A method for precisely adjusting 3D printing orthodontic dental mold layer lines

Abstract

The application discloses a kind of based on 3D printing orthodontic dental model layer line precision adjustment method, it is related to odontology orthodontics and photocuring additive manufacturing technical field, the method includes following component part: S1, data acquisition and early processing step, S2, image selective processing step, S3, 3D printing step and S4, appliance production step;The present application is realized to the precise adjustment of 3D printing orthodontic dental model layer line by introducing layered identification optimization algorithm, multi-scale fusion profile extraction, gradient expansion based on multi-factor coupling formula and other advanced technologies, this method can be according to the specific tooth condition and the treatment requirement of patient, individual adjustment dental model layer line characteristics, such as layer line height, pitch, to significantly improve the precision and individual degree of orthodontic dental model, compared with traditional method, the present application can more accurately copy the anatomical form of tooth, provide more accurate basis for subsequent appliance production, help to improve the treatment effect.

Description

Technical Field

[0001] This invention relates to the field of dental orthodontics and light-cured additive manufacturing technology, specifically a method for precise adjustment of the texture of orthodontic dental mold layers based on 3D printing. Background Technology

[0002] In the field of orthodontics, orthodontic dental molds serve as the basis for the fabrication of orthodontic appliances. Their precision and personalization directly affect the orthodontic results and the patient's treatment experience. With the rapid development of 3D printing technology, light-curing additive manufacturing technology has been widely used in the fabrication of orthodontic dental molds.

[0003] Traditional 3D printing technology for orthodontic dental molds relies primarily on universal printing parameters and simple model processing, making it difficult to personalize them according to the patient's specific dental condition and orthodontic needs. This results in significant deficiencies in the precision and personalization of orthodontic dental molds, failing to accurately replicate the anatomical morphology of teeth, especially the fine structures and special requirements of the tooth surface. Furthermore, the inner surface of orthodontic appliances made using traditional methods is often relatively smooth, lacking the layered texture features that match the tooth surface. This leads to insufficient friction between the appliance and the teeth, affecting the effective transmission of orthodontic forces and reducing treatment efficiency. At the same time, due to the mismatch between the appliance and the tooth shape, patients may experience discomfort when wearing them, and even suffer soft tissue damage in the oral cavity, impacting the patient's treatment experience.

[0004] In view of the problems of insufficient precision, low degree of personalization and poor performance of orthodontic appliances in traditional 3D printing orthodontic dental model technology, the present invention proposes a method for precise adjustment of the layer texture of 3D printed orthodontic dental model, which is of particular importance. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for precise adjustment of the layer texture of 3D-printed orthodontic dental molds. By introducing advanced technologies such as layer recognition optimization algorithms, multi-scale fusion contour extraction, and gradient expansion based on multi-factor coupling formulas, this method achieves precise adjustment of the layer texture of 3D-printed orthodontic dental molds, significantly improving the accuracy and personalization of orthodontic dental molds. At the same time, through layer texture replication enhancement and adaptation optimization technology, the inner surface of the fabricated aligner can perfectly replicate the layer texture features of the orthodontic dental mold, enhancing the friction between the aligner and the tooth surface, improving treatment efficiency, and optimizing the patient's treatment experience.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for precise adjustment of the texture of orthodontic dental molds based on 3D printing, the specific steps of which are as follows:

[0007] S1. Data acquisition and preprocessing steps: A three-dimensional digital model of the patient's teeth is acquired by an intraoral scanner or a high-precision plaster model scanner. The three-dimensional digital model is then imported into 3D printing slicing software for layered slicing to obtain a series of two-dimensional slice images arranged in sequence.

[0008] S2. Selective image processing steps: In the slicing software, odd or even layers are selected as target layers to be processed by a preset layer recognition optimization algorithm. The images of the layers to be processed are then subjected to multi-scale fusion contour extraction, multi-step collaborative region division, gradient expansion based on multi-factor coupling formula, adaptive filtering optimization smoothing, and closed-loop control dynamic correction to obtain the expanded sliced ​​image.

[0009] S3, 3D printing steps: Combine the expanded slice image and the unprocessed slice image in layer order, and transmit them to the photopolymerization 3D printer. Through data adaptation and real-time monitoring technology or multi-nozzle collaborative control technology, photosensitive resin is cured layer by layer according to the image data to obtain an orthodontic dental model with regular concave and convex textures.

[0010] S4. Orthodontic appliance manufacturing steps: Based on the orthodontic dental model, the orthodontic appliance is manufactured using a hot pressing molding process through layer texture replication reinforcement and adaptation optimization technology. The inner surface of the orthodontic appliance replicates the layer texture features of the orthodontic dental model.

