Disclosure of Invention
The invention mainly aims to provide a collaborative control method and a collaborative control system for an air source heat pump aggregation prediction and peak shaving task instruction, which can realize the fine management of oral details, and formulate a more accurate correction track path so as to avoid unnecessary adjustment.
In order to achieve the above object, the present invention provides a three-dimensional image guiding correction planning method, which is characterized by comprising:
acquiring a three-dimensional CT image and a two-dimensional X-ray image of a patient oral cavity, and integrating the three-dimensional CT image and the two-dimensional X-ray image to obtain a corresponding comprehensive image set;
Extracting features of the comprehensive image set to obtain a tooth image and a jawbone image, and carrying out three-dimensional construction according to the tooth image and the jawbone image to obtain an initial three-dimensional model;
Carrying out alignment fusion on the tooth image and the jawbone image to obtain a corresponding oral fusion image;
registering the oral fusion image with the initial three-dimensional model based on a preset registration method to obtain a corresponding oral model;
Performing track analysis on the oral cavity of the patient according to the oral cavity model to obtain an initial track path, and performing iterative optimization on the initial track path to obtain an optimized track scheme;
And simulating the track of the oral cavity model based on the optimized track scheme to obtain a corresponding simulated track result.
Further, the integrating the three-dimensional CT image and the two-dimensional X-ray image to obtain a corresponding integrated image set includes:
performing distortion correction processing on the three-dimensional CT image to obtain a corrected three-dimensional CT image, and performing distortion correction on the two-dimensional X-ray image to obtain a corrected two-dimensional X-ray image;
performing alignment matching on the corrected three-dimensional CT image and the corrected two-dimensional X-ray image to obtain a preliminary alignment image;
performing multi-resolution pyramid decomposition on the preliminary registration images to obtain image levels with different scales, and integrating all the image levels to obtain a multi-scale image set;
Performing wavelet transformation on the multi-scale image set, extracting features of different frequency components to obtain a wavelet coefficient matrix, and performing feature fusion on the wavelet coefficient matrix to obtain fusion feature representation;
and carrying out inverse wavelet transformation on the fusion characteristic representation and reconstructing to obtain the comprehensive image set.
Further, the feature extraction is performed on the comprehensive image set to obtain a tooth image and a jawbone image, and the three-dimensional construction is performed according to the tooth image and the jawbone image to obtain an initial three-dimensional model, which comprises:
Performing image segmentation processing on the comprehensive image set to obtain the segmented dental image and the segmented jawbone image;
extracting tooth profile features of the tooth image according to a preset edge detection algorithm to obtain tooth profile information;
Extracting the characteristics of the jawbone skeleton line of the jawbone image according to a preset skeleton line extraction algorithm to obtain jawbone skeleton line information;
performing three-dimensional reconstruction according to the tooth contour information and the jawbone skeleton line information to obtain a preliminary reconstruction model;
and carrying out smoothing treatment on the model surface of the preliminary reconstruction model to obtain the initial three-dimensional model.
Further, the aligning and fusing the dental image and the jawbone image to obtain a corresponding oral fused image includes:
performing single segmentation processing on the tooth image according to the tooth profile information to obtain corresponding single tooth profile information;
Performing edge detection on the jaw image according to the jaw skeleton line information to obtain jaw boundary information;
performing relationship construction based on single tooth profile information and jaw boundary information to obtain a corresponding tooth jaw corresponding relationship matrix;
Carrying out rigid registration on the tooth image and the jawbone image according to the tooth jawbone corresponding relation matrix to obtain a preliminary alignment result;
Performing non-rigid deformation treatment on the preliminary alignment result to obtain a fine alignment result;
And carrying out interpolation fusion on the tooth image and the jawbone image according to the fine alignment result to obtain the oral fusion image.
Further, the registering the oral fusion image with the initial three-dimensional model based on a preset registering method to obtain a corresponding oral model includes:
Extracting characteristic points of the oral fusion image to obtain a first characteristic point set;
Extracting feature points of the initial three-dimensional model to obtain a second feature point set;
performing coarse registration on the oral fusion image and the initial three-dimensional model according to the first characteristic point set and the second characteristic point set to obtain a coarse registration result;
carrying out fine registration on the oral fusion image and the initial three-dimensional model based on a coarse registration result to obtain a fine registration result;
Carrying out fusion processing on the oral fusion image and the initial three-dimensional model according to the fine registration result to obtain a fusion model;
Carrying out boundary frame analysis on the fusion model to obtain a model boundary frame, and carrying out three-dimensional rule grid construction based on the model boundary frame to obtain a model voxel space;
Densely sampling the surface of the fusion model to obtain surface sampling points;
Detecting whether each voxel of the model voxel space contains a surface sampling point, and marking the voxel containing the surface sampling point as occupied;
Filling voxels from the boundary to the inside of the model voxel space according to a preset filling rule, and marking the inside voxels of the model voxel space as occupied;
coding all occupied voxels according to the model voxel space to obtain a voxelized model;
And carrying out surface reconstruction on the voxelized model to obtain the oral cavity model.
