Spring probe system for rapid registration of articular cartilage based on 3D printing

Abstract

The invention discloses a spring probe system for rapid registration of articular cartilage based on 3D printing, and relates to the technical field of medical imaging and orthopedic surgery navigation. The problem that the joint registration precision is insufficient due to the fact that cartilage cannot be developed in traditional CT imaging is solved. Through personalized customization of the 3D printing guide plate, convenient installation of positioning holes, adjustability of the spring probe assembly and real-time data monitoring, and automatic data processing of the intelligent registration module, manual operation is reduced, the operation efficiency is improved, the spring probe assembly is reasonable in structure and adjustable in elastic force, stable operation of the system in an operation is guaranteed, and the operation safety is improved. Multi-position CT scanning is combined with an advanced image processing and registration technology, an accurate initial model is constructed, a spring probe assembly collects data and is combined with an elastic deformation compensation model to correct registration errors in real time, the registration accuracy is improved, reliable support is provided for orthopedic surgery navigation, the surgery success rate can be improved, and patient wounds and surgery risks are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of medical imaging and orthopedic surgery navigation, and particularly relates to a spring probe system for rapid registration of articular cartilage based on 3D printing. Background Technique

[0002] In orthopedic surgery, the accuracy of joint registration plays a crucial role in the surgical outcome. Traditional joint registration methods have many deficiencies. For example, articular cartilage cannot be visualized in CT imaging, resulting in a significant difference between the preoperative model and the actual joint morphology during the operation, which affects the accuracy of surgical navigation. At the same time, most of the existing registration methods rely on manual experience or single bone feature matching, which not only has low accuracy but also takes a long time, seriously restricting the surgical efficiency and quality. Although MRI can show cartilage, it has high costs, long examination times, and difficulties in fusing with CT data. In addition, existing contact measurement tools are prone to damaging cartilage and cannot provide real-time feedback on cartilage deformation data. Summary of the Invention

[0003] The purpose of the present invention is to provide a spring probe system for rapid registration of articular cartilage based on 3D printing. By obtaining detailed information of the human joint and combining 3D printing technology, accurate positioning and registration of the joint are achieved, providing accurate navigation support for orthopedic surgery, thereby improving the success rate and safety of the surgery, reducing the trauma and recovery time of patients, so as to solve the problems raised in the above background technique.

[0004] To achieve the above purpose, the present invention provides the following technical solutions:

[0005] A spring probe system for rapid registration of articular cartilage based on 3D printing, including a guide plate, a spring probe assembly, and an intelligent registration module. Based on the CT data of the patient, the guide plate is manufactured using 3D printing technology. A plurality of uniformly distributed positioning holes are provided on one side of the guide plate. The spring probe assembly is movably connected to the positioning holes. The intelligent registration module is configured to, based on the sensors integrated at the tip of the spring probe assembly, obtain the cartilage thickness, pressure, and deformation data on the cartilage surface in real time, and combine with the kinematic model of the joint to dynamically adjust the registration parameters of the cartilage region.

[0006] Further, the intelligent registration module further includes: automatically identifying and extracting the key feature points of the bone in the CT data, and based on the point cloud registration algorithm, through introducing a local feature constraint mechanism, achieving preliminary rapid matching of the bone part to obtain an initial registration model.

[0007] Further, the CT data specifically includes:

[0008] Using a CT scanning device to scan the patient's joint in multiple positions to obtain the CT data of the patient's joint. The scanning range includes all the cartilage regions and the hard bone part of the joint;

[0009] Fuse the CT data of multiple body positions obtained, construct a joint image dataset of the patient, and preprocess the joint image dataset to highlight the edge features of bones and cartilage and the internal texture information;

[0010] Extract features from the processed image data, segment the bone and cartilage regions based on the feature extraction results, and label the segmented cartilage regions;

[0011] Construct a three-dimensional joint model of the patient based on the segmented joint image data, optimize the internal structure design of the guide plate based on the three-dimensional joint model of the patient, and determine the fitting conditions of the guide plate with the joint cartilage region and the hard bone part;

[0012] When 3D printing the guide plate, optimize the parameters of layer manufacturing according to the shape and structural characteristics of the guide plate, and obtain the distribution range of the positioning holes according to the joint cartilage region covered by the guide plate.

