Improved manufacturing of dental implants based on digital scan data alignment

Abstract

The invention relates to a method for aligning at least two different digital scan data, each digital scan data representing at least a part of the dentition of the same patient, wherein the different digital scan data are obtained by at least two different image recording methods. The method comprises the following steps: a) recording a three-dimensional volume digital scan data; b) recording a three-dimensional shape digital scan data; c) segmenting and approximating the teeth and / or the jaw of the patient's dentition from the digital scan data; d) performing a key point determination for each tooth; e) aligning the key points obtained in step d) to obtain aligned digital scan data, wherein the alignment between the data of the volume coordinate system and the data of the shape coordinate system is performed by a coordinate transformation of one coordinate system and the coordinate transformation is based at least on the three-dimensional coordinates of the key points in the volume coordinate system and the shape coordinate system.

Description

Technical Field

[0001] This invention relates to a method for aligning at least two different digital scan data, each representing at least a portion of the dentition of the same patient, wherein the different digital scan data were obtained through at least two different image recording methods. The method includes the following steps: a) Record three-dimensional volume digital scan data; b) Record three-dimensional digital scan data; c) Segmentation and approximation of the patient's teeth and / or jawbone from digital scan data; d) Identify key points for each tooth; e) Align the key points obtained in step d) to obtain aligned digital scan data, wherein the alignment between the volume coordinate system data and the shape coordinate system data is performed through a coordinate transformation of the coordinate systems. The coordinate transformation is based at least on the three-dimensional coordinates of the key points in both the volume and shape coordinate systems. The invention also relates to the use of this method in the manufacture of dental reconstructions and to a computer program product comprising computer-readable program code for performing the method. Background Technology

[0002] Planning dental restorations and patient-specific production of dental implants is a challenging task. On the one hand, this is based on the fact that, especially in dentistry, each patient has a very specific tooth configuration. On the other hand, in rare cases, the jawbone and a single tooth can be considered individually, simply because there are always interrelationships between different tooth positions within and between the jawbones. Therefore, complex treatment situations exist and must be considered. To reduce the complexity for dental professionals, dental planning software is used to assist in planning dental reconstructions for a patient's dentition and developing patient-specific solutions. This software may include planning the placement and selection of dental implants, planning orthodontic treatment plans, designing restorations (e.g., crowns, bridges, prostheses, orthodontic appliances), and planning and designing surgical guides used during implant placement and other dental procedures. For proper planning and design, dental professionals must be able to understand and process accurate representations of the patient's dentition, jawbone, and other supporting structures. Imaging devices, such as computed tomography (CT) or cone-beam computed tomography (CBCT) X-ray machines, capture 3D images of the jawbone and teeth, providing detailed information, including jawbone size and density, tooth root location, and neural pathways. As an alternative or supplement, intraoral scanners use optical techniques to capture detailed color surface information of the patient's oral cavity.

[0003] This optical scanner can capture high-resolution details of the visual form of the dental arch, which, together with CT data, can form the basis of planning and manufacturing methods. A challenge in dental planning and manufacturing is the ability to obtain optimal combined information from two types of scans within a shared composite model. Both sensor data types (CBCT x-ray and intraoral optical scanner) capture data in sensor-specific three-dimensional coordinate systems. To merge them, the two scans must be aligned. This process, often referred to as scan matching or registration, has been very laborious and error-prone until now, as the matching operation was either performed manually or, in automated processes, yielded only suboptimal results.

[0004] The optimization of dental planning and manufacturing of dental devices is also disclosed in the patent literature.

[0005] For example, CN 112 200 843A discloses a tooth registration method based on hypervoxel-based CBCT and laser scanning point cloud data. This method includes the following steps: First, extracting tooth models from oral CBCT scan data using the area growing method; second, processing the tooth models using a laser scanner and separating the crowns from the abutment teeth based on tooth characteristics. Then, converting tooth models from different sources into tooth point cloud data and adding color information to the point cloud data to assist in the initial geometric alignment of the tooth point clouds. Simultaneously, downsampling of the point clouds is performed based on the hypervoxel method. Finally, combining the mixed features of the tooth point cloud data and defining a similarity metric accordingly, the weights between color and spatial information in the tooth point cloud data are dynamically adjusted, and the mixed features of the tooth point cloud data are ultimately measured to construct an alignment metric; a matching condition based on the mutual correspondence of the mixed features is established. This registration method is an improved ICP registration algorithm within the improved model's ICP framework. The improved model consists of a point cloud embedding network, an attention-based approximate combinatorial matching module, and a pointer generation layer. Differentiable singular value decomposition layer used to extract the final rigid transformation; it can register two tooth point cloud data with color information from different angles.

