated path. On the other hand, the thinner the edge section thickness, the more the bur needs to deviate from the theoretical path to ensure that the preform does not break, and the greater the cost. Therefore, the anatomical features can represent the magnitude of the cost required for the bur to deviate at different path points, providing a useful data basis for obtaining the simulated path.
[0028] In the small-scale scenario of cutting tooth preparations, even if deviations exist, the simulated path will not deviate significantly from the theoretical path. Furthermore, the key factor affecting the pass rate of preparations lies in whether the preparation surface is protruding or concave, i.e., whether the bur cuts too little or too much. Therefore, the deviation between the simulated path and the theoretical path is concentrated on the degree of proximity to the tooth surface. Moreover, due to the interference of real-world factors in the simulated path, in order to leave a larger safety margin during cutting, the simulated path will be further away from the tooth surface than the theoretical path. Therefore, the simulated path point needs to be set in a direction away from the tooth to improve the accuracy and reliability of the simulation evaluation.
[0029] On the one hand, the cost of the bur deviating at different path points is different. On the other hand, after deviation, since there are multiple simulated path points in the direction away from the tooth, it is necessary to consider which simulated path point the bur actually reaches. Since the distances between multiple simulated path points and their corresponding path points are different, more energy is required to reach a more distant simulated path point. Therefore, the distance between simulated path points and the required energy are positively correlated. Thus, the cost of the bur reaching the simulated path point after deviating from the path point can be reflected by the difference in distance between each simulated path point and its corresponding path point.
[0030] Step S103: Match the simulated path points between adjacent path points one by one to obtain simulated path point pairs. Obtain the path transfer cost between simulated path point pairs based on the distance difference and path deviation cost. Optimize all paths based on the path transfer cost to obtain the local optimal path between adjacent path points. All local optimal paths form the simulated needle center path.
[0031] In this embodiment of the invention, it is considered that the actual cutting path of the sewing needle will not follow the theoretical path, but will move along the simulated path between adjacent path points. Therefore, it is necessary to match the simulated path points between adjacent path points one by one to obtain simulated path point pairs. The simulated path point pairs are the actual local movement path of the sewing needle, which improves the accuracy of the simulation.
[0032] The distance difference and path deviation cost obtained in step S102 represent the cost of the needle reaching different simulated path points and the cost required for the needle to deviate from the path point, respectively. The path transfer cost between simulated path point pairs includes both the cost of the needle moving between the point pairs and the cost required for the needle to initially deviate. Therefore, in order to comprehensively evaluate the combined cost of reaching another simulated path point from one simulated path point, the path transfer cost can be obtained through the distance difference and path deviation cost, thereby achieving adaptive adjustment of the path cost and significantly improving the interpretability and reliability of obtaining the cost between simulated path points.
[0033] During the cutting process, in order to avoid unnecessary energy loss, it is necessary to ensure that the energy of the path followed by the needle movement is minimized. The path transfer cost reflects the energy required for the needle to move under different local paths. The greater the path transfer cost, the greater the energy required. Therefore, in order to obtain the local optimal path with the minimum energy loss, it is necessary to optimize all paths based on the path transfer cost to obtain the local optimal path. This can significantly reduce unnecessary cutting loads and improve the physical feasibility and machining stability of the simulated path.
[0034] Step S104: Based on the degree of difference between the simulated bur center path and the theoretical bur center path, as well as anatomical features, conduct an assessment of tooth defect restoration.
[0035] The core contradiction in minimally invasive restoration lies in the conflict between the extremely thin edge preparation amount and the physical limits of CNC machining. The essence of digital evaluation should be to explicitly project this physical constraint during the design phase. The degree of difference between the simulated bur center path and the theoretical bur center path reflects the deviation between the actual cutting path and the ideal design path. The greater the deviation, the greater the conflict, and the more likely the prepared body is to be unqualified. Conversely, the smaller the deviation, the smaller the conflict, and the more likely the prepared body is to be qualified. The anatomical features include the morphological features and edge section thickness of each path point. The morphological features quantify how the bur moves during cutting, while the edge section thickness quantifies how deep the cut can be. Together, they determine the safety boundary and processing difficulty of each cutting action during the cutting process. Therefore, the degree of difference and anatomical features are highly correlated with the evaluation of tooth defect restoration, enabling a comprehensive and accurate evaluation of minimally invasive tooth defect restoration.