Further, the performing track analysis on the oral cavity of the patient according to the oral cavity model to obtain an initial track path, and performing iterative optimization on the initial track path to obtain an optimized track scheme, including:
Dividing the oral cavity model to obtain a tooth unit and a jaw unit, and analyzing the positions of the tooth unit and the jaw unit to obtain the relative position relationship of the tooth unit and the jaw unit;
Classifying and labeling the tooth units according to the relative position relationship to obtain tooth type information;
Analyzing and processing the tooth type information based on a preset tooth arrangement rule to obtain ideal arrangement positions;
calculating the displacement vector of each tooth unit according to the ideal arrangement position and the current position of the tooth unit, and comprehensively processing the displacement vectors of all the tooth units to obtain the initial track path;
Carrying out three-dimensional decomposition on each displacement vector in the initial track path to obtain a corresponding three-dimensional component vector, and carrying out stress analysis on the three-dimensional component vector to obtain corresponding stress data;
vector adjustment is carried out on the three-dimensional component vector according to the stress data, so that an optimized component vector is obtained;
Synthesizing all the optimized component vectors to obtain optimized displacement vectors;
And carrying out iterative optimization on the initial track path based on the optimized displacement vector to obtain the optimized track scheme.
Further, the performing a simulation track on the oral cavity model based on the optimized track scheme to obtain a corresponding simulation track result includes:
The oral cavity model is subjected to sectional treatment according to the optimized track scheme to obtain a plurality of oral cavity sub-models;
carrying out stress analysis on each oral cavity submodel to obtain a corresponding stress distribution diagram;
Carrying out deformation simulation on each oral cavity submodel based on the stress distribution diagram to obtain deformation submodels, and carrying out combination treatment on all the deformation submodels to obtain a track oral cavity integral model after simulating a track;
comparing and analyzing the track oral cavity integral model with a preset standard oral cavity model to obtain deviation data, and performing fine adjustment treatment on the track oral cavity integral model according to the deviation data to obtain a fine adjustment oral cavity integral model;
and performing three-dimensional rendering treatment on the fine-tuning oral cavity integral model to obtain the simulation track result.
The invention also provides a three-dimensional image guiding correction planning system, which is applied to the three-dimensional image guiding correction planning method of any one of the above items, and comprises the following steps:
The acquisition module is used for acquiring a three-dimensional CT image and a two-dimensional X-ray image of the oral cavity of a patient, and integrating the three-dimensional CT image and the two-dimensional X-ray image to obtain a corresponding comprehensive image set;
The analysis module is used for extracting the characteristics of the comprehensive image set to obtain a tooth image and a jaw image, and performing three-dimensional construction according to the tooth image and the jaw image to obtain an initial three-dimensional model;
the association module is used for carrying out alignment fusion on the tooth image and the jawbone image to obtain a corresponding oral fusion image;
The processing module is used for carrying out registration processing on the oral fusion image and the initial three-dimensional model based on a preset registration method to obtain a corresponding oral model;
the control module is used for carrying out track analysis on the oral cavity of the patient according to the oral cavity model to obtain an initial track path, and carrying out iterative optimization on the initial track path to obtain an optimized track scheme.
The execution module is used for simulating the oral cavity model based on the optimized track scheme to obtain a corresponding simulated track result
The invention also provides a three-dimensional image guiding correction planning device, which comprises:
A memory for storing a program;
And the processor is used for executing the program to realize the steps of the three-dimensional image guiding correction planning method.
The three-dimensional image guiding correction planning method, system, device and medium provided by the invention have the following beneficial effects:
By integrating the three-dimensional CT image and the two-dimensional X-ray image data, the characteristics of teeth and jawbone can be extracted more accurately, so that the construction precision of an oral cavity model is improved, and a more reliable foundation is provided for track planning. By means of alignment fusion of the image data and registration analysis according to the fusion image, fine management of oral details is achieved, more accurate track paths are made, and unnecessary adjustment is avoided. The initial path is subjected to iterative optimization based on an optimized track scheme, so that the track process can be performed efficiently in different stages, unnecessary errors are reduced, and the efficiency of the whole track process is improved. By comprehensively analyzing the oral cavity model and the track path data, a more reasonable track strategy is formulated, and the track scheme is optimized by comprehensively adjusting, so that the track effect is effectively improved, and the treatment time and discomfort of a patient are reduced. And by considering the influence of individual differences on the track scheme, the track strategy can be flexibly adjusted according to the oral characteristics and the requirement change of different patients, so that the track process is more suitable for diversified requirement scenes.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The invention will be further described with reference to the drawings and detailed description.