[0006] Furthermore, CN108 765 474 A describes a method comprising the steps of (1) data preparation, wherein 3D visualization of oral CT data is performed to obtain CT-reconstructed tooth images, optical scan data of the crown and gingival portions are visualized to obtain optical scan crown images, and the CT-reconstructed tooth images and optical scan crown images are placed in a unified world coordinate system; (2) initial registration, wherein feature points are manually selected to process the CT-reconstructed tooth images and optical scan crown images, and an initial registration model is further obtained; and (3) precise registration, wherein feature points are automatically selected to process the initial registration model, and a precise registration model is further obtained. This method solves the problems of data scale transformation and registration at different resolutions, and the registration results converge quickly with small registration errors.

[0007] Furthermore, US102 584 39 B1 discloses a method for manufacturing orthodontic appliances or braces for a patient, the method comprising: determining, by means of a computing device, a representation of one or more anatomical features in an image including the patient's teeth; overlaying the image and a three-dimensional model of the patient's teeth in a three-dimensional space; correcting, at least in part, the orientation of the three-dimensional model observed in the three-dimensional space in an anatomically accurate manner based on the determined representation of one or more anatomical features; after correcting the orientation, measuring the orientation of the occlusal plane of the patient's teeth relative to the determined representation in the three-dimensional space; creating a digital design for the orthodontic appliance or braces using the measured orientation; and manufacturing the orthodontic appliance or braces based on the digital design.

[0008] Such solutions known from the prior art still offer potential for further improvement, particularly regarding the reproducibility and quality of data alignment from different scanning sources, and thus the quality of the resulting orthodontic device generated from the data. Summary of the Invention

[0009] Therefore, the object of the present invention is to overcome at least part of the disadvantages known in the prior art. In particular, the object of the present invention is to provide an improved method that can reliably, faster and less prone to errors register different scan data, and thus enable the production of improved dental devices (e.g., in the form of surgical guides).

[0010] This problem is solved by the method and application of the present invention. Preferred embodiments of the invention are indicated in the specification or drawings, and further features described or shown in the specification or drawings may, individually or in any combination, constitute the purpose of the invention, unless the contrary is clearly drawn from the context.

[0011] According to the present invention, this problem is solved by a method for aligning at least two different digital scan data, wherein each digital scan data represents at least a portion of the dentition of the same patient, and wherein the different digital scan data are obtained by at least two different image recording methods. The method includes at least the following steps: a) Scan the patient's dentition using volumetric image recording methods and record three-dimensional volumetric digital scan data in a volumetric coordinate system; b) Scan the patient's dentition using a shape image recording method and record three-dimensional shape digital scan data in a shape coordinate system; c) Segmenting and approximating the teeth and / or jawbone of the patient's dentition from digital scan data, wherein the segmentation of the teeth is performed in both volume and shape coordinate systems and includes the following steps: c1) Segmenting digital scan data, wherein in the segmentation, the digital scan data is grouped relative to individual teeth and / or jaw structures; c2) Approximate the surface of segmented teeth and / or jaws using triangular meshes, wherein individual representations of the tooth and / or jaw surfaces are obtained in the form of connected triangular meshes; d) Key points are determined for each tooth, where the key points are determined based on the triangular mesh representation of each tooth and the three-dimensional coordinates of the tooth surface features are defined in the volume coordinate system or the shape coordinate system respectively. e) Align the key points obtained in step d) to obtain aligned digital scan data, wherein the alignment between the data in the volume coordinate system and the data in the shape coordinate system is performed by a coordinate transformation of one coordinate system and the coordinate transformation is based at least on the three-dimensional coordinates of the key points in the volume coordinate system and the shape coordinate system.