[0036] In summary: First, this invention filters out irrelevant noise within the mouth to obtain the theoretical bur center path on a static tooth surface model. This mathematically represents the deflection and filtering trajectories generated by the machine tool to balance machining safety and smooth movement, transforming abstract trajectory deviations into clinical indicators. Second, the path deviation cost is obtained by considering the complexity of the morphological features of each path point and the thickness of the edge cross-section, accurately quantifying the cost required for the bur to move under different paths and improving the reliability of the subsequently obtained simulation paths. Then, the optimal simulation path is obtained by optimizing all paths using the path deviation cost, significantly improving the physical interpretability and accuracy of the simulation path. Finally, based on the degree of difference between the simulation path and the theoretical path, the radial offset of the simulated bur center path relative to the theoretical bur center path is calculated. This allows for the simulation evaluation of minimally invasive tooth defect repair without physical trial cutting, quantifying the micron-level machining deviations caused by anti-fracture mechanisms and servo following errors, thereby identifying geometrically continuous but physically unmachinable edge morphologies and improving the accuracy of the evaluation results.
[0037] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the theoretical bur center path on a static tooth surface model includes: The preparation body edge curve is obtained based on the geometric features of the tooth surface. The first tangential vector at each point on the preparation body edge curve is obtained. The normal direction on the plane where the first tangential vector and the optimal placement path vector are located is obtained. Each point on the preparation body edge curve is translated along the normal direction to obtain the theoretical bur center path. The optimal placement path vector is the spindle direction during CNC machining.
[0038] Normal, healthy teeth have smooth geometric features. When defects occur on the tooth surface, the originally smooth geometric features undergo significant abrupt changes, becoming sharp and prominent. These sharp and prominent parts are the defect areas that doctors need to repair for patients using prepared materials. Therefore, it is necessary to obtain the theoretical bur center path based on the areas with distinct geometric features on the static tooth surface model, which can meet the reconstruction requirements of the defect area and ensure a smooth transition during the cutting process within the biosafety range.
[0039] The cutting path of the bur is not made close to the tooth surface, but rather on the preparation body. Therefore, the theoretical bur center path needs to be shifted outward by a certain distance based on the edge curve of the preparation body to obtain the theoretical bur center path.
[0040] In one specific implementation of this invention, the first tangential vector is obtained by curve fitting using the least squares method with a point on the edge curve of the preparatory body and two points before and after it (a total of 5 points). Two tangential vectors on the fitted curve are calculated, and the tangential vector pointing in the direction of needle movement is taken as the first tangential vector.
[0041] The optimal path of placement refers to the common path of placement chosen in denture restoration to ensure optimal fixation and stability of the prepared denture. This is the spindle feed direction during CNC machining, which is usually parallel to the long axis of the tooth. During the preparation machining process, the cutting surface of the bur needs to be parallel to or at a preset angle to the axial wall of the prepared denture. This requires that the bur cannot be translated additionally along the path of placement, otherwise it will lead to overcutting or undercutting during the preparation machining. Therefore, the translation direction needs to be perpendicular to the optimal path of placement vector. In order to ensure that the theoretical bur center path after translation maintains a constant distance from the original edge curve of the prepared denture locally, that is, equidistant translation, the translation direction needs to be perpendicular to the tangent direction at each point. Therefore, the translation direction is the normal direction on the plane where the first tangential vector and the optimal path of placement vector are located.