[0011] Furthermore, the layer recognition optimization algorithm in S2 achieves accurate screening of the layers to be processed by introducing a layer thickness adaptation factor and a dental arch curvature correction coefficient. The formula is as follows: ,in To identify the filtering results, The slice image's layer order index value. For layer thickness adaptation factor, This is the correction factor for the curvature of the dental arch. The current layer slice thickness, The layer thickness adaptation factor is the arch curvature value corresponding to the current layer. The source of the data was determined as follows: 1000 sets of dental mold printing data with different layer thicknesses were collected. A correlation model between layer thickness and screening accuracy was established through multiple linear regression analysis, and the model was used to determine the goodness of fit. The requirements are obtained by solving the problem. The range of values ​​for; dental arch curvature correction factor The data was determined by: obtaining dental arch curvature data from 500 patients with different dental arch morphologies through oral CT scans; combining this with the requirements of different dental arch morphologies for layer distribution in orthodontic clinical practice; and using the analytic hierarchy process (AHP) to quantify the influence weight of dental arch curvature, ultimately determining... The algorithm's implementation steps are as follows: First, extract the layer sequence index of each slice. slice thickness and the curvature of the corresponding dental arch region Then, substitute the parameters into the above formula to calculate the filter result identifier. Finally, based on The value of the marker indicates the layer to be processed and the unprocessed layer, ensuring that the screening process is not only based on the layer sequence, but also adaptable to different layer thicknesses and dental arch morphologies, thereby improving the targeting and accuracy of the screening and avoiding unreasonable layer distribution caused by differences in layer thickness or special dental arch morphologies.

[0012] Furthermore, in S2, the gradient expansion circle based on the multi-factor coupling formula is used to calculate the expansion circle pixel value using the multi-factor coupling formula, which is: ,in For the expanded pixel value, The corrective force affects the weight. To increase the target corrective force, Weights for tooth anatomical features. This represents the anatomical complexity coefficient of the current tooth position. This is a correction factor for the treatment phase. The corrective force influence weight is the value corresponding to the patient's current stage of treatment. The source of this determination is as follows: Based on orthodontic mechanics theory, the efficiency of tooth movement under different orthodontic forces was simulated using finite element method, combined with force-efficiency relationship data from 100 clinical orthodontic cases, and fitted using the least squares method. The optimal solution; weights of tooth anatomical features The determination was based on the following: analysis of the anatomical differences among the anterior teeth, premolars, and molars; invitation of 10 senior orthodontists to conduct Delphi scoring; and normalization of the scoring results. The numerical value; correction coefficient for the treatment stage. The determination of the correction coefficient is based on the following: dividing the orthodontic process into three stages—initial, adjustment, and stabilization—and establishing a mapping relationship between the stage and the correction coefficient by clinically tracking and observing the orthodontic data of 200 patients at each stage, ultimately determining the correction coefficient. The value of is determined by the following steps in implementing the formula: First, the patient's target corrective force increment is determined through clinical examination. Anatomical complexity coefficient of the current tooth position and the current stage of correction Then retrieve the preset , , Weights and coefficients; finally, substitute them into the formula to calculate the expanded pixel value. This ensures that the pixel value for expanding the orthodontic ring not only takes into account the orthodontic force, but also fully adapts to the anatomical features of the teeth and the stage of treatment, so as to achieve personalized and precise control of the amount of expansion.

[0013] Furthermore, the multi-step collaborative region segmentation in S2 is achieved through multi-step collaborative technology. The specific implementation steps are as follows: First, the selected images to be processed are preprocessed, including using an adaptive histogram equalization algorithm to enhance image contrast, remove Gaussian noise and salt-and-pepper noise from the image, and ensure that the boundaries between the tooth contour and the surrounding background are clear. Then, deep features of the image are extracted through a convolutional neural network, including crown edge features, gingival-tooth junction features, and occlusal surface texture features, to construct a multi-dimensional feature vector. Next, the feature vector is input into an improved U-Net semantic segmentation model, which enhances the feature extraction capability of key tooth regions by introducing an attention mechanism module. After that, morphological processing is performed on the segmentation results, including erosion and dilation operations, filling holes in the segmented area, and removing small isolated areas. Finally, the accuracy of the region segmentation is verified by comparing with a standard dental anatomy atlas. If the deviation exceeds a preset threshold, the model parameters are readjusted in the feature extraction stage until the region segmentation accuracy reaches the preset requirements, ensuring that the segmentation of the occlusal surface region, labial-buccal surface region, and gingival margin region is accurate and conforms to the dental anatomy, providing a reliable regional basis for differentiated expansion.

[0014] Furthermore, the contour extraction in S2, achieved through multi-scale fusion, is implemented using multi-scale fusion technology. The specific steps are as follows: First, a multi-scale Canny edge detection algorithm is used, setting three different scale parameters to perform edge detection on the image. The small scale is used to capture fine details of the tooth contour, the medium scale is used to extract the main structure of the contour, and the large scale is used to eliminate background noise interference. Then, the detection results at the three scales are weighted and fused. The weights are determined based on the detection accuracy of different scales, with the medium scale having the highest weight. The weights of the small and large scales are dynamically adjusted based on image quality. Next, contour extraction is performed on the fused edge image through 8-neighborhood connectivity analysis. Each pixel in the image is traversed, and connected edge pixel sets are marked to form an initial contour. Then, polygon approximation processing is performed on the initial contour to reduce redundant points and simplify the contour structure while retaining key feature points. Finally, a curve smoothing algorithm is used to optimize the approximated contour, eliminating jagged protrusions and ensuring the extracted tooth contour is continuous and smooth. This provides an accurate reference contour for subsequent expansion processing, avoiding expansion deviations caused by inaccurate contours.