[0012] The method described above enables accurate alignment or registration of digital scan data from different sources in an automated and user-independent manner. Given that each digital representation of a patient's dentition is provided by the scanning method in its own coordinate system, it is inherently impossible to combine digital representations from different methods. The digital representations must be aligned to provide the same illustration of the patient's dentition. In some cases, this becomes more complex, where different scanning methods provide different "pictures" of the same oral condition. This is, for example, if different scanning methods are used, where one method provides three-dimensional volumetric data of the oral cavity (including teeth and jawbone). If more scanning methods are available, but these methods "only" provide three-dimensional representations of the oral cavity surfaces, matching the data becomes even more difficult based on the different information content of the two scans. It has been found that the method of the present invention is particularly capable of aligning such different data sources, including data representations in different coordinate systems, where more reproducible and better alignment is achieved in a fraction of the time compared to manual or semi-automatic registration routines. Based on this alignment, a unified and more detailed comprehensive view of the patient's teeth and jawbone can be provided, which can often be used to provide more intuitive and accurate oral procedure planning. Beyond this planning, it is particularly possible to obtain more precise data from the matching data for manufacturing patient-specific fitting dental devices. These devices can be customized to specific circumstances and provide better fit solutions compared to planning or manufacturing devices based on non-matching or only semi-optimal matching or alignment data. Attached Figure Description

[0013] Further advantages and advantageous embodiments according to the purposes of the invention are illustrated by the accompanying drawings and explained in the following description. It should be noted that the drawings are merely descriptive and are not intended to limit the invention in any way.

[0014] The attached diagram shows: Figure 1 A schematic diagram of the method according to the present invention; Figure 2 A schematic diagram of the method according to the present invention; Figure 3 Registration of CBCT data and keypoint-based optical scans; Figure 4 According to an exemplary labeling scheme for FDI; Figure 5 Example of a combined 3D view of CBCT and optical scanning; Figure 6 Example of a combined CBCT and optical scan cross-section; Figure 7 An example of segmented and separated tooth meshes from a CBCT scan; Figure 8Example of registration results based on key points; Figure 9 An example of the result after fine registration. Detailed Implementation

[0015] The method of this invention relates to a method for aligning at least two different digital scan data, each representing at least a portion of the dentition of the same patient, wherein the different digital scan data are obtained through at least two different image recording methods. Different intraoral scanning methods provide method-specific data, wherein the illustration of teeth and the oral cavity depends on the measuring device and, in some cases, also on the operator. It is known that there is no direct correlation between different methods, different views, and illustrations; therefore, digital representations cannot simply be combined to produce a common picture, thus providing all the information from both views. Alignment or registration enables combined views, where different data are matched in origin and orientation, and enables combined views of the dentition. Different methods can be applied at different points in time; however, digital representations of at least a shared subgroup of the same teeth and / or jawbones of the same patient must be provided. Different image recording methods provide data based on different physical processes. Therefore, recording methods can be grouped according to the physical means applied (e.g., visible light, X-ray, IR, magnetic spin, or electric spin, etc.).

[0016] The method includes method step a), wherein three-dimensional volumetric digital scan data in a volume coordinate system is recorded by scanning the patient's dentition using a volumetric image recording method. The three-dimensional volumetric digital scan data originates from a scanning method that provides not only surface information but also volumetric information of the dentition at hand. Methods such as CBCT or magnetic resonance tomography (MRT) can provide density-related data, where volumetric structures like teeth or jawbones can be evaluated not only on the surface but also by resolving overall structures containing the same or similar densities. Scanning the patient's dentition means scanning at least a portion of the oral cavity (including dental and / or jawbone structures) and recording data.

[0017] The method includes step b), wherein three-dimensional shape digital scan data in a shape coordinate system is recorded by scanning the patient's dentition using a shape image recording method. The three-dimensional shape digital scan data originates from a scanning method that can provide surface information but not volumetric information of the dentition at hand. For example, optical methods using light as a probe can provide color-correlated data with high resolution and accuracy, but cannot assess the volumetric structures beneath the surface. For example, an intraoral scanner provides a three-dimensional representation of the surface of the patient's dentition (including gingival and jawbone surface structures). In particular, it can detect soft tissue structures compared to typical volumetric scanning methods.