[0042] In one specific implementation of this invention, the method for obtaining the theoretical bur center path is as follows: First, a curvature extremum tracking algorithm (e.g., a ridge extraction method based on discrete curvature tensor) is run on a static tooth surface model to identify the closed feature line with the largest rate of curvature change, which is used as the edge curve of the preparation body; second, the system receives a preset optimal placement path vector, denoted as... This vector, specified by the user, represents the spindle direction during CNC machining; the edge curve of the preparatory body is discretized and sampled at intervals of [missing information]. This yields an ordered sequence of points, for each sampling point on the curve. Calculate the first tangent vector at that point, denoted as . Then, according to and The formula for calculating the translation direction is: , Indicates sampling point The direction of translation, This indicates calculating the magnitude of a vector; it should be noted that if the angle between the optimal positioning vector and the first tangent vector is less than 5 degrees, the translation direction of the previous point is used as the translation direction of the current point; finally, each sampling point... Along the corresponding translation direction Translate by a distance, which is The edge curve of the translated preparatory body is obtained, which is the theoretical needle center path.
[0043] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the morphological features and edge cross-sectional thickness of each path point includes: In the path points, obtain the position coordinates of the path point and the two adjacent path points, and use the discrete curvature formula to calculate the curvature of the path point as a morphological feature. At the path point, a ray is emitted toward the preparation body along the reverse extension direction away from the tooth body, and the distance the ray penetrates the preparation body is taken as the edge section thickness.
[0044] Since the dynamic response characteristics of a machine tool at different locations depend on local geometry and cutting conditions, this embodiment of the invention considers the morphological features and edge section thickness of each path point. Firstly, the morphological features of the path are reflected in the magnitude of local curvature at different path points. The greater the local curvature, the greater the degree of bending at the path point; conversely, the smaller the local curvature, the less the degree of bending at the path point. Therefore, local curvature is used to represent the morphological features. Cutting conditions are reflected in the requirement that the pre-worked object must not undergo brittle fracture during cutting, which means that the edge section thickness of the pre-worked object during cutting needs to be considered.
[0045] In one specific implementation of this invention, the formula for calculating local curvature is: ,in, Indicates the first [number]th ... The morphological features at each point , and These represent the first and second needles on the theoretical needle center path. The point, the first The point and the first One point, and These represent finding the magnitude of a vector and the cube of its magnitude, respectively. The above formulas are the discrete curvature formulas, a technique well-known to those skilled in the art; their specific meanings will not be elaborated upon here.
[0046] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the path deviation cost of each path point based on the anatomical features includes: The morphological similarity between the morphological features of each path point and the most prominent morphology is obtained, as well as the thickness comparison features between the edge section thickness of each path point and the thickness threshold. Based on the morphological similarity and thickness comparison features, the path deviation cost is obtained by weighted fusion with acceleration-sensitive weight and cutting force load coefficient. The thickness threshold, acceleration-sensitive weight, and cutting force load coefficient are obtained by experimental calibration.
[0047] By comparing the morphological features of each path point with the most prominent morphological feature, similarity calculations can map the morphological features of all path points to a unified relative interval [0, 1]. Higher morphological similarity indicates that the path point's morphological features are closer to the most prominent morphology, resulting in higher processing risk and cost; therefore, it should be given higher weight in the path deviation cost. By comparing the edge section thickness of each path point with a thickness threshold, a relative measure of material fragility can be established. By calculating the thickness contrast feature, this relative measure can be mapped to the path deviation cost in a normalized manner. The thickness contrast feature represents the closeness between the edge section thickness and the thickness threshold. The closer the edge section thickness is to the thickness threshold, the greater the processing risk, i.e., the more likely fracture will occur, resulting in higher cost; therefore, it should be given higher weight in the path deviation cost. The acceleration sensitivity weight and cutting force load coefficient represent the machine tool's dynamic following error tendency during high-speed reversal and the average rate of force change under different cutting angles, respectively, i.e., the physical amplification factor of the machine tool for morphological similarity and thickness contrast features under different morphological paths. In a specific implementation of this invention, the formula for calculating the path deviation cost is: ; in: The first digit on the theoretical needle center path Path deviation cost at each point; : Basic servo rigidity constant; Acceleration sensitivity weight; Cutting force load coefficient; The first digit on the theoretical needle center path Morphological features at each point; : The maximum morphological value at each point along the theoretical needle center path; The first digit on the theoretical needle center path Edge section thickness at each point; Thickness threshold; : An exponential function with the natural constant as its base.