[0015] Furthermore, the dynamic correction of the closed-loop control in S2 is achieved through closed-loop control technology. The specific implementation steps are as follows: First, after the expansion processing is completed, a real-time preview module is activated to perform pixel-level comparison analysis on the expanded slice image, calculate the spacing value of adjacent teeth after expansion, and determine whether there is a contour intersection problem. If contour intersection is detected, the system automatically locates the coordinate range of the intersection area and extracts the tooth contour features of that area, including contour curvature and the relative positional relationship of adjacent teeth. Then, based on the features of the intersection area, a correction algorithm is activated, decreasing the expansion pixel value of the intersection area by a preset step size. After each reduction, the spacing between adjacent teeth is recalculated. Simultaneously, the impact of the corrected expansion contour on the transmission of orthodontic force is simulated using finite element analysis to ensure that the orthodontic force is not excessively reduced during the correction process. If the spacing still does not reach the safe threshold after one reduction, the above reduction and simulation steps are repeated until the spacing is ≥ the safe threshold. Finally, a detailed difference report is generated, marking the original parameters, adjusted parameters, and simulation verification results of the intersection area, and sent to the doctor's terminal. After the doctor confirms that there are no errors, the corrected image data is stored in the printing queue to ensure that the expansion process avoids contour intersections and does not affect the effective transmission of orthodontic force.

[0016] Furthermore, the smoothing process of adaptive filtering optimization in S2 is achieved through adaptive filtering optimization technology. The specific implementation steps are as follows: First, pixel-level analysis is performed on the expanded contour image to extract the curvature value of each pixel on the contour. Based on the curvature value, the contour is divided into smooth regions and abrupt change regions. Smooth regions are areas with small and stable curvature values, while abrupt change regions are areas with large or rapidly changing curvature values. Then, differentiated filtering parameters are selected for different regions. For smooth regions, a larger Gaussian filter kernel is used to ensure the overall smoothness of the contour. For abrupt change regions, a smaller Gaussian filter kernel is used, and an edge preservation factor is introduced to avoid losing key abrupt changes in the contour during the filtering process. Next, the filter kernel is applied to the contour image, and a weighted average is calculated for each pixel. The weights are determined based on the distance between the pixel and its neighboring pixels and the difference in gray level. The closer the distance and the smaller the gray level difference, the greater the weight. After filtering, the contour is resampled to evenly distribute the pixels on the contour, ensuring the continuity and consistency of the contour. Finally, the effect of smoothing is verified by calculating the similarity of the contour before and after filtering and the offset of key feature points. If the similarity is lower than the preset threshold or the offset of feature points is too large, the filtering parameters are readjusted and filtering is performed again until the contour has no jagged edges and retains the original anatomical features of the teeth, providing accurate and smooth contour data for subsequent 3D printing.

[0017] Furthermore, the specific implementation steps of the data adaptation and real-time monitoring technology in S3 for 3D printing are as follows: First, the slice image data after layer-by-layer combination is converted to a standard format recognizable by the photopolymerization 3D printer. Simultaneously, the data integrity is checked to identify any data loss, image distortion, or other issues. If any problems are found, the original data is automatically retransmitted and recombined. Then, based on the processing type of the slice image, a preset printing parameter library is called. The parameters in the library are optimized through numerous printing experiments to ensure that different types of slice layers can achieve good curing results. During the printing process, the printer's real-time monitoring module is activated, using a high-speed camera... The curing process of each layer is filmed, and data such as the height and contour accuracy of the cured layer are collected. The collected data is compared with the design values ​​in real time, and the deviation value is calculated. If the deviation value exceeds the allowable range, the system automatically analyzes the cause of the deviation. If it is due to insufficient curing caused by insufficient exposure time, the exposure time of the corresponding layer is appropriately extended. If it is due to layer deviation caused by resin liquid level fluctuation, the liquid level stabilization device of the printer is adjusted. After printing, the dental model is 3D scanned to obtain the actual layer data, which is comprehensively compared with the design layer data to generate a printing accuracy report. This provides a reference for subsequent orthodontic appliance fabrication, ensuring that the printed orthodontic dental model layer pattern meets the design requirements and the accuracy meets clinical application standards.

[0018] Furthermore, the specific implementation steps of the multi-nozzle collaborative control technology in S3 to achieve layer texture adjustment of the dental mold are as follows: First, the nozzles of the multi-nozzle photopolymerization 3D printer are calibrated, including position calibration, jet volume calibration, and jet speed calibration, to ensure that the jet parameters of each nozzle are consistent and to avoid uneven layer texture caused by nozzle deviation; then, according to the layer sequence index of the slice image, the printing task is assigned to two groups of nozzles, one group of nozzles is responsible for printing the layer to be processed, and the other group of nozzles is responsible for printing the unprocessed layer; for the nozzle responsible for the layer to be processed, the program controls it to increase the jet volume when jetting photosensitive resin according to the preset layer texture height requirement. The increased jet volume is determined through previous printing experiments to ensure that the height of the formed protruding layer texture meets the design requirements; during printing... During printing, the spraying status of the two sets of nozzles is monitored in real time. Sensors collect data on the spraying pressure and resin flow rate of the nozzles. If abnormal data is detected, the system automatically pauses printing, adjusts the nozzle parameters, and then resumes. Simultaneously, a high-speed camera is used to capture the forming effect of each layer, observe the shape and height of the raised layers, and compare them with the design values. If deviations are found, they are corrected by adjusting the additional spraying volume or spraying speed of the nozzles. After printing, the layer structure of the dental model is inspected, including indicators such as layer height difference and spacing uniformity, to ensure that the layer texture meets the requirements for orthodontic appliance fabrication. This technology, through the coordinated work of the nozzles, can achieve precise adjustment of the dental model layer texture without the need for complex enlargement processing of the sliced ​​images, providing more diverse implementation paths for technical solutions.