Referring to fig. 1, the present invention provides a three-dimensional image guiding correction planning method, which includes:
Step S1, acquiring a three-dimensional CT image and a two-dimensional X-ray image of a patient oral cavity, and integrating the three-dimensional CT image and the two-dimensional X-ray image to obtain a corresponding comprehensive image set;
step S2, extracting features of the comprehensive image set to obtain a tooth image and a jawbone image, and carrying out three-dimensional construction according to the tooth image and the jawbone image to obtain an initial three-dimensional model;
S3, carrying out alignment fusion on the tooth image and the jawbone image to obtain a corresponding oral fusion image;
Step S4, carrying out registration processing on the oral fusion image and the initial three-dimensional model based on a preset registration method to obtain a corresponding oral model;
s5, carrying out track analysis on the oral cavity of the patient according to the oral cavity model to obtain an initial track path, and carrying out iterative optimization on the initial track path to obtain an optimized track scheme;
and S6, carrying out simulation track on the oral cavity model based on the optimized track scheme to obtain a corresponding simulation track result.
Based on the above steps, the detailed procedure is as follows:
Step S1, acquiring detailed three-dimensional CT image data of the oral cavity of a patient by reading a CT scanning program. The data includes a general view of the oral structures of the teeth, gums, and jawbone.
By reading the X-ray machine, two-dimensional X-ray images of the mouth, in particular side and front images, are obtained, ensuring that the contrast and sharpness of the X-ray images are sufficient to show details of the teeth and jaw.
And denoising the acquired three-dimensional CT image, and reducing noise in the image by using non-local mean filtering or other image denoising algorithms. And carrying out contrast enhancement on the two-dimensional X-ray image, and adopting a histogram equalization or self-adaptive contrast enhancement technology to promote detail display of the image.
The CT image and the X-ray image are respectively subjected to distortion correction, and the distortion caused by the movement of the imaging device or the patient is corrected by using a geometric transformation method.
Registering the preprocessed three-dimensional CT image and the preprocessed two-dimensional X-ray image, and aligning the images of the two modes by adopting a registration algorithm (such as SIFT, SURF and the like) based on feature point matching.
And integrating the registered images by using a multi-mode image fusion technology (such as a fusion algorithm based on mutual information) to obtain a comprehensive image set.
And S2, applying an image segmentation algorithm (such as an algorithm based on region growing, threshold segmentation or active contour model) to the comprehensive image set to extract images of the tooth region and the jawbone region respectively.
And processing the tooth image by adopting an edge detection algorithm (such as Canny edge detection) to extract the outline characteristics of the teeth. And processing the jaw bone image by using a bone line extraction algorithm (such as morphological skeleton extraction) to extract bone line characteristics of the jaw bone.
And (3) carrying out three-dimensional construction by adopting a three-dimensional reconstruction algorithm (such as Marching Cubes algorithm based on volume rendering) according to the extracted tooth image and the extracted jawbone image, so as to generate a preliminary three-dimensional model of the tooth and the jawbone. And (3) carrying out grid optimization treatment on the preliminary three-dimensional model, optimizing the model surface by using a grid simplification and grid smoothing algorithm (such as Laplacian smoothing), reducing the number of polygons and improving the surface smoothness.
And according to the optimized three-dimensional grid model, texture mapping technology is adopted to carry out texture mapping on the surface of the model, and an initial three-dimensional model with real surface characteristics is obtained.
And S3, applying a feature point extraction algorithm to the tooth image and the jaw image in the preliminary three-dimensional model to respectively extract key feature point sets of the tooth and the jaw. And (3) performing rigid registration on the tooth image and the jawbone image according to the extracted feature point set by adopting an Iterative Closest Point (ICP) algorithm to obtain a preliminary alignment result.
Based on the preliminary alignment result, a non-rigid registration algorithm (such as a thin plate spline deformation algorithm) is adopted to carry out fine alignment on the tooth image and the jawbone image, so as to obtain a fine alignment result.
And (3) carrying out interpolation fusion on the tooth image and the jawbone image according to the fine alignment result, and obtaining a final oral fusion image by adopting a linear interpolation or higher interpolation method.
And S4, preprocessing the fused image of the oral cavity and the initial three-dimensional model, such as denoising, normalization and coordinate system unification, so that the registration algorithm can work effectively. And extracting key characteristic points in the oral fusion image and the initial three-dimensional model, and performing primary matching by adopting a characteristic point matching algorithm (such as SIFT, SURF and the like).
Initializing registration parameters, performing rigid registration based on the preliminarily matched feature points, and adjusting translation and rotation parameters so that the oral fusion image is approximately aligned with the initial three-dimensional model.
The non-rigid registration method is adopted for fine registration, and the method of combining global deformation and local deformation is utilized for further adjusting the image and the model to make the spatial position and the morphology of the image and the model highly consistent.