[0048] Basic servo stiffness constant This represents the machine tool servo system's ability to resist path deviation. Its function is to amplify the overall path deviation cost benchmark, that is, to convert the path deviation cost into an absolute value. This allows the final path deviation cost to reflect the differences in machining stability under different servo stiffnesses, thereby improving the adaptability of path planning to the dynamic characteristics of the machine tool. The positive integer 1 represents the normalized relative risk coefficient, limiting the value range to... This indicates the additional costs associated with morphological features and material risks; This term represents morphological similarity. Since the numerator cannot be greater than the denominator, the larger the value, the higher the similarity. However, because tooth defects in reality have complex structures, the bur will not make just one cut, and the cutting path will not be a standard straight line; therefore, there is no denominator. The case where it is 0, while the acceleration-sensitive weight This reflects the tendency of dynamic following error under different morphological characteristics. It has a high correlation with morphological characteristics and adopts and The product can increase the cost in high curvature regions, simulating tracking errors; This term performs a negative correlation mapping on the thickness contrast feature. Since the smaller the edge section thickness relative to the thickness threshold, the higher the probability of pre-body fracture, and the smaller the ratio, the greater the cost. The thickness contrast feature and path deviation cost are negatively correlated. By taking the negative sign, the negative correlation is mapped to a positive correlation. The function makes the edge section thickness When the thickness approaches or falls below the threshold, the path deviation cost increases exponentially, designed to simulate the conservative machining strategy employed by machine tools to prevent edge chipping; cutting force load coefficient It represents the average rate of change of force under different cutting angles of attack, which can amplify the risk of cutting instability caused by insufficient material thickness and avoid the problems of being too conservative or too aggressive caused by the rigid constraint of a single thickness threshold (i.e., thickness threshold).
[0049] In another specific implementation of this invention, the CNC machine tool employs a three-axis or five-axis linkage cutting machine, and the ceramic preform material is lithium disilicate glass ceramic. The basic servo rigidity constant, thickness threshold, acceleration sensitivity weight, and cutting force load coefficient are obtained through standard test block cutting experiments. The thickness threshold can be set as follows: (Minimum processing thickness of lithium disilicate glass ceramic); the acceleration sensitivity weight setting range is [0.5, 2], set to 0.75; the basic servo rigidity constant is set to 2; the cutting force load coefficient setting range is [0.1, 0.5], set to 0.3.
[0050] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the direction away from the tooth includes: At each path point, calculate the second tangential vector of the path point, and obtain the cross-sectional circle of the path point based on the second tangential vector and the position coordinates of the path point; obtain the corresponding point of the path point on the edge curve of the preparation body, calculate the Euclidean distance between each point on the cross-sectional circle and the corresponding point, and take the direction of the path point pointing to the point on the cross-sectional circle corresponding to the maximum value of the Euclidean distance as the direction away from the tooth body.
[0051] To ensure that the unnecessary impact of the bur on the preparation body due to material properties and machine tool machining deviations during the cutting process is minimized, that is, to provide the maximum safety buffer space for machining errors, material deformation and machine tool vibration, it is necessary to obtain the direction that is furthest away from the tooth body. The point on the cross-sectional circle corresponding to the maximum Euclidean distance reflects the farthest point on the cross-sectional circle from the corresponding point on the edge of the preparation body. Using the direction from the path point to this point as the direction away from the tooth body can maximize the expansion of redundant space in the direction away from the tooth body, thereby providing the maximum safety margin for the cutting process when there are material removal and machine tool tracking errors.