[0019] Furthermore, the specific implementation steps of the layer texture replication enhancement and adaptation optimization technology in S4 for orthodontic appliance fabrication are as follows: First, the printed orthodontic dental model undergoes surface pretreatment to remove residual resin and supporting structures from the model surface. A fine polishing tool is used to lightly polish the layer texture surface to ensure its integrity and smoothness, avoiding sharp protrusions that could scratch the appliance material. Then, the pretreated dental model is fixed on the mold base of a thermoforming machine. Based on the characteristics of the appliance material, appropriate thermoforming temperature, pressure, and time parameters are set, and the thermoforming process is initiated to ensure that the appliance material fully conforms to the dental model surface and accurately replicates the layer texture features. After thermoforming, the appliance is allowed to cool to room temperature. The aligner is removed from the dental mold, and its edges are trimmed to remove excess material, ensuring smooth edges and a good fit to the gum line. Next, a 3D optical scanner is used to scan the inner surface of the aligner, obtaining the actual parameters of layer depth and spacing, which are compared with the design parameters. If deviations are found, the cause is analyzed. If incomplete layer replication is due to improper hot-pressing parameters, the hot-pressing temperature, pressure, or time is adjusted, and the aligner is remade. Finally, the finished aligner is sterilized, and necessary fine-tuning is made according to the patient's oral fit to ensure comfortable wear and effective contact between the layers and the tooth surface, enhancing friction and improving orthodontic force.

[0020] Compared with existing technologies, this method for precise adjustment of the texture of orthodontic dental molds based on 3D printing has the following advantages:

[0021] I. This invention achieves precise adjustment of the layer texture of 3D-printed orthodontic dental molds by introducing advanced technologies such as layer recognition optimization algorithm, multi-scale fusion contour extraction, and gradient expansion based on multi-factor coupling formula. This method can personalize the layer texture features of the dental mold, such as layer height and spacing, according to the patient's specific dental condition and orthodontic needs, thereby significantly improving the accuracy and personalization of orthodontic dental molds. Compared with traditional methods, this invention can more accurately replicate the anatomical morphology of teeth, providing a more accurate basis for the subsequent fabrication of orthodontic appliances and helping to improve the orthodontic effect.

[0022] Second, this invention utilizes layer texture replication enhancement and adaptation optimization technology. The inner surface of the orthodontic appliance manufactured by this invention can perfectly replicate the layer texture characteristics of the orthodontic dental model. This design significantly enhances the friction between the appliance and the tooth surface, which helps to transmit orthodontic force more effectively and improves orthodontic efficiency. At the same time, because the layer texture characteristics of the inner surface of the appliance are highly matched with the patient's tooth morphology, it is more comfortable to wear, reducing the discomfort and soft tissue damage to the oral cavity that may be caused by traditional orthodontic appliances, thereby optimizing the patient's orthodontic experience. In addition, this method also ensures high precision and stability in the dental model and appliance manufacturing process through closed-loop control dynamic correction and adaptive filtering optimization smoothing technology, further improving the overall performance of the appliance.

[0023] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0025] Figure 1 A flowchart of a method for precise adjustment of the texture of orthodontic dental molds based on 3D printing;

[0026] Figure 2 This is a flowchart illustrating a method for precise adjustment of the texture of orthodontic dental mold layers based on 3D printing.

[0027] Figure 3 This is a flowchart for a health assessment method based on a precise adjustment method for the texture of orthodontic dental molds using 3D printing. Detailed Implementation

[0028] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0029] Example 1

[0030] A comprehensive oral scan was performed on a 14-year-old adolescent patient, accurately acquiring a 3D digital model of all teeth, including crowded areas of the anterior teeth. This ensured the model clearly represented the arrangement of the anterior teeth, crown morphology, and gingival margins. The acquired 3D digital model was imported into professional 3D printing slicing software, and layered slicing was performed according to layer thickness parameters adapted to the adolescent's dental arch morphology. The final result was a series of 2D slice images arranged sequentially, with each slice fully preserving the relative position and contour details of the anterior teeth and adjacent teeth. Figure 1 As shown.

[0031] In the slicing software, activate the preset layer recognition optimization algorithm, introducing a layer thickness adaptation factor and an arch curvature correction coefficient, using the following formula: ,in To identify the filtering results, The slice image's layer order index value. For layer thickness adaptation factor, This is the correction factor for the curvature of the dental arch. The current layer slice thickness, Given the arch curvature value corresponding to the current layer, and considering the layer thickness characteristics and arch curvature of the patient's anterior teeth region, odd-numbered layers are selected as target layers to be processed. First, multi-scale contour extraction is performed. Using the Canny edge detection algorithm at three different scales, fine edge textures and the main contour structure of the anterior teeth are captured while background noise is eliminated. After weighted fusion, the initial contour is extracted through 8-neighborhood connectivity analysis. Then, the structure is simplified using polygon approximation while retaining key feature points. Finally, a curve smoothing algorithm is used to eliminate jagged protrusions on the contour.