[0013] Further, segmenting the bone and cartilage regions based on the feature extraction results further includes:

[0014] Traverse each pixel point in the segmentation result image, screen and integrate the pixel points belonging to the bone and cartilage regions, and generate a pixel gray value group corresponding to the bone region and the cartilage region based on the integration result;

[0015] Calculate the pixel gray value group of the bone region to obtain the gray mean value of the bone region, and obtain the gray standard deviation of the bone region based on the difference between each pixel gray value and the gray mean value;

[0016] Calculate the pixel gray value group of the cartilage region to obtain the gray mean value of the cartilage region, and obtain the gray standard deviation of the cartilage region based on the difference between each pixel gray value and the gray mean value;

[0017] Determine the weight coefficients of the bone and cartilage regions in the segmentation result image based on the area ratio, importance in the joint, and contrast with the surrounding tissues of the bone and cartilage regions;

[0018] Calculate a threshold based on the weight coefficients of the bone and cartilage regions, and the gray mean values and standard deviations of the bone and cartilage regions, and this threshold can effectively distinguish the bone and cartilage regions from the background;

[0019] Perform binary processing on the segmentation result image based on the threshold, and optimize the boundaries of the bone and cartilage regions based on the processing result.

[0020] Further, obtaining an initial registration model specifically includes:

[0021] Filter the joint image dataset after fusion processing and preprocessing, and adjust the parameters of the filter according to the texture complexity and gray-scale change degree of different regions in the image;

[0022] Adjust the gray-scale distribution of the image to a predetermined distribution form, and match it with the gray-scale distribution characteristics of the standard joint image;

[0023] Based on the matching result, identify and extract the key feature points of the bones in the joint image, obtain the coordinate information of the key feature points of the bones, and convert the coordinate information of the key feature points of the bones in the three-dimensional space into point cloud data;

[0024] According to the density distribution of the points in the point cloud, divide the points with connected density into different clusters, identify the outlier points with significantly lower density than the surrounding area and remove them;

[0025] Use a curvature-based sampling algorithm to resample the remaining point cloud, increase the number of sampling points in the area with larger point cloud curvature, and reduce the number of sampling points in the flat area with smaller curvature to generate optimized point cloud data of the key feature points of the bones.

[0026] Further, obtaining the initial registration model also includes:

[0027] Based on the optimized point cloud data of the key feature points of the bones, obtain the local features of each key feature point of the bones, generate feature vectors, and construct a feature vector database;

[0028] Combined with the local feature constraint mechanism, perform point cloud registration with the reference point cloud as the target, dynamically adjust the corresponding point search range and matching weight according to the local feature matching situation, and construct the initial registration model;

[0029] Further, dynamically adjust the registration parameters of the cartilage area, specifically including:

[0030] Process the probe data collected by the spring probe assembly, including cartilage thickness, pressure, and deformation data;

[0031] Construct an elastic deformation compensation model to describe the law of elastic deformation of cartilage;

[0032] Input the preprocessed probe data into the elastic deformation compensation model, calculate the elastic deformation amount of the cartilage area, and according to the basic principle of point cloud registration, convert the deformation amount of the cartilage into the adjustment amount of the registration parameters, the parameters to be adjusted for registration;

[0033] Apply the calculated adjustment amount of the registration parameters to the initial registration model, update the registration parameters of the cartilage area, and realize the dynamic correction of the registration model.

[0034] Furthermore, the other side of the guide plate is attached to the cartilage-free part of the joint. The guide plate is provided with a fixing clip that penetrates the guide plate and abuts against the cartilage-free part of the joint, and an anti-slip pattern is provided on the side close to the joint. The side of the guide plate with positioning holes covers the cartilage area of the joint and is used to carry the spring probe assembly.