[0018] In step c), the teeth and / or jawbones of the patient's dentition are segmented and approximated from the digital scan data, wherein tooth segmentation is performed in both volume and shape coordinate systems. Both recording methods provide a digital representation of the oral cavity (including teeth and jawbones). This data is processed in this step to isolate structures of interest, such as individual teeth or jawbones, from the overall scan of the dentition. The method is performed independently for the two scan datasets and produces two datasets, one containing three-dimensional volume digital scan data and the other containing three-dimensional shape digital scan data, which contain information about the isolated individual structures in the oral cavity. In this case, separation means that data about individual structures in the oral cavity (e.g., teeth) can be further processed. The data representing individual structures are recombined and approximated, for example, to reduce the overall data volume.

[0019] Segmentation step c) includes substep c1), in which segmentation of the digital scan data is performed, wherein the digital scan data is grouped with respect to individual teeth and / or jaw structures during segmentation. The segmentation of the digital scan data for each method can be performed manually or (semi-)automatically by methods known to those skilled in the art. As described above, in segmentation, the overall data is divided into separate groups, where each distinct group represents a single marked structure within the entire oral cavity. The markings reflect the type and location (if applicable) of the structure (e.g., a specific tooth or maxilla / mandible). These separated structures can then be further processed and handled individually.

[0020] Segmentation step c) further includes sub-step c2), where an approximation is performed on the surface of the segmented teeth and / or jaws using a triangular mesh, wherein the individual representation of the tooth and / or jaw surface is obtained as a connected triangular mesh. To reduce system complexity and data volume, it has been found useful to transform the separated structures into a triangular mesh representation. Mesh representations of surface structures are known in the dental field, and such transformations are known to those skilled in the art. The transformation is performed on the individual structure of each segment of the 3D volumetric digital scan data and the 3D shape digital scan data. It is also possible that the scanning method computes and returns such a mesh representation itself during the scanning process. In this case, separate additional transformations are naturally unnecessary.

[0021] In step d), keypoint determination is performed for each tooth. Keypoint determination is based on a triangular mesh representation of each tooth and defines the three-dimensional coordinates of tooth surface features in either volume or shape coordinates. Tooth keypoints are prominent and visible structures on the outer surface of the tooth. Possible keypoints can be selected for convenience and may include the tooth perimeter at a specific height, the midpoint of the tooth's upper surface, the edge position of the tooth, etc. These tooth keypoints are determined for each segment of tooth in both the volume and shape models. Keypoints can be represented by the three-dimensional coordinates of this structure for a specific tooth. Based on the fact that keypoints have been determined for all teeth, the positions of keypoints on individual teeth in a given model and the three-dimensional relationships between all teeth are obtained.

[0022] In step e), alignment is performed on the keypoints obtained in step d) to obtain aligned digital scan data, wherein the alignment between the volume coordinate system data and the shape coordinate system data is performed through a coordinate transformation of one coordinate system, and the coordinate transformation is based at least on the three-dimensional coordinates of the keypoints in both the volume and shape coordinate systems. Based on the correspondence of tooth keypoints belonging to the same tooth in each model, a mathematical transformation can be derived that makes the two keypoint sets match. This correspondence can be established using tooth-specific location markers from step c1). The transformation of keypoints in one model can, for example, include translation and / or rotation transformations of the entire keypoint set of a scanning method. A translation and / or rotation transformation of a keypoint set produces a representation where the keypoints in both datasets include the same or similar three-dimensional coordinates. This transformation is the basis for representing both scanning methods within the same coordinate system and obtaining more detailed and comprehensive images. Based on the combined model, better planning of dental applications and patient-specific fabrication of dental devices can be performed.

[0023] In a preferred aspect of this method, within or after step c) and before step e), a jawbone position-sensitive marker in a dental symbol coordinate system can be assigned to each tooth in the volume and shape scans, and the alignment in step e) may further include comparing the matching of the jawbone position-sensitive markers of the teeth in the volume and shape coordinate systems. In step c) or d), the segmented teeth can be labeled according to any dental symbol coordinate system. In such a system, a specific descriptor is typically appended to the teeth as a function of the tooth's position within the jawbone. One possible known symbol is the FDI symbol, as described below. However, any symbol system can also be used, where a different but unique descriptor can be used for each tooth. Such a marker is added to the segmented teeth in both the volume and shape datasets, and the alignment of the two datasets is also performed as a function of the correspondence between the tooth markers in the two datasets. This method can accelerate the matching process between the two coordinate systems and can further improve repeatability, especially in cases of difficult jawbone structures or tooth arrangements in the form of dental bridges or several missing teeth. More reliable assignment can be achieved, particularly in such cases.