[0052] In one specific implementation of this invention, the second tangential vector is obtained by curve fitting using the least squares method with the previous path point and two path points before and after (a total of 5 points) in the theoretical needle center path. Two tangential vectors on the fitted curve are calculated, and the tangential vector pointing in the needle movement direction is taken as the second tangential vector. With the path point position as the center of the cross-sectional circle, the second tangential vector is the normal vector of the cross-sectional circle, and the radius is... A cross-sectional circle is obtained by fitting the least squares method.
[0053] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining multiple simulated path points includes: Multiple sampling points are obtained by sampling along the straight line corresponding to each path point in the direction away from the tooth, and these multiple sampling points are used as simulated path points.
[0054] In actual cutting, the bur exhibits two types of errors: undercut and overcut. These errors manifest as the bur center being on a ray originating from the theoretical path point and moving away from the tooth, and on a ray originating from the theoretical path point and moving closer to the tooth. Therefore, the simulated path point lies on the straight line in the direction away from the tooth. In a specific implementation of this invention, the sampling method is equidistant sampling, with a sampling interval of [missing information]. The sampling range is from the theoretical path point in the direction away from the tooth to... From there, samples were collected towards the tooth structure. Place.
[0055] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the distance difference between each simulated path point and its corresponding path point includes: For each path point and its corresponding simulated path point, calculate the Euclidean distance between the simulated path point and the corresponding path point, and use this distance difference as the distance difference.
[0056] The distance between each simulated path point is different. Euclidean distance is used as the distance difference, which can represent the cost of the needle deviating from the theoretical position. The farther the distance, the greater the cost, and the closer the distance, the smaller the cost.
[0057] Preferably, in some possible implementations of the embodiments of the present invention, the method of performing point-by-point matching of simulated path points between adjacent path points to obtain simulated path point pairs, and obtaining the path transfer cost between simulated path point pairs based on distance differences and path deviation costs includes: In two adjacent path points, the simulated path point corresponding to the current step path point is matched with the simulated path point corresponding to the next step path point point one by one to obtain multiple simulated path point pairs; Between two adjacent path points, calculate the squared difference between the distance difference between the current step path point and the next step path point, and multiply the squared difference by the path deviation cost of the current step path point as the motion potential energy term; Calculate the square of the distance difference between the current step path points as the geometric potential energy term; use the sum of the geometric potential energy term and the motion potential energy term as the path transfer cost between simulated path point pairs.
[0058] Because the needle is affected by material properties and machine tool processing errors, the needle moves on simulated path points between two adjacent path points. Therefore, it is necessary to match the simulated path points point by point to obtain multiple simulated path point pairs.
[0059] In one specific implementation of this invention, the formula for calculating the path transition cost between single-step simulated path point pairs is as follows: ; in: : The path transition cost between path point pairs in a single step simulation; : The distance difference between the k-th path point and the j-th simulated path point; : The distance difference between the (k-1)th path point and the mth simulated path point; : Path deviation cost of the (k-1)th path point.
[0060] The geometric potential energy term represents the penalty for the needle deviating from its theoretically designed position; squaring it amplifies this penalty. The kinetic potential energy term represents the penalty for a radial abrupt change in the path, where... and These represent the distance difference between the simulated path point pair and the distance difference between the simulated path point pair and the previous simulated path point, respectively. The difference between the two distance differences reflects the degree of radial offset between the simulated path point pairs. The larger the difference, the greater the radial offset, and the greater the cost of the needle moving between the simulated path point pairs. The purpose of squaring is to amplify this cost. This represents the path deviation cost of the path point corresponding to the previous simulated path point in the simulated path point pair. This is because the movement of the needle is directional; that is, the needle moves from the simulated path point corresponding to the (k-1)th path point to the simulated path point corresponding to the kth path point. Using the path deviation cost of the previous path point (i.e., the (k-1)th path point) allows the motion potential energy term to be combined with the current starting point of motion. and The product introduces a cost for the needle to detach from the simulated path point, on top of the cost of the needle moving between simulated path point pairs. This further improves the physical integrity of the path transfer cost. Moreover, the existence of the path deviation cost makes the path more inclined to remain smooth in areas with complex features (such as acute angles or thin walls), thereby reducing unnecessary manufacturing deviations.
