System for the automatic generation of dental surfaces from temporomandibular joint parameters

Abstract

Computer-implemented data processing system for the automatic generation of patient-specific occlusal crown geometries, comprising: a) an input interface (110) configured to receive b) a space definition module (120), configured to determine a c) a normal propagation module (130), configured to calculate d) a mesh synthesis module (140), configured to generate a e) an output interface (150) configured to provide the synthesized crown geometry as a three-dimensional model for downstream manufacturing, simulation or diagnostic systems.;

Description

II. Technical Field

[0001] The invention relates to a computer-implemented data processing system and a synthesized dental crown geometry in the field of dental prosthetics and computer-aided design (CAD). In particular, the invention relates to the automatic generation of individual occlusal surface shapes from temporomandibular joint parameters and three-dimensional dental geometries, such as those generated by commercially available intraoral scanners. The system reverses the conventional design approach: instead of loading an occlusal surface from a library and subsequently checking it against articulation, the occlusal surface shape is directly CALCULATED from the individual joint parameters and the available space—the crown is mathematically derived, not designed. III. State of the art

[0002] The relationship between temporomandibular joint geometry and tooth morphology has been known in principle since Hanau (1926) and Gysi and Lundeen & Wirth (1973). Condylar inclination determines cusp inclination, the Bennett angle shapes the fissure patterns, and the immediate side shift defines molar morphology in a 1:1 ratio (Lundeen & Wirth, 1973). Oancea et al. (2018) quantitatively confirmed that altered articulation parameters necessarily change crown morphology. Existing CAD systems, at best, utilize this relationship in the reverse direction. • CEREC Biogeneric (Mehl & Blanz, 2005): PCA-based crown design from a library. Joint parameters can be imported optionally, but are not the primary generator of the morphology. The crown is generated statistically and morphologically and subsequently checked. • CICERO System (Olthoff, van der Zel, 2000): Axiography data for dynamic occlusal surface adaptation. First static design, then dynamic correction. Causal direction not reversed. System no longer on the market. • Deep learning / GAN approaches (2022–2024): Learning correlations from large datasets, not causality. No individual joint parameters as primary input. Promising results (RMS 0.114 mm), but without understanding the cause-and-effect relationship. • Digital complete dentures (Expert Consensus 2025): Individual joint parameters are optional; the standard case is a mean-value articulator with prefabricated teeth from set libraries. NONE of the known systems directly and causally generates the occlusal occlusal surface shape from the individual temporomandibular joint parameters. The direction is always: Design (from library / statistics) → subsequent articulation check. The invention disclosed here reverses this direction: Joint parameters + space → occlusal surface. Task IV.

[0003] The object of the invention is to provide a data processing system that automatically generates a patient-specific occlusal geometry from individual temporomandibular joint parameters and commercially available three-dimensional dental geometries, without requiring a tooth library, a template, or an iterative design-verification cycle. Furthermore, a synthesized dental crown geometry is to be created whose occlusal morphology is a direct mathematical consequence of the individual temporomandibular joint geometry and which is dynamically compatible with all jaw movements of the patient. V. Solution

[0004] The problem is solved by a data processing system according to claim 1, a synthesized crown geometry according to claim 9 and a browser-based arrangement according to claim 12. Advantageous embodiments are specified in dependent claims 2-8, 10-11 and 13.

[0005] The system receives three-dimensional dental geometries and individual temporomandibular joint parameters, determines the available target volume for the crown, calculates a normal map based on the simulated jaw movements using normal propagation, and synthesizes the occlusal chewing surface through iterative mesh synthesis while adhering to the target volume. VI. Description of Preferred Embodiments 1. Technical Background: Standards as a Universal Language

[0006] Each digital tooth surface is a triangulated surface mesh (STL). Each triangle is defined by three vertices and a surface normal. The surface normal of a triangle on the occlusal surface IS the local cusp inclination, fissure slope, or marginal ridge inclination—precisely the angle that Hanau described as "cusp inclination" and Lundeen as "cusp slope." The invention is based on the understanding that the micronormals (surface normals of each occlusal triangle) are the mathematical consequence of the macronormals (condylar inclination, Bennett angle, ISS). The condylar geometry propagates throughout the entire jaw and determines the angle of each triangle on the occlusal surface. 2. Preferred embodiment: The CrownForge workflow