[0024] In a preferred embodiment of the method, the comparison of the specified location-sensitive markers includes the alignment of keypoints in the volume and shape coordinate systems prior to step e). For the robustness and robustness of the entire method, it has been found advantageous to perform a marker comparison between the teeth in the two coordinate systems before keypoint alignment. This comparison is particularly effective if only one or more teeth are missing in a single dataset, or if complex crown or bridge structures exist. Therefore, using only tooth keypoints available in both scan datasets for keypoint alignment increases robustness and reduces complexity.

[0025] Within the preferred features of this method, the comparison of marker matching is based on residuals in the form of the sum of squared distances between corresponding tooth positions, and the comparison is performed multiple times, generating several different marker sets, wherein each marker set differs in the designation of at least one tooth marker. During step c2), keypoint correspondences based on the designated markers are highly helpful in obtaining good registration results. The quality of this registration can be evaluated as a function of the residuals, in the form of the sum of squared distances between corresponding tooth keypoint positions after the transformation is applied. This procedure can be performed multiple times, with several different sets of suitable marker error assumptions, and thus different correspondence designations, used to increase robustness to marker error errors. The final keypoint registration result is the one with the smallest residuals. Such a step may be necessary, especially if the tooth position markers from step c2) are incorrect, potentially leading to inconsistent markers across different datasets. Particularly in the case of interdental spaces or bridges, the marker procedure may cause positional shifts that are typically small. If an interdental space is detected only in one dataset but not in another, an incorrect correspondence is established, and therefore the registration result will not be optimal. Applying registration to all possible keypoint combinations between two datasets and comparing the results would be a computationally expensive and time-consuming task. Therefore, checking suitable, for example, random subsets of the original corresponding specified matrix limits the number of combinations to a manageable dataset. This randomized label offset scheme may help find better matching solutions. Based on the comparison on the basis of segmented teeth, this comparison and matching can be performed quite quickly and accelerates the following comparison steps based on grid representation.

[0026] According to a preferred embodiment of the method, the differences in the markings can be set to a single location. The amount of substitution for the marking correspondence can be further reduced by applying the following scheme: Considering the assumption that if a tooth location is incorrectly marked, adjacent teeth will also be moved by the same amount. This increment can reduce the number of combinations to a manageable size. If a tooth location in the marking scheme is empty, all subsequent possible consecutive subsets will be moved by one, resulting in a much smaller set of marking hypotheses, where the corresponding registration results can be derived and compared within a reasonable amount of processing time. This further comparison may help identify any misclassified markings in the previous marking specification steps.

[0027] In another preferred embodiment of the method, each tooth in both the volume and shape coordinate systems is assigned an FDI (World Dental Federation notation) mark according to ISO 3950 "Signage of Tooth Positions and Oral Regions". For the rapid and reliable handling of difficult tooth setups, it has been found useful to use the FDI system for tooth marking and, consequently, to compare these marks. Based on the symmetry of this notation, comparable results can be readily obtained in two scan sets. Nevertheless, any consistent nomenclature scheme is also applicable.

[0028] According to a further preferred embodiment of the method, in the key point determination in step d), the crown peak of each tooth can be determined. For reliable matching based on key points, the crown peak of each tooth has been found to be useful for comparison and matching. This key point is relatively small compared to other structures and can usually be captured well by different scanning methods. The crown peak of a tooth is defined as the highest point extending into the oral cavity from the tooth, and the lowest point for maxillary teeth.

[0029] In a preferred aspect of this method, the three-dimensional volumetric digital scan data is obtained by a method selected from the group consisting of computed tomography (CT), cone-beam computed tomography (CBCT), or magnetic resonance imaging (MRT). The method of the present invention has been found to be particularly useful for the registration of volumetric data obtained by the methods described above. Registration or matching between different methods can be performed very reliably, and the registration produces a better combined view, which forms the basis for more accurate planning and device manufacturing methods.

[0030] According to a further preferred embodiment of the method, three-dimensional shape digital scan data can be obtained using an optical dental scanner. The method of the present invention has been found to be particularly suitable for registering shape data obtained by an optical scanner. Registration or matching between dental scanner data and three-dimensional volume digital scan data can be performed very reliably, and the registration produces a better combined or registered view, which forms the basis for more accurate planning and device fabrication.