[0035] Furthermore, the spring probe assembly includes a probe tip, a spring, and a probe body. The spring is fixedly installed in the probe tip and connected to one side of the probe body. A chute for the movement of the probe tip is provided in the probe body, and the probe body is movably connected to the probe tip through the chute.

[0036] Furthermore, a micro-capacitance sensor is integrated at the tip of the probe head for real-time monitoring of the probe's telescopic amount and pressure data to calculate the actual thickness of the cartilage. A spring force adjustment plate is installed in the probe body. One side of the spring force adjustment plate is fixedly connected to an adjustment column, and the adjustment column penetrates the probe body and is connected to an adjustment handle.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] Through personalized customization of the 3D-printed guide plate, the positioning holes are convenient for installation. The spring probe assembly is adjustable and data monitoring is real-time. The intelligent registration module automatically processes data, reducing manual operations and improving surgical efficiency. The anti-slip design of the guide plate fixing clip stably connects to the joint. The spring probe assembly has a reasonable structure and adjustable spring force, ensuring the stable operation of the system during surgery. Multi-position CT scanning combined with advanced image processing and registration technology constructs an accurate initial model. The data collected by the spring probe assembly combined with the elastic deformation compensation model can correct the registration error in real-time, improving the registration accuracy, providing reliable support for orthopedic surgical navigation, helping to improve the surgical success rate, and reducing patient trauma and surgical risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the spring probe system for rapid registration of articular cartilage of the present invention;

[0040] Figure 2 It is a partial cross-sectional view of the spring probe assembly of the present invention;

[0041] Figure 3 It is a flowchart of the spring probe system for rapid registration of articular cartilage based on 3D printing of the present invention.

[0042] In the figure: 1. Guide plate; 11. Positioning hole; 12. Fixing clip; 2. Spring probe assembly; 21. Probe tip; 22. Spring; 23. Probe body; 24. Spring force adjustment plate; 25. Adjustment column. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0044] To solve the technical problem of insufficient joint registration accuracy caused by the inability of cartilage to be visualized in traditional CT imaging, please refer to Figures 1 - 3 , the following technical solutions are provided in this embodiment:

[0045] A spring probe system for rapid registration of articular cartilage based on 3D printing, including a guide plate 1, a spring probe assembly 2 and an intelligent registration module. Based on the CT data of the patient, the guide plate 1 is manufactured using 3D printing technology. A uniformly distributed positioning hole 11 is provided on one side of the guide plate 1, and the layout and spacing of the hole positions are optimized to meet the measurement requirements of different regions of articular cartilage. The spring probe assembly 2 is movably connected to the positioning hole 11;

[0046] In this embodiment, the other side of the guide plate 1 is attached to the cartilage-free part of the joint. A fixing clip 12 is provided on the guide plate 1. The fixing clip 12 penetrates the guide plate 1 and abuts against the cartilage-free part of the joint, and anti-slip lines are provided on the side close to the joint. The side of the guide plate 1 with the positioning hole 11 covers the articular cartilage area for carrying the spring probe assembly 2. The material is selected as a medical-grade polymer material with biocompatibility, high strength and low elastic modulus. While ensuring the mechanical properties of the guide plate, this material can reduce the adverse effects on surrounding tissues and has a certain degradation period in the body. Materials with appropriate degradation characteristics can be selected according to surgical needs;

[0047] In this embodiment, the spring probe assembly 2 includes a probe head 21, a spring 22 and a probe body 23. The probe head adopts a special triangular pyramid design. This shape can not only pierce the cartilage more easily but also provide more stable support during the measurement to ensure the accuracy of the measurement data. The spring 22 is fixedly installed in the probe head 21 and connected to one side of the probe body 23. A chute for the movement of the probe head 21 is provided in the probe body 23. The probe body 23 is movably connected to the probe head 21 through the chute. A micro-capacitance sensor is integrated at the tip of the probe head 21 for real-time monitoring of the probe telescopic amount and pressure data to calculate the actual thickness of the cartilage. A spring force adjustment plate 24 is installed in the probe body 23. One side of the spring force adjustment plate 24 is fixedly connected to an adjustment column 25. The adjustment column 25 penetrates the probe body 23 and is connected to an adjustment handle. By rotating the adjustment column 25, the spring force adjustment plate 24 is adjusted, thereby adjusting the stiffness of the spring 22 to achieve the adjustment of the preset spring force threshold;