[0061] Preferably, in some possible implementations of the embodiments of the present invention, the method for optimizing all paths to obtain the locally optimal path between adjacent path points includes: Select the starting point of the path from all path points. For each simulated path point pair between the path point and the starting point, traverse all path transition costs on the complete path formed by connecting the simulated path point pairs end to end from the current step path point to the starting point. The simulated path point pair contained in the path with the smallest sum of path transition costs is taken as the local optimal path. The local optimal path satisfies the closure constraint.
[0062] Since the cutting process is continuous, the current needle position strongly depends on the previous position and motion state. To obtain a true path that conforms to physical inertia, this embodiment of the invention uses dynamic programming (Viterbi algorithm) to solve for the trajectory with the minimum total energy in the discrete state space. Because the local optimal path needs to satisfy the closure constraint of the true path, in the Viterbi algorithm, the simulated path point index of the path start point and the simulated path point index of the path end point coincide or are adjacent in space. The solution process steps are as follows: First, since the path is closed, a random position is selected as the reference for the starting point. (L represents the number of simulated path points corresponding to a path point), definition ,in, This represents the cumulative energy of the j-th simulated path point of the first path point. This represents the distance difference between the j-th simulated path point and the first path point. Then, for each subsequent path point... Simulated path point corresponding to path point k Iterate through all simulated path points from the starting point to the current path point k. Find the predecessor state index that minimizes the accumulated energy. : ; in, This represents the cumulative energy of the j-th simulated path point at the k-th path point. This represents the cumulative energy of the m-th simulated path point at the (k-1)-th path point. Let $\frac{m}{j}$ represent the path transition cost between the m-th simulated path point of the (k-1)-th path point and the j-th simulated path point of the k-th path point. This represents the function that iterates through all m simulated path points to find the minimum value within {}. Simultaneously, it utilizes a backtracking pointer matrix. Record path source: ; in, This represents the index of the best predecessor state of the j-th simulated path point of the k-th path point, i.e., in the optimal path from the starting point to the k-th path point, in the j-th path point... The simulated path point at which a given path point is located. This represents the simulated path point that minimizes the expression within the curly braces {}. Finally, when the recursion reaches the endpoint... N represents the number of path points in the total energy. Find the index with the minimum energy. Subsequently, using By backtracking the matrix, the complete optimal state index sequence is obtained. The set of simulated path points corresponding to the optimal index sequence is taken as the simulated needle center path; the simulated needle center path includes all simulated manufacturing deviations caused by material properties and machine tool dynamics limitations.
[0063] Preferably, in some possible implementations of the embodiments of the present invention, the method for evaluating tooth defect restoration based on the degree of difference between the simulated bur center path and the theoretical bur center path, as well as anatomical features, includes: The difference in positional coordinates between the simulated path point on the simulated needle center path and the path point on the theoretical needle center path is calculated. Based on the magnitude of the difference in positional coordinates, the size of the machining gap is determined. Based on the difference between the anatomical features and the preset thresholds in the corresponding dimensions, the causes of the machining gap are analyzed. The size of the machining gap and the causes of the machining gap are used as the evaluation results.
[0064] The core contradiction in minimally invasive repair lies in the conflict between the extremely thin edge preparation amount and the physical limits of CNC machining. The essence of digital assessment should be to explicitly project this physical constraint in the design stage. Therefore, by performing reverse analysis on the energy-minimizing path (i.e., simulating the center path of the bur), and by reusing the state variables in the path optimization process, the machining gap can be directly quantified and its causes diagnosed, which can guide doctors to make targeted geometric adjustments in physically unmachinable areas.
[0065] The simulated burr center path is the center trajectory after considering tool radius compensation. Its offset relative to the theoretical burr center path directly reflects the positional error of the cutting surface. Therefore, defining the gap directly using the distance difference between simulated path points on the simulated burr center path can avoid complex solid envelope surface calculations.