[0031] In a preferred aspect of the method, the alignment in step e) is performed in a two-step procedure, wherein in a first coarse alignment, a first coordinate transformation is performed based on the 3D positions of key points in the volume and shape coordinate systems, and in a second fine alignment, a second coordinate transformation is performed based on a comparison of the triangular mesh representations of the teeth in the volume and shape coordinate systems. For the overall system performance and the reliability of finding correct registration between different coordinate systems, it has been found useful to divide the overall alignment into two steps. The first step is performed guided by key points and generally results in fairly good alignment between coordinate systems. If the highest precision matching is required, fine alignment can be performed at the already found positions. Fine alignment is more time-consuming because this step is based on a comparison of triangular mesh data, and these data sets are much larger than key points. Furthermore, this stepwise approach can also be useful in very difficult cases where one or more teeth are missing in different locations or sequentially within the jawbone.

[0032] According to a preferred embodiment of the method, coarse alignment can be based on optimizing the sum of squared distances between corresponding keypoints in the volume coordinate system and the shape coordinate system. For fast and reproducible alignment, it has been found useful to use the sum of squared distances between corresponding keypoints as the alignment optimization objective function. Minimizing this function yields the required transformation parameters. The computation can be very fast, and the mathematical model allows for optimal intermediate results, which can be further processed in fine registration.

[0033] According to a further preferred embodiment of the method, the fine alignment step can be based on the Iterative Closest Point (ICP) algorithm. To further improve alignment accuracy, a specific version of the Iterative Closest Point (ICP) algorithm can be used. The goal of this algorithm is to align two point clouds of the same object to each other. It achieves this by first finding the point correspondences between the two clouds, and then finding a transformation (3D rotation and translation) that minimizes a certain metric, which in the simplest case is the sum of the squared distances between the correspondences (the so-called point-to-point metric). The transformation is applied to the first point cloud, and these two steps are repeated until a convergence criterion is met, and the algorithm returns the final cumulative transformation matrix. For example, the vertices of two triangular mesh datasets from a CBCT scanner and an intraoral optical scanner can be interpreted as two point clouds from the same jawbone. Due to the available triangle information within each mesh, the surface normal of each vertex can be calculated. Therefore, a more efficient computation of the "point-to-plane" metric can be applied in the transformation optimization step. Thus, the sum of the squared distances between each vertex and its corresponding surface defined by the points and normals of the other point cloud is minimized.

[0034] Based on the fact that the separation mesh for each tooth has already been calculated, the possible combinations of points from the same corresponding tooth in another scan can be limited in the corresponding matching step. This produces computationally efficient, robust, and accurate registration results.

[0035] Furthermore, the invention relates to the use of the method according to the invention in the manufacture of dental reconstructions for a patient's dentition. Based on better registration of scan data from different sources as described above, better dental treatment planning can be provided, and the design and manufacture of any device can be adapted to be used during treatment or for dental reconstructions tailored to the specific needs of a particular patient. Treatment failures are avoided, and treatment problems are reduced due to better device fitting, and in some cases, accelerated healing can be achieved.

[0036] In a preferred embodiment, the dental reconstruction may be selected from the group consisting of: dental implants; a group of prostheses selected from crowns, bridges, prostheses, and orthodontic appliances; a surgical guide used during implant placement; or a combination of at least two of these components. In particular, the components of the aforementioned group of dental reconstruction devices can benefit from the improved method according to the invention. The reconstruction can be precisely adapted to the patient's needs and can avoid any misconstruction or particular incompatibility.

[0037] Furthermore, dental reconstructive devices are also within the scope of this invention, wherein the manufacture of such devices includes device design steps based on the method of this invention. The advantages of the devices are particularly those of the method of this invention and the uses of this invention.

[0038] Furthermore, the present invention is a computer program product comprising computer-readable program code for performing the method according to the present invention.