[0032] Next, a multi-step collaborative region segmentation process is performed. First, an adaptive histogram equalization algorithm is used to enhance image contrast and remove Gaussian and salt-and-pepper noise. Then, a convolutional neural network is used to extract deep features such as the anterior tooth crown margin, gingival-teeth junction, and occlusal texture. A multi-dimensional feature vector is constructed and input into an improved U-Net semantic segmentation model with an attention mechanism module. The segmentation results are subjected to erosion and dilation operations to fill holes and remove isolated areas. Finally, the segmentation is compared with a standard anterior tooth anatomical atlas to ensure accurate segmentation of the crowded anterior teeth. Subsequently, gradient expansion is performed based on a multi-factor coupling formula. Taking into account the target orthodontic force increment for the patient's crowded anterior teeth, the anatomical complexity coefficient of the anterior teeth, and the stage characteristics of the initial treatment period, personalized expansion pixel values ​​are calculated to achieve targeted expansion. The formula is as follows: ,in For the expanded pixel value, The corrective force affects the weight. To increase the target corrective force, Weights for tooth anatomical features. This represents the anatomical complexity coefficient of the current tooth position. This is a correction factor for the treatment phase. For the patient's current orthodontic stage, after expansion, adaptive filtering optimization is applied for smoothing. Pixel-level analysis divides the anterior tooth contour into smooth and abrupt regions. A larger Gaussian filter kernel is used for smooth regions, while a smaller Gaussian filter kernel is used for abrupt regions, and an edge preservation factor is introduced. After filtering and resampling, contour similarity and feature point offset are verified to ensure smooth contours while preserving original anatomical features. Finally, closed-loop dynamic correction is initiated, performing pixel-level comparisons on the expanded slice images to detect contour intersections between anterior teeth. If intersections are found, the intersection area is located, contour features are extracted, and the expansion pixel value is decreased by a preset step size. Simultaneously, finite element simulation is used to simulate the impact of orthodontic force transmission until the distance between adjacent anterior teeth reaches a safe threshold. A difference report is then generated and sent to the doctor's terminal for confirmation.

[0033] The expanded slice images are combined with the unprocessed slice images in their original layer order. The combined data is then format-converted to a standard format recognizable by the photopolymerization 3D printer. Simultaneously, data integrity is verified, checking for data loss, image distortion, and other issues. If any anomalies are found, the original data is automatically retransmitted and recombined. Based on the processing type of the slice images, a preset printing parameter library is invoked. This parameter library has been optimized through numerous experiments printing dental molds for adolescent anterior orthodontic treatment. Figure 2 As shown, during the printing process, a real-time monitoring module is activated, using a high-speed camera to capture the curing process of each layer in real time, collecting data such as layer height and contour accuracy, and comparing them with the design values ​​in real time. When insufficient curing of certain layers is detected due to insufficient exposure time, the system automatically extends the exposure time of the corresponding layers; when resin level fluctuations are detected causing layer deviations, the printer's liquid level stabilization device is adjusted promptly. After printing, the orthodontic dental model is 3D scanned to obtain actual layer data and comprehensively compare it with the design data, generating a detailed printing accuracy report to confirm that the dental model accurately reproduces the anatomical morphology of crowded anterior teeth and the designed layer features.

[0034] The printed orthodontic dental model undergoes surface pretreatment to remove residual photosensitive resin and support structures. A fine polishing tool is used to lightly polish the layered surface, ensuring the layers are intact and free of sharp protrusions to avoid scratching the aligner material. The pretreated model is then fixed onto the mold base of a thermoforming machine. Based on the characteristics of the aligner material, the thermoforming temperature, pressure, and time parameters are set to adapt to the replication of the anterior tooth layered texture. The thermoforming process is then initiated to ensure the aligner material fully adheres to the model surface. After the aligner cools to room temperature, it is removed, and the edges are finely trimmed to remove excess material, ensuring a smooth edge that conforms to the gingival morphology. A 3D optical scanner is used to scan the layered texture on the inner surface of the aligner, obtaining actual parameters such as layer depth and spacing, which are compared with the design parameters. Simultaneously, the conformity is verified according to the health assessment process. Figure 3 As shown, after confirming that there are no deviations, the orthodontic appliance is strictly disinfected, and the patient is allowed to try it on. Minor adjustments are made according to the oral fit to ensure that the texture of the inner surface of the appliance is in full contact with the surface of the anterior teeth, thereby increasing friction to effectively transmit the orthodontic force and improving the treatment effect of anterior crowding.

[0035] Example 2

[0036] A high-precision intraoral scanner was used to scan the oral cavity of a 28-year-old adult patient, focusing on acquiring three-dimensional data of the rotated molars and surrounding teeth. The rotation angle, crown morphology, and relative positional relationship with adjacent teeth were precisely captured to construct a full-mouth three-dimensional digital model of the teeth. This three-dimensional digital model was imported into 3D printing slicing software, and layered slicing was performed according to the layer thickness parameters adapted to the anatomical structure of the molar region. This resulted in a series of two-dimensional slice images arranged in a layered sequence, ensuring that the irregular contour of the rotated molars and their positional relationship with adjacent teeth were accurately represented in the slice images. Figure 1 As shown.