The registered oral model is checked to ensure spatial consistency of all critical structures, such as teeth and jawbone. If registration errors are found, the method of local optimization is adopted for correction, and the oral cavity model with high precision and high consistency is obtained.
And S5, carrying out track analysis according to the teeth and the jawbone data in the oral cavity model, and identifying key problems in the oral cavity of the patient, such as tooth dislocation, too large or too small tooth gaps, asymmetric jawbone and the like.
Based on the trajectory objective (e.g., aligning teeth, adjusting bite relationships, etc.), a preliminary trajectory path is generated.
And carrying out repeated iterative optimization on the preliminary track path, wherein each optimization is adjusted and improved based on the result of the previous step.
And simulating the track effect of each iteration by adopting a simulation technology, predicting an oral cavity model after track, and evaluating the difference between the oral cavity model and the target state.
And according to the simulation result, adjusting parameters and strategies of the optimization algorithm, and gradually approaching to an optimal track scheme.
After multiple iterative optimization, a track scheme with the best comprehensive effect is determined.
Comparing the final track scheme with the oral cavity model, verifying that the final track scheme meets the expected requirements in various aspects (such as functions, appearance, stability and the like), and finally generating a detailed optimized track scheme.
And S6, carrying out simulation track on the oral cavity model based on the optimized track scheme, wherein the simulation track comprises mechanical simulation, tissue response simulation and other modules.
And gradually adjusting the oral model according to the steps and the plan of the optimized track scheme. Each adjustment step requires precise displacement and rotation of the teeth and jaw, after each adjustment step, model data of the current state is recorded and its gap from the target state is evaluated.
In the process of simulating the track, mechanical analysis is carried out to evaluate the stress and strain conditions of the teeth and the jawbone under the action of track force.
And after all track steps are completed, evaluating a final simulated track result. The tracked oral model is checked for compliance with desired objectives including tooth alignment, bite relationship, jaw symmetry, etc.
And generating a comparison analysis report before and after the track, and displaying each key step and the final effect in the track process. And carrying out feedback analysis according to the simulation track result. If a larger gap is found between the simulation result and the expected result, the optimized track scheme is readjusted, and the simulation is performed again, so that a final simulation track result is obtained.
According to the three-dimensional image guiding correction planning method, the three-dimensional CT image and the two-dimensional X-ray image data are integrated, so that the characteristics of teeth and jawbone can be extracted more accurately, the construction precision of an oral cavity model is improved, and a more reliable basis is provided for track planning. By means of alignment fusion of the image data and registration analysis according to the fusion image, fine management of oral details is achieved, more accurate track paths are made, and unnecessary adjustment is avoided. The initial path is subjected to iterative optimization based on an optimized track scheme, so that the track process can be performed efficiently in different stages, unnecessary errors are reduced, and the efficiency of the whole track process is improved. By comprehensively analyzing the oral cavity model and the track path data, a more reasonable track strategy is formulated, and the track scheme is optimized by comprehensively adjusting, so that the track effect is effectively improved, and the treatment time and discomfort of a patient are reduced. And by considering the influence of individual differences on the track scheme, the track strategy can be flexibly adjusted according to the oral characteristics and the requirement change of different patients, so that the track process is more suitable for diversified requirement scenes.
In one embodiment, the integrating processing of the three-dimensional CT image and the two-dimensional X-ray image to obtain a corresponding integrated image set includes:
And acquiring a three-dimensional CT image generated by scanning the target area by the CT scanner. And the geometric deformation and the uneven intensity generated in the shooting process are eliminated through a geometric correction algorithm and an intensity balancing algorithm, so that a corrected three-dimensional CT image is obtained.
And acquiring a two-dimensional X-ray image generated by scanning the same target area by an X-ray machine. The two-dimensional X-ray image is corrected for distortion by applying image correction techniques such as nonlinear transformation and histogram equalization.
And performing alignment matching on the corrected three-dimensional CT image and the corrected two-dimensional X-ray image. Key feature points of the two images are detected and matched through feature point extraction and matching algorithms, such as SIFT (scale invariant feature transform) or SURF (speeded up robust feature). And performing spatial alignment on the three-dimensional CT image and the two-dimensional X-ray image according to the detected key feature points by using a rigid or non-rigid registration algorithm, so as to obtain a preliminary alignment image.
The multi-resolution pyramid decomposition is performed on the preliminary alignment images, including downsampling the preliminary alignment images layer by layer to generate a series of image levels of different resolutions. Each level represents details and structural information of the image at different scales.
And integrating all the image levels to form a multi-scale image set. The multi-scale image set contains image information of the target area under different resolutions.
And extracting a wavelet coefficient matrix by carrying out wavelet transformation on the multi-scale image set to obtain the characteristics of the image under different frequency components.
And carrying out feature fusion on the wavelet coefficient matrix, and carrying out comprehensive treatment on the features of different frequency components to obtain fusion feature representation.