[0048] The intelligent registration module is configured to, based on the sensors integrated at the tip of the spring probe assembly 2, obtain in real time the cartilage thickness, pressure, and deformation data on the cartilage surface, combine with the kinematic model of the joint, dynamically adjust the registration parameters of the cartilage region, achieve real-time correction of the registration error, generate a high-precision navigation model, automatically identify and extract the key skeletal feature points in the CT data, and has higher accuracy and efficiency compared with the traditional method of manually selecting feature points. Based on the point cloud registration algorithm, by introducing a local feature constraint mechanism, it realizes the preliminary and rapid matching of the skeletal part, obtains an initial registration model, and while ensuring the rapid matching of the skeletal part, significantly improves the matching accuracy.

[0049] In this embodiment, taking the knee joint registration as an example: Data acquisition and modeling: Obtain the CT data of the patient's knee joint, and segment the bone and cartilage regions (the cartilage region is marked as the "non-visualizable area"); 3D print a guide plate that matches the cartilage region and configure a spring probe on the guide plate. Intraoperative registration operation: Rigid registration: Completely fit the guide plate with the key cartilage part. Cartilage measurement: Insert the spring probe into the preset hole position and gently press it to the preset elastic force threshold; the capacitive sensor records the deformation amount in real time and calculates the actual cartilage thickness (for example: the original model thickness is 5 mm, the measured value is 4.2 mm, and the registration offset is compensated). Dynamic correction: The registration algorithm fuses the data of multiple probes to generate a corrected navigation model.

[0050] In this embodiment, the guide plate 1 has a unique shape that highly adapts to the joint cartilage region and part of the hard bone surface. Its surface is provided with a precise positioning structure, which can precisely fit with the specific anatomical landmarks of the joint part to ensure the stable placement of the guide plate during the operation. The spring probe assembly 2 integrates advanced spring mechanics feedback, capacitive length measurement, and pressure monitoring triple technologies, can accurately preset the elastic force threshold of the spring according to the characteristics of different joint cartilages and surgical requirements, thereby precisely controlling the compression depth of the cartilage, effectively avoiding cartilage damage caused by excessive compression, and real-time and accurately monitoring the telescopic amount of the probe, being able to monitor the pressure change of the probe during the process of contacting the cartilage in real time. Once the pressure exceeds the safe range, a warning signal is immediately issued to ensure the safety of the surgical operation.

[0051] In this embodiment, the CT data specifically includes:

[0052] Use a CT scanning device to scan the patient's joint in multiple positions, including standard anteroposterior, lateral, and special functional positions such as the flexion position and extension position of the knee joint, obtain the CT data of the patient's joint, and the scanning range includes all the cartilage regions and hard bone parts of the joint;

[0053] Fuse the CT data of multiple body positions obtained to construct a joint image dataset of the patient. The joint image dataset comprehensively covers all the cartilage regions and related hard bone parts of the joint, providing richer information for subsequent analysis. Then, preprocess the joint image dataset. Use an image enhancement algorithm based on wavelet transform to highlight the edge features and internal texture information of the bones and cartilage, and enhance the features of different tissues specifically, so that the contrast between the bones and cartilage in the image is significantly improved, facilitating subsequent segmentation operations. At the same time, use algorithms such as median filtering to remove noise in the image, improving the quality and stability of the image;

[0054] Extract features from the processed image data, segment the bone and cartilage regions based on the feature extraction results, and mark the segmented cartilage regions. The marking information includes key parameters such as the position, shape, and thickness of the cartilage;

[0055] Construct a three-dimensional joint model of the patient based on the segmented joint image data, and optimize the internal structure design of the guide plate 1 based on the three-dimensional joint model of the patient to determine the fitting situation between the guide plate 1 and the joint cartilage region and hard bone part;