[0039] Figure 1 Different steps of the method of the present invention are illustrated. In the first stage, three-dimensional volume digital scan data and three-dimensional shape digital scan data are processed separately. Each scan dataset is segmented, and after segmentation, individual structures in the oral cavity (such as teeth or jawbones) are parsed and can be processed individually. For each segmented structure, a triangular mesh representation of the surface structure is generated. It is possible that the scan data already exists in the mesh representation. In this case, the triangular mesh transformation step can be skipped. Based on the triangular mesh representation of the tooth surface structure, key points of each tooth can be detected. Possible key points can be the crown peaks of each tooth. Registration, i.e., the alignment of two scan coordinate systems, is performed based on the three-dimensional position of the key points. For alignment, for example, the sum of the squared distances between corresponding teeth in different scan coordinate systems can be used as a quality parameter. Translation and rotation transformations are implemented during the optimization process. Based on the registration, a combined image is obtained to represent the details of the three-dimensional shape digital scan data and the three-dimensional volume digital scan data in a common view. Based on the matched dataset, dental devices can be better planned and manufactured.

[0040] Figure 2A preferred embodiment of the method of the present invention is illustrated. In the first stage, three-dimensional volumetric digital scan data and three-dimensional shape digital scan data are processed separately. Each digital scan dataset is segmented, and after segmentation, individual structures in the oral cavity (such as teeth or jawbones) are parsed and can be processed individually. For each segmented structure, a triangular mesh representation of the surface structure is generated. It is possible that the scan data already exists in the mesh representation. In this case, the triangular mesh transformation step can be skipped. Based on the triangular mesh representation of the tooth surface structure, key points for each tooth can be detected. Possible key points could be the crown peaks of each tooth. In addition to key point determination, markers are attached to all teeth, where the marker is a function of the tooth position in the jawbone. Possible markers could be FDI markers. Marker attachment can be performed before or after key point calculation. The registration step, i.e., the alignment of the two scan coordinate systems, is performed incrementally. In the first coarse registration step, the two coordinate systems are aligned based on the markers and key points. The alignment as a function of tooth markers can be repeated several times further, where different markers can be attached to at least one tooth among all teeth in that coordinate system. This step may detect any erroneous tooth markers and may produce better matches. After alignment is performed using the function of the marker correspondence and keypoints, registration can be further optimized by considering the entire mesh of a single tooth instead of just the keypoints. In the second step of registration, the two coordinate systems can be further aligned based on the triangular mesh representation of the surface structure. In this step, the overall transformation between the two coordinate systems is much smaller compared to the coarse registration step. However, in this step, the "optimal" alignment between the two coordinate systems can be achieved. Based on the registration, a combined and accurate image representing the details of the 3D shape digital scan data and the 3D volume digital scan data is obtained. Translation and rotation transformations are implemented during the optimization process. Based on the matched dataset, dental devices can be better planned and manufactured.

[0041] Figure 3 Examples are shown of three-dimensional shape digital scan data obtained from the surface of a patient's teeth and three-dimensional volume digital scan data in the form of a three-dimensional representation of the surface and internal structure of the jawbone and teeth. The two representations are independent of each other, and in the method of the present invention, surface data, for example, from an intraoral scanner is matched with volume data from a CBCT scan. Registration results in a transformation to combine the two datasets.

[0042] Figure 4 Possible tooth marking schemes are illustrated. In the marking step, each tooth is designated using a marker (e.g., 11, 21, and 22) according to the FDI symbol. In optimization, if the marker position is changed, the marking of at least one tooth is altered. In this figure, the marker for tooth 12 can be changed to 13 to detect the possibility of an incorrect marker being added during the marking step. This marking modification is very useful in cases of missing teeth.

[0043] Figure 5 A 3D visualization of the aligned data is shown. The optical scan reveals a bright, smooth structure above the darker, coarser structure found in a CBCT scan. In particular, since soft tissue structures are absent in CBCT scans, the optical scan expands the information content of the view, thus enabling better planning capabilities.

[0044] Figure 6 A cross-sectional view of the aligned data is shown. The small white dashed lines represent optical scans. It again shows the extent of soft tissue structures like the gums, relative to structures with high density.

[0045] Figure 7 The segmentation results of the CBCT scan, transformed into a triangular mesh, are shown. Each tooth is separated from the jawbone and can be processed individually for further analysis. In some cases, this, along with designated location markers, forms the basis of the registration process.

[0046] Figure 8 The registration results based on keypoints are shown. The optical scan, shown as a structured white grid, is very close to the maxillary grid from the CBCT scan (shown as solid dark gray). Nevertheless, the registration results are not as perfect as indicated by the black double-headed arrows. Disparities still exist between the two scan datasets. This could occur if the keypoint locations specified in the two datasets do not match.