[0037] A layer recognition optimization algorithm was run in the slicing software. Combining the layer thickness adaptation factor and the arch curvature correction coefficient of the patient's molar region, even-numbered layers were selected as the target layers to be processed. First, multi-scale fusion contour extraction was performed. The Canny edge detection algorithm at three different scales was used to capture the irregular contour details and main structure of the twisted molar and remove background interference. After weighted fusion, the initial contour was extracted by 8-neighborhood connectivity analysis. The structure was simplified by polygon approximation and key feature points were retained. Then, the jagged protrusions on the contour were eliminated by curve smoothing algorithm. Next, multi-step collaborative region segmentation was performed. First, the adaptive histogram equalization algorithm was used to enhance image contrast and remove Gaussian noise and salt-and-pepper noise. Then, the complex texture features of the crown edge, gingival-teeth junction and occlusal surface of the molar were extracted by convolutional neural network. Multi-dimensional feature vectors were constructed and input into the improved U-Net semantic segmentation model with an attention mechanism module. The segmentation results were subjected to erosion and dilation operations to fill holes and remove isolated areas. Finally, the results were compared with the standard molar anatomical atlas to ensure the accuracy of the segmentation of the twisted molar region. If the deviation exceeded the preset threshold, the model parameters were readjusted. Subsequently, gradient expansion was performed based on a multi-factor coupling formula. This comprehensively considered the patient's target orthodontic force increment for the rotated molar, the high anatomical complexity coefficient of the molar, and the stage characteristics of the orthodontic adjustment period, calculating personalized expansion pixel values ​​for targeted expansion. After expansion, adaptive filtering optimization was applied for smoothing. A larger Gaussian filter kernel was used for smooth areas of the molar contour, while a smaller Gaussian filter kernel was used for abrupt changes such as the occlusal surface, and an edge preservation factor was introduced. After filtering and resampling, contour similarity and feature point offset were verified to ensure the contour was jagged and the anatomical features of the molar were fully preserved. Finally, closed-loop control was activated for dynamic correction. Pixel-level comparisons were performed on the expanded slice images to detect contour intersections between the rotated molar and adjacent teeth. If intersections were found, the intersection area was located, and features such as contour curvature and relative positional relationships were extracted. The expansion pixel value was decreased incrementally by a preset step size. Simultaneously, the influence of orthodontic force transmission was simulated through finite element simulation until the distance between adjacent teeth reached a safe threshold. A difference report was then generated and sent to the doctor's terminal for confirmation.

[0038] A multi-nozzle photopolymer 3D printer was used. First, both sets of nozzles underwent comprehensive calibration, including position calibration, ejection volume calibration, and ejection speed calibration, to ensure consistent ejection parameters and prevent uneven layer textures due to nozzle deviation. Based on the layer sequence index of the sliced ​​image, the printing tasks were assigned to the two sets of nozzles. One set was responsible for printing the processed even-numbered layers, and the other for printing the unprocessed layers. For the nozzles responsible for the processed layers, the program controlled them to increase the ejection volume of photosensitive resin according to the preset layer height requirements. This additional ejection volume was determined through previous printing experiments on orthodontic dental models for molars. Figure 2As shown, during the printing process, sensors collect real-time data on the jet pressure and resin flow rate of the two sets of nozzles. If any abnormal data is detected, the system automatically pauses printing, adjusts the nozzle parameters, and then resumes. Simultaneously, a high-speed camera captures real-time images of each layer's forming effect, observing the shape and height of the raised layers and comparing them with the design values. If deviations are found, corrections are made by adjusting the additional jet volume or jet speed of the nozzles. After printing, the layer structure of the dental model undergoes a specialized inspection, focusing on verifying the uniformity of layer height difference and spacing to ensure it meets the requirements for manufacturing molar rotation orthodontic appliances.

[0039] The printed orthodontic dental model undergoes surface pretreatment to remove residual photosensitive resin and support structures. The surface of the textured layer is gently polished with a fine polishing tool to ensure its integrity and smoothness, avoiding any impact on the replication of the orthodontic appliance's texture. The pretreated model is then fixed onto the mold base of a thermoforming machine. Based on the characteristics of the appliance material and the complexity of the molar texture, appropriate thermoforming temperature, pressure, and time parameters are set, and the thermoforming process is initiated to ensure the appliance material fully adheres to the model surface, accurately replicating the texture features. After the appliance cools to room temperature, it is removed, and the edges are finely trimmed to remove excess material, ensuring a smooth edge that conforms to the gingival morphology. A 3D optical scanner is used to scan the texture of the inner surface of the appliance, obtaining actual parameters such as texture depth and spacing, which are compared with the design parameters. A pass / fail assessment is then performed according to the health assessment process. Figure 3 As shown, if the layer replication is incomplete due to hot-pressing parameters, the hot-pressing temperature, pressure, or time should be adjusted and the appliance remade. After confirming that the layer parameters meet the requirements, the appliance should be strictly sterilized. The patient should try on the appliance, and necessary fine-tuning should be made according to the fit in the molar area to ensure the appliance is comfortable to wear and that the inner surface layer conforms closely to the surface of the rotated molars, effectively transmitting corrective force and promoting the gradual return of the molars to their normal position.