[0056] When 3D printing the guide plate 1, according to the shape and structural characteristics of the guide plate 1, optimize the parameters of layer manufacturing, select an appropriate layer thickness. For parts with complex shapes and high precision requirements, use a thinner layer thickness such as 0.1 - 0.2 mm to ensure that the surface of the printed guide plate is smooth and the details are clear; for parts with relatively simple shapes, appropriately increase the layer thickness such as 0.3 - 0.5 mm to improve the printing efficiency. During the printing process of each layer, by adjusting the movement trajectory, speed, and extrusion amount of the printing nozzle, optimize the stacking method of the material to ensure the bonding strength between layers, and at the same time reduce deformation and defects during the printing process. After printing, perform post-processing on the guide plate, such as grinding and polishing, to further improve the surface quality and adaptability of the guide plate; and obtain the distribution range of the positioning holes 11 according to the joint cartilage region covered by the guide plate 1 to ensure the position accuracy and angular accuracy of the spring probe assembly 2 during the installation process

[0057] In this embodiment, segmenting the bone and cartilage regions based on the feature extraction results further includes:

[0058] Traverse each pixel point in the segmented result image, screen and integrate the pixel points belonging to the bone and cartilage regions, and generate a pixel gray value group corresponding to the bone region and cartilage region based on the integration result;

[0059] Calculate the pixel gray value group of the bone region to obtain the gray mean value of the bone region, and obtain the gray standard deviation of the bone region based on the difference between each pixel gray value and the gray mean value;

[0060] Calculate the pixel gray value group of the cartilage area to obtain the gray mean value of the cartilage area, and obtain the gray standard deviation of the cartilage area based on the difference between each pixel gray value and the gray mean value;

[0061] Based on the area ratio, importance in the joint, and contrast with surrounding tissues of the bone and cartilage areas, determine the weight coefficients of the bone and cartilage areas in the segmentation result image. For example, construct a weight coefficient determination model based on regional features and clinical needs. If the bone area is large and plays a key role in positioning in surgical planning, assign it a relatively high weight; if the contrast between the cartilage area and surrounding tissues is low, to highlight its boundary, appropriately adjust the proportion of the cartilage area in the weight calculation. Train this model with a large amount of sample data, where the samples cover CT images of different patients and different joint parts and their corresponding segmentation results. Finally, according to the specific features of the current image to be segmented, input the trained model to obtain the weight coefficients of the bone and cartilage areas suitable for this image;

[0062] Calculate a threshold based on the weight coefficients of the bone and cartilage areas, as well as the gray mean values and standard deviations of the bone and cartilage areas. This threshold can effectively distinguish the bone and cartilage areas from the background;

[0063] In this embodiment, the weighted average method is used. Multiply the gray mean value of the bone area by its weight coefficient, multiply the gray mean value of the cartilage area by its weight coefficient, add the two together, and then add an empirical offset determined according to a large amount of experimental data. At the same time, considering the regional gray fluctuation situation reflected by the gray standard deviation, introduce a regulation factor related to the standard deviation, and this factor is dynamically adjusted according to the stability of the regional gray fluctuation. For example, if the gray standard deviation of a certain area is large, indicating large gray fluctuations, to avoid misjudgment, appropriately increase the influence of the regulation factor on the threshold. Through such a calculation method, a precise threshold is obtained by integrating various factors, and this threshold can effectively distinguish the bone and cartilage areas from the background;

[0064] Perform binary processing on the segmentation result image based on the threshold, and optimize the boundaries of the bone and cartilage areas based on the processing result.