And (3) performing inverse wavelet transform on the fused characteristic representation, and reducing the fused characteristic to an image form. And reconstructing a comprehensive image set containing all the characteristic information through inverse wavelet transformation. In the reconstruction process, an interpolation algorithm and a resampling technology are adopted to ensure the spatial resolution and quality of the image.
According to the embodiment, through the distortion correction of the three-dimensional CT image and the two-dimensional X-ray image, the accuracy of the image is improved, and errors caused by distortion in the acquisition process are reduced. And the corrected images are aligned and matched, so that the spatial consistency of multi-source image data is ensured, and the basic precision of image fusion is improved. Through multi-resolution pyramid decomposition and multi-scale image set integration, the method can capture image details and global structures under different scales, and the comprehensiveness and the fineness of image analysis are enhanced. The wavelet transformation and the feature extraction further extract the multi-frequency components of the image, so that the comprehensiveness and richness of the image features are ensured. Through feature fusion and inverse wavelet transformation, the generated comprehensive image set not only maintains key features of the original image, but also eliminates redundant information, and improves the quality and practicability of the image.
In one embodiment, feature extraction is performed on the integrated image set to obtain a dental image and a jaw image, and three-dimensional construction is performed according to the dental image and the jaw image to obtain an initial three-dimensional model, including:
And performing image segmentation processing on the comprehensive image set to separate dental images and jawbone images. Techniques for image segmentation processing include thresholding, region growing, graph cut algorithms, and the like.
On the basis of the obtained tooth image, a preset edge detection algorithm is adopted to extract the tooth profile characteristics, and the edge detection algorithm comprises Canny edge detection, sobel operator, laplacian operator and the like.
And extracting the characteristics of the jawbone skeleton line of the segmented jawbone image by adopting a preset skeleton line extraction algorithm, wherein the skeleton line extraction algorithm adopts a skeleton extraction method based on gradient or a skeleton extraction method based on morphology.
And carrying out three-dimensional reconstruction according to the extracted tooth profile information and the extracted jawbone skeleton line information to obtain a preliminary reconstruction model, wherein a surface reconstruction algorithm such as Marching Cubes algorithm, poisson surface reconstruction and the like is adopted in the three-dimensional reconstruction.
And carrying out smoothing treatment on the surface of the preliminary reconstruction model through a preset smoothing algorithm (Laplacian smoothing), so as to obtain an initial three-dimensional model.
According to the embodiment, through image segmentation processing and feature extraction, a tooth image and a jawbone image can be obtained respectively, so that finer separation and recognition are realized. This not only improves the accuracy of the data processing, but also provides a reliable basis for subsequent three-dimensional reconstruction. By adopting a preset edge detection algorithm and a bone line extraction algorithm, tooth profile and jaw bone line information can be efficiently obtained, and the three-dimensional reconstruction accuracy is ensured. By carrying out surface smoothing treatment on the preliminary reconstruction model, the attractiveness and the authenticity of the model are further improved, and errors in the subsequent design of the trackers are reduced. The accuracy of the three-dimensional model provides scientific basis for the establishment of a track scheme, and reduces risks and uncertainty in the track process.
In one embodiment, the tooth image and the jawbone image are aligned and fused to obtain a corresponding oral fusion image, which comprises:
and analyzing the tooth image according to the tooth profile information, and dividing each tooth from the whole tooth image to obtain the profile information of each single tooth.
And performing edge detection on the jawbone image according to the bone line information of the jawbone, and extracting the boundary information of the jawbone by using a gradient-based edge detection method.
Based on the obtained single tooth profile information and jaw boundary information, a corresponding relation matrix of the teeth and the jaw, namely a corresponding relation matrix of the teeth and the jaw, is constructed, and a jaw area corresponding to each tooth is calculated. The dental jaw correspondence matrix reflects the spatial relationship between the teeth and the jaw.
And carrying out rigid registration on the tooth image and the jaw image according to the tooth and jaw corresponding relation matrix to obtain a preliminary alignment result, wherein the purpose of the rigid registration is to enable the tooth image and the jaw image to be preliminarily aligned in space through rotation and translation operation.
On the basis of the initial alignment result, the image is subjected to non-rigid deformation treatment so as to obtain a fine alignment result, and the purpose of the non-rigid deformation treatment is to further adjust the tooth image and the jawbone image so that the tooth image and the jawbone image are more consistent in local detail.
And carrying out interpolation fusion on the two image data, namely the tooth image and the jawbone image, according to the fine alignment result and by adopting a tri-linear interpolation method to obtain an oral fusion image. The purpose of interpolation fusion is to fuse the tooth image and the jawbone image on the basis of the fine alignment result, so as to generate a unified oral cavity image. Interpolation fusion can also use nearest neighbor interpolation or methods based on image fusion algorithms.