[0047] Figure 9 The results of the fine registration are shown. The optical scan, shown as a structured white grid, matches perfectly with the maxillary tooth grid from the CBCT scan, shown in dark gray. The two datasets are not entirely consistent due to different scanning procedures and different processing and approximation processes. However, the alignment is accurate enough to allow viewing the complete dataset in a common view and to provide an impression of soft tissue structures relative to high-density structures like bone and teeth.

Claims

1. A method for aligning at least two different digital scan data, each digital scan data representing at least a portion of the dentition of the same patient, wherein said different digital scan data are obtained by at least two different image recording methods, the method comprising at least the following steps: a) Scan the patient's dentition using a volumetric image recording method and record three-dimensional volumetric digital scan data in a volumetric coordinate system; b) Scan the patient's dentition using a shape image recording method, and record three-dimensional shape digital scan data in a shape coordinate system; c) Segmenting and approximating the teeth and / or jawbone of the patient's dentition from the digital scan data, wherein the segmentation of the teeth is performed in the volume coordinate system and the shape coordinate system and includes the following steps: c1) Segmenting the digital scan data, wherein in the segmentation the digital scan data is grouped relative to individual teeth and / or jaw structures; c2) The surface of the segmented teeth and / or jaws is approximated by a triangular mesh, wherein an individual representation of the surface of the teeth and / or jaws is obtained in the form of a connected triangular mesh; d) Key point determination for each tooth, wherein the key point determination is based on the triangular mesh representation of each tooth and the three-dimensional coordinates of the tooth surface features are defined in the volume coordinate system or the shape coordinate system respectively; e) Align the key points obtained in step d) to obtain aligned digital scan data, wherein the alignment between the data in the volume coordinate system and the data in the shape coordinate system is performed by a coordinate transformation of one coordinate system and the coordinate transformation is based at least on the three-dimensional coordinates of the key points in the volume coordinate system and the shape coordinate system. Within or after step c) and before step e), a jawbone position sensitive marker in a tooth symbol coordinate system is assigned to each tooth in the three-dimensional volume digital scan data and the three-dimensional shape digital scan data, and the alignment in step e) further includes comparing the matching of the jawbone position sensitive markers of the teeth in the volume coordinate system and the shape coordinate system; the comparison of the position sensitive markers includes step e) before aligning the key points in the volume coordinate system and the shape coordinate system; The comparison of the marker matching is based on residuals in the form of the sum of squared distances between corresponding tooth positions, and the comparison is performed multiple times, wherein several different marker sets are generated, wherein the marker sets are different in at least one tooth marker designation, and the difference in the markers is set to a position.

2. The method according to claim 1, wherein, Assign a World Dental Federation notation to each tooth in the volume coordinate system and the shape coordinate system according to ISO 3950 "Identification of tooth position and oral cavity area".

3. The method according to claim 1, wherein, In the determination of key points in step d), the crown peak of each tooth is determined.

4. The method according to claim 1, wherein, The three-dimensional volume digital scan data is obtained through computed tomography, cone-beam computed tomography, or magnetic resonance imaging.

5. The method according to claim 1, wherein, The three-dimensional digital scan data was obtained using an intraoral optical scanner.

6. The method according to claim 1, wherein, The alignment in step e) is performed in a two-step procedure, wherein in the first coarse alignment, a first coordinate transformation is performed based on the three-dimensional positions of key points in the volume coordinate system and the shape coordinate system, and in the second fine alignment, a second coordinate transformation is performed based on a comparison of the triangular mesh representation of the teeth in the volume coordinate system and the shape coordinate system.

7. The method according to claim 6, wherein, The coarse alignment is based on the optimization of the sum of the squared distances between the corresponding key points in the volume coordinate system and the shape coordinate system.

8. The method according to claim 7, wherein, The fine alignment step is based on the iterative nearest point algorithm.

9. Use of the method according to any one of claims 1-8 in the manufacture of dental reconstructions of a patient's dentition.

10. The use according to claim 9, wherein, The dental reconstruction is selected from the group consisting of at least one of the following: dental implants, prostheses, and surgical guides used during implant placement, wherein the prosthesis is selected from the group consisting of dental crowns, dental bridges, dentures, and orthodontic appliances.

11. A computer program product comprising computer-readable program code for performing the method according to any one of claims 1-8.

Images