[0065] In this embodiment, the calculated threshold is applied to the preliminary segmentation result image, and each pixel in the image is judged. If the gray value of the pixel is greater than the threshold, it is determined as a bone or cartilage area and assigned a foreground pixel value such as 255; if the gray value is less than the threshold, it is determined as a background pixel and assigned a background pixel value such as 0, completing the binarization process of the image. To further optimize the boundary, erosion and dilation operations in morphological operations are adopted. First, the erosion operation is performed to remove the isolated pixel points and small protrusions on the boundary of the bone and cartilage areas in the binarized image, making the boundary initially smooth; then the dilation operation is performed to restore the area reduced by erosion and connect the possibly disconnected boundaries. During the erosion and dilation processes, according to the morphological characteristics of the bone and cartilage areas, structural elements with appropriate shapes and sizes such as circles and rectangles are selected to ensure that while optimizing the boundary, the original shape and size of the area will not be overly changed, thus more clearly defining the boundaries of the bone and cartilage.

[0066] In this embodiment, obtaining the initial registration model specifically includes:

[0067] Performing filtering processing on the fused and preprocessed joint image dataset, and adjusting the parameters of the filter according to the texture complexity and gray value change degree of different regions in the image;

[0068] Based on the boundaries of the bone and cartilage areas, a smaller filtering window and a lower filtering intensity are adopted to retain edge details. For relatively smooth areas, the filtering window and intensity are increased to further remove residual noise;

[0069] Using the histogram specification technique, the gray distribution of the image is adjusted to a predetermined distribution form and matched with the gray distribution characteristics of the standard joint image, enhancing the visual effect of the bone and cartilage tissues in the joint image, improving the contrast between different tissues, and facilitating more accurate identification and extraction of the key feature points of the bone in the subsequent process;

[0070] Based on the matching result, identifying and extracting the key feature points of the bone in the joint image, obtaining the coordinate information of the key feature points of the bone, and converting the coordinate information of the key feature points of the bone in the three-dimensional space into point cloud data;

[0071] According to the density distribution of the points in the point cloud, the points with connected densities are divided into different clusters, and the outlier points with significantly lower density than the surrounding areas are identified and removed;

[0072] Adopting a curvature-based sampling algorithm to resample the remaining point cloud, and for the areas with larger point cloud curvature, that is, the parts where the bone surface changes more violently, increasing the number of sampling points to more accurately reflect the shape details of the bone, reducing the number of sampling points in the flat areas with smaller curvature to avoid data redundancy, and at the same time improving the calculation efficiency of subsequent registration, generating the optimized point cloud data of the key feature points of the bone.

[0073] Based on the optimized point cloud data of the key skeletal feature points, obtain the local features of each key skeletal feature point, generate a feature vector representing the local shape and spatial relationship of the points, and construct a feature vector database;

[0074] Adjust the neighborhood radius according to the average point spacing of the point cloud, determine the neighborhood range of each key skeletal feature point, and within the neighborhood, calculate geometric relationships such as the distance and normal angle between the key skeletal feature point and the neighborhood points, and generate the FPFH feature vector of each key skeletal feature point;

[0075] Combined with the local feature constraint mechanism, perform point cloud registration with the reference point cloud as the target, dynamically adjust the corresponding point search range and matching weight according to the local feature matching situation. For regions with high local feature similarity, narrow the corresponding point search range to improve the search efficiency, and increase the weight of these matching point pairs when calculating the transformation parameters, making the registration result more dependent on reliable matches; for regions with large local feature differences, appropriately expand the search range to find more potential corresponding points, and at the same time reduce the weight of the matching point pairs in these regions to reduce the impact of false matches. Stop the iteration when the matching error is less than the preset threshold, and construct an initial registration model;

[0076] In this embodiment, apply the initial registration model to the test data with known real joint structures for preliminary verification, calculate indexes such as the average distance error and the maximum distance error between the registered point cloud and the point cloud of the real joint structure. If the error exceeds the acceptable range, readjust the local feature matching parameters or perform outlier detection and elimination again. Verify the accuracy of the model through actual tests, and optimize the model targeted for those that do not meet the accuracy requirements to ensure that the final registration model can meet the accuracy requirements of practical applications such as clinical surgical navigation, and improve the reliability and practicality of the entire joint cartilage rapid registration system.