The embodiment ensures that the spatial relationship between the teeth and the jawbone is accurately expressed through multi-step fine operations such as segmentation, edge detection, rigid registration, non-rigid deformation treatment and the like, so as to generate high-quality oral fusion images. Compared with the traditional single image processing mode, the method can display the internal structure of the oral cavity more comprehensively, helps doctors to make a more accurate track scheme, and improves the treatment effect and the comfort level of patients. In addition, the visual effect of the image is further improved through the interpolation fusion technology, so that a doctor can more intuitively observe and judge the image during planning and track treatment implementation, and the convenience and efficiency of clinical operation are improved. In a word, the method not only improves the accuracy and quality of image processing, but also brings significant clinical application value to the field of oral track.
In one embodiment, the registering processing is performed on the oral fusion image and the initial three-dimensional model based on a preset registering method to obtain a corresponding oral model, including:
And extracting characteristic points of the oral fusion image to obtain a first characteristic point set. The feature points of the first feature point set are key points in the image, such as edges of teeth or other anatomical feature points. And extracting the characteristic points of the initial three-dimensional model to obtain a second characteristic point set, wherein the characteristic points of the second characteristic point set correspond to the characteristic points of the first characteristic point set.
And performing rough registration on the oral fusion image and the initial three-dimensional model according to the first characteristic point set and the second characteristic point set, wherein the purpose of rough registration is to approximately align the two data sets so as to provide a good initial position. And (3) in the course of coarse registration, analyzing by an iterative nearest point algorithm to obtain a coarse registration result.
And carrying out fine registration on the oral fusion image and the initial three-dimensional model according to the coarse registration result, wherein the fine registration is carried out on the basis of an optimization algorithm, such as a least square method or other optimization methods for adjustment. By fine registration, a more accurate fine registration result is obtained.
And carrying out fusion processing on the oral fusion image and the initial three-dimensional model according to a method of combining the fine registration result with weighted average to obtain a fusion model. The fusion process involves combining two data sets together while retaining the advantages of each.
And carrying out boundary box analysis on the fusion model, and determining the external boundary of the fusion model to obtain a model boundary box.
And constructing a three-dimensional regular grid based on the model boundary box to obtain a model voxel space, wherein the three-dimensional regular grid divides the model into a plurality of small voxels.
And densely sampling the surface of the fusion model to obtain surface sampling points, wherein the surface sampling points are key points of the surface of the model. Detecting whether each voxel of the model voxel space contains a surface sampling point, labeling the voxels containing the surface sampling point as occupied, and thus determining which voxels are part of the model, is also possible.
And filling voxels of the model voxel space from the boundary to the inside according to a preset filling rule, and marking the internal voxels of the model voxel space as occupied, wherein the filling rule comprises a geometric shape-based filling rule, a voxel exclusion rule, a voxel density rule and a voxel morphology rule.
Wherein, the rule of voxel exclusion is to set voxels which are not filled under certain specific conditions. For example, voxels that are filled outside the model or marked as invalid regions are avoided.
And (3) a voxel density rule, namely controlling the density of filling voxels according to a preset density requirement. Ensuring that the filled model has uniform density distribution and avoiding overfilling or underfilling.
And the voxel morphology rule is to follow the natural morphology of the model, so that voxels which do not accord with the actual morphology are avoided being generated in the filling process. For example, the generation of protruding voxels on smooth surfaces is avoided.
And coding all occupied voxels according to the model voxel space to obtain a voxelized model, wherein the voxelized model is a combination of a plurality of small voxels represented by the three-dimensional model. And carrying out surface reconstruction on the voxelized model to obtain a final oral cavity model.
The embodiment provides a good initial alignment position through the steps of feature point extraction and coarse registration, and greatly reduces the calculation complexity and time of subsequent fine registration. The fine registration is based on an optimization algorithm, so that the two data sets can be aligned with high precision, and the accuracy of registration is improved. The fusion processing retains the advantages of respective data sets through weighted average and other technologies, and ensures the accuracy and the integrity of a fusion model. And the boundary box analysis and the three-dimensional regular grid construction enable the division of the model voxel space to be more standard and easier to process. And the steps of dense sampling and voxel marking not only ensure the integrity of the model, but also improve the reliability of data. The final voxelized model and surface reconstruction step provides a highly accurate oral model through accurate encoding and reconstruction algorithms. The embodiment provides reliable three-dimensional data support for the oral track, is beneficial to improving track effect and patient satisfaction, and remarkably improves convenience and efficiency of clinical application.
In one embodiment, track analysis is performed on a patient's mouth according to a mouth model to obtain an initial track path, and iterative optimization is performed on the initial track path to obtain an optimized track scheme, including:
The obtained oral cavity model is divided into a tooth unit and a jaw unit. The segmentation process can utilize image segmentation algorithms such as region growing and level set methods. After the segmentation is completed, position analysis is carried out on the tooth unit and the jaw bone unit, the relative position relation between the tooth unit and the jaw bone unit is determined, the position analysis comprises the steps of calculating geometric features such as center point coordinates, direction vectors and the like of each unit, and the spatial relation between the tooth unit and the jaw bone unit is established.