[0066] In one specific implementation of this invention, the distance difference between the k-th simulated path point and its corresponding path point on the simulated needle center path is defined as the simulated edge machining gap. : when When the simulated bur center is located outside the theoretical path, it means that in order to meet the machine tool dynamics constraints or prevent material breakage, the bur performs a "deflection" operation, resulting in material that should have been removed remaining at the edge of the prepared body. This residual material will form physical interference when the restoration is inserted, causing the restoration to be lifted, thus creating a gap that cannot be closed at the edge.
[0067] when When the bur is in contact with or slightly overcuts the theoretical path, this is generally acceptable in clinical practice because the adhesive can fill the small overcut space without hindering placement.
[0068] To enable physicians to understand how to correct pre-existing structures, it is necessary to clearly identify the physical causes of gaps and set a clinically tolerable gap threshold. For all For anomalies exceeding this threshold, the extracted anatomical features are used for attribution determination, establishing a mapping logic from physical features to clinical procedures. In this embodiment of the invention, attribution determination includes material brittleness protection attribution and kinematic filtering attribution: First, to attribute the cause to the material's brittleness protection, the system checks the edge section thickness at the anomaly point. If the edge section thickness is less than a preset thickness threshold, the system determines that the gap at that point is caused by the machine tool's anti-splitting protection mechanism. This means that the edge at that point is prepared too thin, and the machine tool control system forcibly limits the feed depth of the bur to prevent the ceramic from shattering during cutting. At this point, the system generates a "increase thickness" correction command.
[0069] Then, for kinematic filtering attribution, if the edge section thickness is within a safe range, the system checks the morphological features at the abnormal point. If the radius of the corner corresponding to the morphological feature is smaller than the effective cutting radius of the machine tool's kerf or exceeds the dynamic response bandwidth of the machine tool's axis, the system determines that the backlash at that point is caused by the kinematic filtering effect of the servo system. This means that the edge corner at that point is too sharp, and the machine tool's physical axis cannot track the rapidly changing path with zero error during high-speed movement, resulting in a follow-through error. At this time, the system generates a "rounded corner" correction command.
[0070] After quantitative and attribution analysis, the quantitative data and attribution results are overlaid on a static tooth surface model using augmented reality, intuitively displaying the location of tooth defects and suggested repairs. The system uses color coding (green for acceptable, red for out-of-tolerance) to differentiate between these defects. The numerical values are rendered on the edge area of the prepared volume. For the red areas, the system automatically attaches text labels, displaying specific correction suggestions (e.g., "Insufficient thickness, needs to be increased"). (Or "The corner is too sharp and needs to be rounded"). Following the on-screen prompts, the doctor uses the mobile device directly inside the patient's mouth to fine-tune the prepared body. After the adjustment is completed, the system triggers a new round of evaluation until the simulated gaps in all edge areas are below the clinically tolerable gap threshold. This eliminates potential manufacturing defects before physical processing occurs.
[0071] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0072] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A digital simulation evaluation method for minimally invasive restoration of tooth defects, characterized in that, The method includes: Obtain the theoretical bur center path on the static tooth surface model; The anatomical features of each path point on the theoretical bur center path are obtained, including the morphological features and edge section thickness of each path point; the path deviation cost of each path point is obtained based on the anatomical features; multiple simulated path points are obtained at each path point along the direction away from the tooth, and the distance difference between each simulated path point and the corresponding path point is obtained. The simulated path points between adjacent path points are matched point by point to obtain simulated path point pairs. The path transfer cost between simulated path point pairs is obtained based on the distance difference and path deviation cost. All paths are optimized based on the path transfer cost to obtain the local optimal path between adjacent path points. All local optimal paths form the simulated needle center path. The assessment of tooth defect restoration is based on the degree of difference between the simulated bur center path and the theoretical bur center path, as well as anatomical features.
2. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method for obtaining the theoretical bur center path on a static tooth surface model includes: The preparation body edge curve is obtained based on the geometric features of the tooth surface. The first tangential vector at each point on the preparation body edge curve is obtained. The normal direction on the plane where the first tangential vector and the optimal placement path vector are located is obtained. Each point on the preparation body edge curve is translated along the normal direction to obtain the theoretical bur center path. The optimal placement path vector is the spindle direction during CNC machining.
3. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The methods for obtaining the morphological features and edge cross-sectional thickness of each path point include: In the path points, obtain the position coordinates of the path point and the two adjacent path points, and use the discrete curvature formula to calculate the curvature of the path point as a morphological feature. At the path point, a ray is emitted toward the preparation body along the reverse extension direction away from the tooth body, and the distance the ray penetrates the preparation body is taken as the edge section thickness.
4. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method for obtaining the path deviation cost of each path point based on the anatomical features includes: The morphological similarity between the morphological features of each path point and the most prominent morphology is obtained, as well as the thickness comparison features between the edge section thickness of each path point and the thickness threshold. Based on the morphological and thickness comparison features, the path deviation cost is obtained by weighted fusion of acceleration-sensitive weight and cutting force load coefficient. The thickness threshold, acceleration-sensitive weight, and cutting force load coefficient are obtained by experimental calibration.
5. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The direction away from the tooth includes: At each path point, calculate the second tangential vector of the path point, and obtain the cross-sectional circle of the path point based on the second tangential vector and the position coordinates of the path point; obtain the corresponding point of the path point on the edge curve of the preparation body, calculate the Euclidean distance between each point on the cross-sectional circle and the corresponding point, and take the direction of the path point pointing to the point on the cross-sectional circle corresponding to the maximum value of the Euclidean distance as the direction away from the tooth body.
6. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method for obtaining multiple simulated path points includes: Multiple sampling points are obtained by sampling along the straight line corresponding to each path point in the direction away from the tooth, and these multiple sampling points are used as simulated path points.
7. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method for obtaining the distance difference between each simulated path point and its corresponding path point includes: For each path point and its corresponding simulated path point, calculate the Euclidean distance between the simulated path point and the corresponding path point, and use this distance difference as the distance difference.
8. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method of performing point-by-point matching of simulated path points between adjacent path points to obtain simulated path point pairs, and obtaining the path transfer cost between simulated path point pairs based on distance differences and path deviation costs, includes: In two adjacent path points, the simulated path point corresponding to the current step path point is matched with the simulated path point corresponding to the next step path point point one by one to obtain multiple simulated path point pairs; Between two adjacent path points, calculate the squared difference of the distance difference between the current path point and the next path point. Multiply the squared difference by the path deviation cost of the current path point as the motion potential energy term. Calculate the squared distance difference of the current path point as the geometric potential energy term. Sum the geometric potential energy term and the motion potential energy term as the path transfer cost between the simulated path point pairs.
9. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method for optimizing all paths to obtain the locally optimal path between adjacent path points includes: Select the starting point of the path from all path points. For each simulated path point pair between the path point and the starting point, traverse all path transition costs on the complete path formed by connecting the simulated path point pairs end to end from the current step path point to the starting point. The simulated path point pair contained in the path with the smallest sum of path transition costs is taken as the local optimal path. The local optimal path satisfies the closure constraint.
10. The digital simulation evaluation method for minimally invasive restoration of tooth defects according to claim 1, characterized in that, The method for assessing tooth defect restoration based on the degree of difference between the simulated bur center path and the theoretical bur center path, as well as anatomical features, includes: The difference in positional coordinates between the simulated path point on the simulated needle center path and the path point on the theoretical needle center path is calculated. Based on the magnitude of the difference in positional coordinates, the size of the machining gap is determined. Based on the difference between the anatomical features and the preset thresholds in the corresponding dimensions, the causes of the machining gap are analyzed. The size of the machining gap and the causes of the machining gap are used as the evaluation results.
