CN-122023305-A - Automatic tooth grinding positioning 3D image construction method based on CT and oral cavity scanning

CN122023305ACN 122023305 ACN122023305 ACN 122023305ACN-122023305-A

Abstract

The invention discloses an automatic tooth grinding positioning 3D image construction method based on CT and oral cavity scanning, which comprises the core steps of acquiring CT and oral cavity scanning data, extracting core parameters, calculating elastic deformation correction coefficients and correcting data based on multiple parameters, calculating fusion weights by combining the correction coefficients with micro-morphology parameters and the like, distributing the fusion weights, calculating compensation coefficients based on mineralization degree and the like, fusing gray values, and extracting characteristic points to reconstruct a 3D image. The method solves the problem of insufficient positioning precision caused by time sequence difference elastic deformation and cross-equipment data isomerism in the prior art, realizes accurate positioning of molar teeth, and improves the construction reliability of 3D images.

Inventors

HE RUIXING
LI KELIANG
LI HUIWEN

Assignees

佛山市创昕医疗器械有限公司

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (10)

1. The automatic tooth grinding positioning 3D image construction method based on CT and oral cavity scanning is characterized by comprising the following steps of: S1, CT scanning data and intraoral scanning data are obtained, and tooth grinding position characteristic parameters, tooth grinding geometric characteristic parameters, periodontal ligament elastic modulus characteristic parameters, scanning time sequence difference parameters, image signal-to-noise ratio parameters and enamel-dentin interface micro-morphology characteristic parameters are extracted; S2, calculating a molar time sequence difference elastic deformation correction coefficient based on the molar position characteristic parameter, the molar geometric characteristic parameter, the periodontal ligament elastic modulus characteristic parameter and the scanning time sequence difference parameter, and carrying out elastic deformation correction on the CT scanning data and the intraoral scanning data based on the correction coefficient; S3, calculating an anisotropic fusion weight of the enamel-dentin interface based on the enamel-dentin interface micro-morphological feature parameter, the image signal to noise ratio parameter and the correction coefficient, and distributing weights of two groups of corrected data based on the fusion weight; And S4, calculating a nonlinear normalization compensation coefficient based on the tooth grinding mineralization degree characteristic parameter, the scanning angle parameter and the equipment gain parameter, combining the fusion weights to perform gray value fusion on the two groups of data, extracting tooth grinding characteristic points, reconstructing a 3D image, and outputting tooth grinding positioning 3D image data.
2. The method of claim 1, wherein the molar geometry parameters in S1 comprise molar crown height parameters, periodontal film thickness parameters, and the molar position parameters comprise proximal position parameters, distal position parameters, neutral position parameters.
3. The method according to claim 1, wherein S1 further comprises extracting a CT gray value raw parameter corresponding to CT scan data and an intra-oral scan gray value raw parameter corresponding to intra-oral scan data, and the image signal-to-noise ratio parameter comprises a CT scan data signal-to-noise ratio parameter and an intra-oral scan data signal-to-noise ratio parameter.
4. The method of claim 1, wherein the molar feature points in S4 comprise molar crown feature points, molar root feature points, enamel-dentin interface feature points.
5. The method according to claim 1, wherein the molar timing difference elastic deformation correction coefficient in S2 is calculated by a preset molar timing difference elastic deformation correction coefficient calculation formula, the calculation formula correlating the molar position characteristic parameter, the molar geometry characteristic parameter, the periodontal ligament elastic modulus characteristic parameter, and the scan timing difference parameter.
6. The method according to claim 5, wherein the enamel-dentin interface anisotropy fusion weight in S3 is calculated by a preset enamel-dentin interface anisotropy fusion weight calculation formula, and the calculation formula is related to the enamel-dentin interface micro-morphology feature parameter, the image signal-to-noise ratio parameter, and the molar timing difference elastic deformation correction coefficient.
7. The method of claim 6, wherein the nonlinear normalized compensation coefficient in S4 is calculated by a preset nonlinear normalized compensation coefficient calculation formula that correlates the molar mineralization level characteristic parameter, the scan angle parameter, and the device gain parameter.
8. The method according to claim 7, wherein the gray value fusion in S4 is calculated by a preset nonlinear normalized compensated fused gray value calculation formula, and the calculation formula correlates the nonlinear normalized compensation coefficient, the enamel-dentin interface anisotropic fusion weight, the CT gray value raw parameter, and the intra-oral scan gray value raw parameter.
9. The method according to claim 4, wherein the extracting of the molar feature points comprises the steps of extracting the molar crown feature points in the fused-on data, extracting the molar root feature points, extracting the enamel-dentin interface feature points, and extracting the molar geometric feature parameters according to claim 2.
10. The method of claim 1, wherein reconstructing the 3D image comprises creating a coordinate system with the fused data geometric center as an origin, mapping the extracted molar feature points to the coordinate system, constructing a mesh using a triangulation algorithm, performing texture mapping based on the mesh, and outputting molar positioning 3D image data.