[0040] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for precise adjustment of the texture of orthodontic dental mold layers based on 3D printing, characterized in that, The specific steps of this method are as follows: S1. Data acquisition and preprocessing steps: A three-dimensional digital model of the patient's teeth is acquired by an intraoral scanner or a high-precision plaster model scanner. The three-dimensional digital model is then imported into 3D printing slicing software for layered slicing to obtain a series of two-dimensional slice images arranged in sequence. S2. Selective image processing steps: In the slicing software, odd or even layers are selected as target layers to be processed by a preset layer recognition optimization algorithm. The images of the layers to be processed are then subjected to multi-scale fusion contour extraction, multi-step collaborative region division, gradient expansion based on multi-factor coupling formula, adaptive filtering optimization smoothing, and closed-loop control dynamic correction to obtain the expanded sliced ​​image. S3, 3D printing steps: Combine the expanded slice image and the unprocessed slice image in layer order, and transmit them to the photopolymerization 3D printer. Through data adaptation and real-time monitoring technology or multi-nozzle collaborative control technology, photosensitive resin is cured layer by layer according to the image data to obtain an orthodontic dental model with regular concave and convex textures. S4. Orthodontic appliance manufacturing steps: Based on the orthodontic dental model, the orthodontic appliance is manufactured using a hot pressing molding process through layer texture replication reinforcement and adaptation optimization technology. The inner surface of the orthodontic appliance replicates the layer texture features of the orthodontic dental model.

2. The method for precise adjustment of the texture of orthodontic dental molds based on 3D printing according to claim 1, characterized in that, The layer recognition optimization algorithm in S2 introduces a layer thickness adaptation factor and a dental arch curvature correction coefficient, and the formula is as follows: ,in To identify the filtering results, The slice image's layer order index value. For layer thickness adaptation factor, This is the correction factor for the curvature of the dental arch. The current layer slice thickness, This represents the arch curvature value corresponding to the current layer.

3. The method for precise adjustment of the texture of orthodontic dental molds based on 3D printing according to claim 1, characterized in that, In S2, the gradient expansion based on a multi-factor coupling formula is used to calculate the expanded pixel value using the multi-factor coupling formula, which is: ,in For the expanded pixel value, The corrective force affects the weight. To increase the target corrective force, Weights for tooth anatomical features. This represents the anatomical complexity coefficient of the current tooth position. This is a correction factor for the treatment phase. This represents the patient's current stage of treatment.

4. The method for precise adjustment of the texture of orthodontic dental mold layers based on 3D printing according to claim 1, characterized in that, The multi-step collaborative region segmentation in S2 is achieved through multi-step collaborative technology. The specific implementation steps are as follows: First, the selected images to be processed are preprocessed, including using an adaptive histogram equalization algorithm to enhance image contrast and remove Gaussian noise and salt-and-pepper noise from the images. Then, deep features of the images are extracted through a convolutional neural network, including crown edge features, gingival-teeth junction features, and occlusal surface texture features, to construct a multi-dimensional feature vector. Next, the feature vector is input into an improved U-Net semantic segmentation model, which enhances the feature extraction capability of key areas of the teeth by introducing an attention mechanism module. After that, morphological processing is performed on the segmentation results, including erosion and dilation operations, filling holes in the segmented areas, and removing small isolated areas. Finally, the accuracy of the region segmentation is verified by comparing it with a standard dental anatomy atlas. If the deviation exceeds a preset threshold, the model parameters are readjusted in the feature extraction stage until the region segmentation accuracy reaches the preset requirements.

5. The method for precise adjustment of the texture of orthodontic dental mold layers based on 3D printing according to claim 1, characterized in that, The contour extraction in S2, achieved through multi-scale fusion, is implemented using a multi-scale fusion technique. The specific steps are as follows: First, a multi-scale Canny edge detection algorithm is used, setting three different scale parameters to perform edge detection on the image. The small scale is used to capture fine details of the tooth contour, the medium scale to extract the main structure of the contour, and the large scale to eliminate background noise interference. Then, the detection results at the three scales are weighted and fused. The weights are determined based on the detection accuracy at different scales, with the medium scale having the highest weight. The weights of the small and large scales are dynamically adjusted based on image quality. Next, contour extraction is performed on the fused edge image using 8-neighborhood connectivity analysis. Each pixel in the image is traversed, and connected edge pixel sets are marked to form an initial contour. Then, polygon approximation processing is performed on the initial contour while retaining key feature points. Finally, a curve smoothing algorithm is used to optimize the approximated contour, eliminating jagged protrusions on the contour.