[0077] In this embodiment, dynamically adjust the registration parameters of the cartilage region, specifically including:

[0078] Process the probe data collected by the spring probe assembly 2, including cartilage thickness, pressure, and deformation data;

[0079] Construct an elastic deformation compensation model to describe the law of cartilage elastic deformation. For example, use the finite element method to model the cartilage, divide the cartilage into multiple tiny units, and simulate the deformation of the cartilage under different parts and different force conditions. Through the analysis of a large number of simulation results and actual measurement data, determine the parameters in the model, such as the relationship expression between the elastic modulus and pressure and deformation. At the same time, consider the dynamic load on the cartilage during joint movement, so that the model can more realistically reflect the elastic deformation behavior of the cartilage in the actual surgical scenario;

[0080] Input the preprocessed probe data into the elastic deformation compensation model, calculate the elastic deformation amount of the cartilage region, and according to the basic principle of point cloud registration, convert the deformation amount of the cartilage into the adjustment amount of the registration parameters, which are the parameters to be adjusted for registration;

[0081] Apply the calculated adjustment amount of the registration parameters to the initial registration model, update the registration parameters of the cartilage region, and achieve the dynamic correction of the registration model.

[0082] In this embodiment, collect and process the probe data of the spring probe assembly, construct an elastic deformation compensation model to simulate the cartilage deformation, and determine the model parameters by analyzing the simulation results and the actual measurement data; then, input the preprocessed data into the model to calculate the elastic deformation amount, convert it into the adjustment amount of the registration parameters and apply it to the initial registration model to achieve dynamic correction, which not only improves the registration accuracy, enhances the adaptability, but also provides personalized treatment for patients, reduces the surgical risk, and optimizes the surgical process.

[0083] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A spring probe system for rapid registration of articular cartilage based on 3D printing, characterized in that, It includes a guide plate (1), a spring probe assembly (2) and an intelligent registration module. Based on the patient's CT data, the guide plate (1) is manufactured using 3D printing technology. A side of the guide plate (1) is provided with evenly distributed positioning holes (11). The spring probe assembly (2) is movably connected to the positioning holes (11). The intelligent registration module is configured to, based on the sensors integrated at the tips of the spring probe assembly (2), obtain in real time the cartilage thickness, pressure and deformation data on the cartilage surface, and combine with the kinematic model of the joint to dynamically adjust the registration parameters of the cartilage region.

2. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 1, wherein The intelligent registration module further includes: automatically identifying and extracting the key feature points of the bones in the CT data, and based on the point cloud registration algorithm, realizing the preliminary and rapid matching of the bone part by introducing a local feature constraint mechanism to obtain an initial registration model.

3. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 2, characterized in that, The CT data specifically includes: Using a CT scanning device to scan the patient's joint in multiple positions to obtain the CT data of the patient's joint, and the scanning range includes all the cartilage regions and the bony parts of the joint; Performing fusion processing on the obtained CT data in multiple positions to construct a joint image data set of the patient, and preprocessing the joint image data set to highlight the edge features and internal texture information of the bones and cartilage; Performing feature extraction on the processed image data, segmenting the bone and cartilage regions based on the feature extraction results, and marking the segmented cartilage regions; Constructing a three-dimensional model of the patient's joint based on the segmented joint image data, and optimizing the internal structure design of the guide plate (1) based on the three-dimensional model of the patient's joint to determine the fitting condition of the guide plate (1) with the joint cartilage region and the bony part; When 3D printing the guide plate (1), optimizing the parameters of the layered manufacturing according to the shape and structural characteristics of the guide plate (1), and obtaining the distribution range of the positioning holes (11) according to the joint cartilage region covered by the guide plate (1).

4. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 3, characterized in that Segmenting the bone and cartilage regions based on the feature extraction results further includes: Traversing each pixel point in the segmented result image, screening and integrating the pixel points belonging to the bone and cartilage regions, and generating a pixel gray value group corresponding to the bone region and the cartilage region based on the integration result; Calculating the pixel gray value group of the bone region to obtain the gray mean value of the bone region, and obtaining the gray standard deviation of the bone region based on the difference between each pixel gray value and the gray mean value; Calculating the pixel gray value group of the cartilage region to obtain the gray mean value of the cartilage region, and obtaining the gray standard deviation of the cartilage region based on the difference between each pixel gray value and the gray mean value; Determining the weight coefficients of the bone and cartilage regions in the segmented result image based on the area ratio, importance in the joint and contrast with the surrounding tissues of the bone and cartilage regions; Calculating a threshold based on the weight coefficients of the bone and cartilage regions, and the gray mean values and standard deviations of the bone and cartilage regions, and this threshold can effectively distinguish the bone and cartilage regions from the background; Performing binary processing on the segmented result image based on the threshold, and optimizing the boundaries of the bone and cartilage regions based on the processing result.

5. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 2, characterized in that, Obtaining the initial registration model specifically includes: Filter the fused and preprocessed joint image dataset, and adjust the parameters of the filter according to the texture complexity and gray-scale change degree of different regions in the image; Adjust the gray-scale distribution of the image to a predetermined distribution form, and match it with the gray-scale distribution characteristics of the standard joint image; Based on the matching result, identify and extract the key feature points of the bones in the joint image, obtain the coordinate information of the key feature points of the bones, and convert the coordinate information of the key feature points of the bones in the three-dimensional space into point cloud data; According to the density distribution of the points in the point cloud, divide the points with connected density into different clusters, identify the outlier points with significantly lower density than the surrounding areas and remove them; Use a curvature-based sampling algorithm to resample the remaining point cloud, increase the number of sampling points in the area with larger curvature of the point cloud, and reduce the number of sampling points in the flat area with smaller curvature to generate optimized point cloud data of the key feature points of the bones.

6. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 5, characterized in that, Obtain the initial registration model, and it also includes: Based on the optimized point cloud data of the key feature points of the bones, obtain the local features of each key feature point of the bones, generate feature vectors, and construct a feature vector database; Combine the local feature constraint mechanism, perform point cloud registration with the reference point cloud as the target, dynamically adjust the corresponding point search range and matching weight according to the local feature matching situation, and construct the initial registration model.

7. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 1, wherein Dynamically adjust the registration parameters of the cartilage area, specifically including: Process the probe data collected by the spring probe assembly (2), including cartilage thickness, pressure, and deformation data; Construct an elastic deformation compensation model to describe the law of elastic deformation of cartilage; Input the preprocessed probe data into the elastic deformation compensation model, calculate the elastic deformation amount of the cartilage area, and according to the basic principle of point cloud registration, convert the deformation amount of the cartilage into the adjustment amount of the registration parameters, and the parameters to be adjusted for registration; Apply the calculated adjustment amount of the registration parameters to the initial registration model, update the registration parameters of the cartilage area, and realize the dynamic correction of the registration model.

8. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 1, characterized in that, The other side of the guide plate (1) is attached to the part of the joint without cartilage. The guide plate (1) is provided with a fixing clip (12). The fixing clip (12) penetrates through the guide plate (1) and abuts against the part of the joint without cartilage, and anti-slip lines are provided on the side close to the joint. The side of the guide plate (1) provided with the positioning holes (11) covers the cartilage area of the joint and is used to carry the spring probe assembly (2).

9. The spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 1, characterized in that, The spring probe assembly (2) includes a probe head (21), a spring (22), and a probe body (23). The spring (22) is fixedly installed in the probe head (21) and connected to one side of the probe body (23). A sliding groove for the probe head (21) to move is provided in the probe body (23), and the probe body (23) is movably connected to the probe head (21) through the sliding groove.

10. A spring probe system for rapid registration of articular cartilage based on 3D printing according to claim 1, characterized in that, The tip of the probe head (21) is integrated with a micro-capacitance sensor for real-time monitoring of the probe telescopic amount and pressure data to calculate the actual thickness of the cartilage. A spring force adjustment plate (24) is installed in the probe body (23). One side of the spring force adjustment plate (24) is fixedly connected to the adjustment column (25), and the adjustment column (25) penetrates through the probe body (23) and is connected to the adjustment handle.

Images