Description

Automatic tooth grinding positioning 3D image construction method based on CT and oral cavity scanning Technical Field The invention relates to the technical field of stomatology image processing and computer vision, in particular to an automatic tooth grinding positioning 3D image construction method based on CT and oral scanning. Background In oral clinical diagnosis and repair, molar positioning 3D image construction based on CT and intra-oral scanning is a key basis of accurate diagnosis and treatment. In the prior art, when CT and intraoral scanning data are fused to construct a molar 3D image, the problem of periodontal ligament elastic deformation caused by time sequence difference of CT static scanning and intraoral dynamic scanning is not generally considered. The elastic deformation can cause distortion of scanned data, in the subsequent data fusion process, the prior art mostly adopts unified weight distribution and linear normalization treatment, and cannot adapt to anisotropic characteristics of enamel-dentin interface micro-morphology and heterogeneous characteristics of cross-equipment gray values, so that molar positioning accuracy is insufficient, and the clinical accurate diagnosis and treatment requirement is difficult to meet. Based on the above problems, a technical scheme for realizing high-precision tooth grinding positioning 3D image construction by solving the problems of elastic deformation and cross-device data fusion adaptability caused by time sequence difference is needed. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides an automatic tooth grinding positioning 3D image construction method based on CT and oral scanning, which comprises the following steps: S1, CT scanning data and intraoral scanning data are obtained, and tooth grinding position characteristic parameters, tooth grinding geometric characteristic parameters, periodontal ligament elastic modulus characteristic parameters, scanning time sequence difference parameters, image signal-to-noise ratio parameters and enamel-dentin interface micro-morphology characteristic parameters are extracted; S2, calculating a molar time sequence difference elastic deformation correction coefficient based on the molar position characteristic parameter, the molar geometric characteristic parameter, the periodontal ligament elastic modulus characteristic parameter and the scanning time sequence difference parameter, and carrying out elastic deformation correction on the CT scanning data and the intraoral scanning data based on the correction coefficient; S3, calculating an anisotropic fusion weight of the enamel-dentin interface based on the enamel-dentin interface micro-morphological feature parameter, the image signal to noise ratio parameter and the correction coefficient, and distributing weights of two groups of corrected data based on the fusion weight; And S4, calculating a nonlinear normalization compensation coefficient based on the tooth grinding mineralization degree characteristic parameter, the scanning angle parameter and the equipment gain parameter, combining the fusion weights to perform gray value fusion on the two groups of data, extracting tooth grinding characteristic points, reconstructing a 3D image, and outputting tooth grinding positioning 3D image data. Preferably, the molar geometry parameters in S1 include molar crown height parameters and periodontal film thickness parameters, and the molar position parameters include a proximal position parameter, a distal position parameter and a neutral position parameter. Further preferably, the step S1 further includes extracting a CT gray value original parameter corresponding to CT scan data and an intra-oral scan gray value original parameter corresponding to intra-oral scan data, where the image signal-to-noise ratio parameters include a CT scan data signal-to-noise ratio parameter and an intra-oral scan data signal-to-noise ratio parameter. Further preferably, the molar feature points in S4 include molar crown feature points, molar root feature points, enamel-dentin interface feature points. Further preferably, the molar time sequence difference elastic deformation correction coefficient in S2 is calculated by a preset molar time sequence difference elastic deformation correction coefficient calculation formula, and the calculation formula is related to the molar position characteristic parameter, the molar geometric characteristic parameter, the periodontal ligament elastic modulus characteristic parameter and the scanning time sequence difference parameter. Further preferably, the enamel-dentin interface anisotropic fusion weight in S3 is calculated by a preset enamel-dentin interface anisotropic fusion weight calculation formula, and the calculation formula is related to the enamel-dentin interface micro-morphology feature parameter, the image signal to noise ratio parameter and the molar time sequence difference ela