The tooth units are classified and marked according to the relative position relation between the tooth units and the jaw units so as to identify the type of each tooth, such as incisors, canines, premolars, molars and the like. The classification process is by employing a rule-based approach.
After the tooth type information is obtained, the tooth type information is analyzed and processed based on a preset tooth arrangement rule. The tooth arrangement rules are set based on the expertise of dentistry, including ideal tooth spacing, bite relationship, etc. An ideal arrangement position is determined for each tooth by the tooth arrangement rules.
The displacement vector of each tooth element from the current position to the ideal arrangement position is calculated, and the process involves coordinate transformation and vector calculation in three-dimensional space. And (3) comprehensively processing the displacement vectors of all the tooth units to obtain an initial track path. The initial trajectory path represents a preliminary movement scheme of the tooth from the current state to the ideal state.
And carrying out three-dimensional decomposition on each displacement vector to obtain component vectors in the directions of x, y and z, and carrying out stress analysis on the component vectors. This step also includes various forces that may be encountered during tooth movement, such as periodontal ligament resistance, adjacent tooth effects, and the like. And calculating by establishing a mechanical relation to obtain stress data corresponding to each component vector.
The three-dimensional component vector is adjusted based on the result of the force analysis, and the adjustment process comprises operations related to scaling, rotation or redirection of the vector. The purpose of the adjustment is to make the trajectory process smoother and more efficient while reducing the effects of unnecessary forces, the adjusted vector being called the optimized component vector.
Synthesizing all the optimized component vectors to obtain a new optimized displacement vector, wherein the optimized displacement vector forms an optimized track path.
The optimization process is repeatedly performed, and iteration is continuously performed until a preset termination condition is reached, such as that the displacement variation is smaller than a certain preset threshold value, the iteration times reach an upper limit, and the like. Finally, the personalized tooth track scheme of the patient is obtained, namely the optimized track scheme.
According to the embodiment, through accurate three-dimensional segmentation and position analysis of the oral cavity model, the relative position relation between the tooth unit and the jaw unit can be accurately obtained, so that a high-precision data base is provided for subsequent track planning. And secondly, based on a preset tooth arrangement rule and classification labels, the system can automatically generate ideal tooth arrangement positions, so that errors of manual operation are reduced, and the scientificity and rationality of a track scheme are improved. Furthermore, through three-dimensional decomposition and stress analysis of the initial track path, various mechanical problems possibly encountered in the tooth moving process can be comprehensively considered, and the track path is optimized, so that the track process is smoother and more efficient. Finally, the iterative optimization process ensures the continuous perfection of each track scheme, and the finally obtained optimized track scheme is more personalized and accurate.
In one embodiment, performing a simulated track on the oral model based on the optimized track scheme to obtain a corresponding simulated track result, including:
The method comprises the steps of carrying out segmentation processing on an oral cavity model according to an optimized track scheme, and supposing that the oral cavity model comprises tooth arrangement of a maxilla and a mandible, dividing the whole oral cavity model into a plurality of sub-models according to the optimized track scheme, wherein each sub-model corresponds to a part of teeth. Specifically, the maxilla may be divided into a left anterior tooth region, a right anterior tooth region, a left posterior tooth region, and a right posterior tooth region, and the mandible may be divided into a left anterior tooth region, a right anterior tooth region, a left posterior tooth region, and a right posterior tooth region. In this way, the entire oral model is decomposed into eight sub-models.
And carrying out stress analysis on each oral cavity submodel to obtain a corresponding stress distribution diagram. Stress analysis is performed on each sub-model to simulate the distribution of track forces in teeth and periodontal tissue. For example, the stress situation of the left anterior tooth area is analyzed to obtain the stress magnitude and distribution situation of each tooth in different directions in the area. Similarly, the same stress analysis is performed on the other seven sub-models to obtain respective stress distribution diagrams.
And carrying out deformation simulation on each oral cavity submodel based on the stress distribution diagram to obtain a deformation submodel. In this process, the deformation of the teeth under the action of the track force is simulated according to the stress distribution diagram. For example, the teeth in the left anterior tooth zone may be displaced and rotated under stress, and after deformation simulation, a deformation sub-model of the zone is obtained. The other seven sub-models are subjected to similar deformation simulation to obtain respective deformation sub-models.
And combining all the deformation submodels to obtain a track oral cavity integral model after simulating the track. Specifically, the eight deformation submodels are combined according to the actual structure of the oral cavity model to form a complete track oral cavity model, and the model shows the overall change condition of the oral cavity in the track process under the optimized track scheme.