6. The method for precise adjustment of the texture of orthodontic dental molds based on 3D printing according to claim 1, characterized in that, The dynamic correction of the closed-loop control in S2 is achieved through closed-loop control technology. The specific implementation steps are as follows: First, after the expansion processing is completed, the real-time preview module is started to perform pixel-level comparison and analysis on the expanded slice image, calculate the spacing value of adjacent teeth after expansion, and determine whether there is a contour intersection problem; if contour intersection is detected, the system automatically locates the coordinate range of the intersection area and extracts the tooth contour features of the area. Then, based on the characteristics of the intersection area, the correction algorithm is activated, and the expansion pixel value of the intersection area is decreased step by step according to the preset step size. After each decrease, the distance between adjacent teeth is recalculated. At the same time, the influence of the corrected expansion contour on the orthodontic force transmission is simulated through finite element simulation. Finally, a detailed difference report is generated, marking the original parameters, adjusted parameters and simulation verification results of the intersection area, and sent to the doctor's terminal.

7. The method for precise adjustment of the texture of orthodontic dental mold layers based on 3D printing according to claim 1, characterized in that, The smoothing process in S2, achieved through adaptive filtering optimization, is implemented using the following steps: First, pixel-level analysis is performed on the expanded contour image to extract the curvature value of each pixel on the contour. Based on the curvature value, the contour is divided into smooth regions and abrupt change regions. Smooth regions are those with small curvature values ​​and stable changes, while abrupt change regions are those with large curvature values ​​or rapid changes. Then, differentiated filtering parameters are selected for different regions. For smooth regions, a larger Gaussian filter kernel is used; for abrupt change regions, a smaller Gaussian filter kernel is used, and an edge preservation factor is introduced. Next, the filter kernel is applied to the contour image, and a weighted average is calculated for each pixel. After filtering, the contour is resampled to evenly distribute the pixels on the contour. Finally, the smoothing effect is verified by calculating the similarity of the contour before and after filtering and the offset of key feature points. If the similarity is lower than a preset threshold or the feature point offset is too large, the filtering parameters are readjusted, and filtering is performed again until the contour has no jagged edges.

8. The method for precise adjustment of the texture of orthodontic dental molds based on 3D printing according to claim 1, characterized in that, The specific implementation steps of the data adaptation and real-time monitoring technology in S3 for 3D printing are as follows: First, the slice image data after layer-by-layer combination is converted into a standard format that can be recognized by the photopolymerization 3D printer. At the same time, the data integrity is checked to see if there are any problems such as data loss or image distortion. If there are any problems, the original data is automatically retransmitted and recombined. Then, according to the processing type of the slice image, the preset printing parameter library is called. During the printing process, the real-time monitoring module of the printer is started, and the curing process of each layer is captured by a high-speed camera to collect data such as the height of the cured layer texture and the contour accuracy. The collected data is compared with the design value in real time to calculate the deviation value. If the deviation value exceeds the allowable range, the system automatically analyzes the cause of the deviation. If it is due to insufficient curing caused by insufficient exposure time, the exposure time of the corresponding layer is appropriately extended. If it is due to layer texture deviation caused by resin liquid level fluctuation, the liquid level stabilization device of the printer is adjusted. After printing, the dental model is 3D scanned to obtain the actual layer texture data, which is then compared with the designed layer texture data to generate a printing accuracy report.

9. The method for precise adjustment of the texture of orthodontic dental mold based on 3D printing according to claim 1, characterized in that, The specific implementation steps of the multi-nozzle collaborative control technology in S3 to achieve dental mold layer texture adjustment are as follows: First, the nozzles of the multi-nozzle photopolymerization 3D printer are calibrated; then, according to the layer sequence index of the slice image, the printing task is assigned to two groups of nozzles, one group of nozzles is responsible for printing the layer to be processed, and the other group of nozzles is responsible for printing the unprocessed layer. For the printheads responsible for the layers to be processed, the program controls them to increase the spray volume of photosensitive resin according to the preset layer height requirements. During the printing process, the spraying status of the two sets of printheads is monitored in real time. Sensors collect data on the spraying pressure and resin flow rate of the printheads. If abnormal data is detected, the system automatically pauses printing, adjusts the printhead parameters, and then continues. At the same time, a high-speed camera is used to capture the forming effect of each layer, observe the shape and height of the raised layers, and compare it with the design values. If there are deviations, they are corrected by adjusting the extra spray volume or spraying speed of the printheads. After printing is completed, the layer structure of the dental mold is inspected.

10. The method for precise adjustment of the texture of orthodontic dental mold based on 3D printing according to claim 1, characterized in that, The specific implementation steps of the layer replication enhancement and adaptation optimization technology described in S4 for the fabrication of orthodontic appliances are as follows: First, the printed orthodontic dental model undergoes surface pretreatment, and the layer surface is lightly polished using a fine polishing tool; then, the pretreated dental model is fixed on the mold base of a thermoforming machine, and appropriate thermoforming temperature, pressure, and time parameters are set according to the characteristics of the appliance material, and the thermoforming process is started; after thermoforming, the appliance is removed from the dental model after cooling to room temperature, and the edges of the appliance are trimmed to remove excess material; next, a three-dimensional optical scanner is used to scan the layer texture on the inner surface of the appliance to obtain the actual parameters of layer texture depth and spacing, and compare them with the design parameters; if there is a deviation, the cause of the deviation is analyzed. If the layer texture replication is incomplete due to improper thermoforming parameters, the thermoforming temperature, pressure, or time is adjusted, and the appliance is remade; finally, the completed appliance is disinfected, and necessary fine adjustments are made according to the patient's oral fit.

Images