And comparing and analyzing the track oral cavity integral model with a preset standard oral cavity model to obtain deviation data. By contrast analysis, the differences between the track oral cavity model and the standard oral cavity model are identified, and the deviation data particularly comprise the tooth positions, angles, arrangement and the like. And carrying out fine adjustment treatment on the track oral cavity integral model according to the deviation data. For example, if the position of a certain tooth deviates from the standard model, the position of the tooth is adjusted to the standard position through fine adjustment, and the fine adjustment is performed for a plurality of times, so that a fine adjustment oral cavity integral model is finally obtained.
And performing three-dimensional rendering treatment on the fine-tuning oral cavity integral model to obtain a simulation track result. In the process, the three-dimensional rendering software is utilized to render the trimmed oral cavity model, and a high-precision three-dimensional image is generated. The image can intuitively display the oral cavity model after the track, including the arrangement, the form and other information of the teeth.
According to the method, the oral cavity model is subjected to sectional processing and stress analysis based on the optimized track scheme, so that the distribution condition of track force in the oral cavity can be simulated more accurately, and the accuracy of the track scheme is improved. Through deformation simulation and combination treatment, the overall change of the oral cavity in the track process can be comprehensively displayed, so that the track result is more visual and intuitive. The track oral cavity integral model is compared with the standard model for analysis, and fine adjustment processing is carried out according to deviation data, so that errors possibly existing in the track process can be effectively reduced, and the accuracy of the track effect is improved. Finally, the high-precision three-dimensional image obtained through three-dimensional rendering processing provides visual reference for doctors, so that the method is more convenient and efficient in evaluating and optimizing the track scheme. In the whole, the method of the embodiment not only improves the accuracy and the visibility of the track scheme, but also remarkably improves the track effect and provides reliable decision basis for doctors.
Referring to fig. 2, the present invention further provides a three-dimensional image guiding correction planning system, which is applied to any one of the three-dimensional image guiding correction planning methods, including:
The acquisition module is used for acquiring a three-dimensional CT image and a two-dimensional X-ray image of the oral cavity of the patient, and integrating the three-dimensional CT image and the two-dimensional X-ray image to obtain a corresponding comprehensive image set;
The analysis module is used for extracting features of the comprehensive image set to obtain a tooth image and a jaw image, and carrying out three-dimensional construction according to the tooth image and the jaw image to obtain an initial three-dimensional model;
the association module is used for carrying out alignment fusion on the tooth image and the jawbone image to obtain a corresponding oral fusion image;
the processing module is used for carrying out registration processing on the oral fusion image and the initial three-dimensional model based on a preset registration method to obtain a corresponding oral model;
the control module is used for carrying out track analysis on the oral cavity of the patient according to the oral cavity model to obtain an initial track path, and carrying out iterative optimization on the initial track path to obtain an optimized track scheme.
The execution module is used for simulating the locus of the oral model based on the optimized locus scheme to obtain a corresponding simulated locus result
According to the three-dimensional image guiding correction planning system, the three-dimensional CT image and the two-dimensional X-ray image data are integrated, so that the characteristics of teeth and jawbone can be extracted more accurately, the construction precision of an oral cavity model is improved, and a more reliable foundation is provided for track planning. By means of alignment fusion of the image data and registration analysis according to the fusion image, fine management of oral details is achieved, more accurate track paths are made, and unnecessary adjustment is avoided. The initial path is subjected to iterative optimization based on an optimized track scheme, so that the track process can be performed efficiently in different stages, unnecessary errors are reduced, and the efficiency of the whole track process is improved. By comprehensively analyzing the oral cavity model and the track path data, a more reasonable track strategy is formulated, and the track scheme is optimized by comprehensively adjusting, so that the track effect is effectively improved, and the treatment time and discomfort of a patient are reduced. And by considering the influence of individual differences on the track scheme, the track strategy can be flexibly adjusted according to the oral characteristics and the requirement change of different patients, so that the track process is more suitable for diversified requirement scenes.
Referring to fig. 3, the present invention further provides a three-dimensional image guiding correction planning apparatus, including:
A memory for storing a program;
And the processor is used for executing the program to realize the steps of the three-dimensional image guiding correction planning method.
In this embodiment, the processor and the memory may be connected by a bus or other means. The memory may include volatile memory, such as random access memory, or nonvolatile memory, such as read only memory, flash memory, hard disk, or solid state disk. The processor may be a general-purpose processor, such as a central processing unit, a digital signal processor, an application specific integrated circuit, or one or more integrated circuits configured to implement embodiments of the present invention.
The invention also provides a storage medium storing computer instructions for causing a computer to perform the method of any one of the above.
It should be noted that, for convenience and brevity of description, the specific working process of the above-described system and each module may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes using the descriptions and drawings of the present invention or direct or indirect application in other related technical fields are included in the scope of the present invention.