[0007] Step 1 -- Data import: The system receives two categories of data via the input interface (110): (A) Geometry data as three-dimensional dental meshes -- preparation stump, adjacent teeth, antagonist -- especially in STL format from commercially available intraoral scanners; (B) Temporomandibular joint parameters -- TMJ, Bennett angle, ISS, IKA -- especially as a CondylaMap parameter set from an upstream inverse parameter determination system.

[0008] Step 2 -- Space Definition: The space definition module (120) determines the three-dimensional target volume. Boundaries: above, the antagonist or the occlusal plane; below, the prepared tooth plus cement gap; mesially and distally, the adjacent teeth; buccally and palatally, the crown contour.

[0009] Step 3 -- Normal Propagation: The normal propagation module (130) translates the macro-normals into movement paths. For protrusion, left laterotrusion, and right laterotrusion, it calculates the path the antagonist will take across the planned occlusal surface. For each triangle on this path, the required surface normal is calculated to allow the antagonist to glide without collision. The result is the normal map ( Fig. 2, Fig. 210).

[0010] Step 4 -- Mesh Synthesis: The mesh synthesis module (140) generates the occlusal surface. Starting from a start mesh, vertices are shifted so that each triangle assumes its required normal, remains within the defined space, and meets minimum wall thickness requirements.

[0011] Step 5 -- Validation: The optional validation module (160) simulates all jaw movements and calculates a collision and contact index as a quality measure.

[0012] Step 6 - Output: The synthesized crown geometry is provided as an open STL mesh via the output interface (150). 3. The antagonist as a negative form

[0013] In a preferred embodiment, the antagonist inversion module (135) utilizes the geometry of the existing antagonist as a negative shape: antagonist cusps define the position of the missing fossae, and antagonist fossae define the position of the missing cusps. The occlusal surface is the only shape that simultaneously fits into the space with which the antagonist statically harmonizes and is dynamically compatible with the joint movements. 4. Calculation without antagonists

[0014] The system is also configured to generate images even without an antagonist scan: The temporomandibular joint parameters define the movement paths, while the adjacent teeth define the occlusal plane and Spee / Wilson curves. From this, an anatomically correct occlusal surface can be derived, although its accuracy is lower than with an antagonist scan, but its basic shape is mathematically determined. 5. Emergent Mesh Synthesis (Swarm Algorithm)

[0015] In another preferred embodiment, the mesh synthesis module (140) implements a swarm algorithm: Each triangle of the occlusal surface acts as an agent with local rules—normal consistency (cohesion), antagonist freedom (separation), neighbor compatibility (alignment), and force distribution (structure). Cusps emerge because, where the antagonist has a fossa, the triangles are allowed to swarm upwards. Fissures emerge because, where the antagonist has a cusp, the triangles must avoid it. 6. Browser-based implementation

[0016] A preferred embodiment of the arrangement according to claim 12 is entirely browser-based (WebGL, in particular Three.js). The dental geometries are processed client-side and do not leave the user's computer (GDPR compliance). Normal calculation uses computeVertexNormals(), collision detection uses BVH-accelerated methods, and kinematics uses 4x4 transformation matrices. Export is as an open STL file. VII. List of reference symbols 10 Stump Mesh (Preparation Geometry) 11 Antagonist Mesh 12 adjacent tooth meshes 13 TMJ parameters (CondylaMap) 20 Synthesized crown geometry (edition) 100 Data processing system (total system) 110 Input interface 120 Room Definition Module 130 Normal Propagation Module 135 Antagonist Inversion Module 140 Mesh Synthesis Module 150 output interface 160 Validation module 200 target volume 210 regular cards 220 Startmesh 230 Synthesized occlusal surface 300 CAD / CAM manufacturing system 310 3D printing system 320 Virtual Articulation System 400 client computers with browser 410 3D Viewport (WebGL) 420 Parameter Panel 430 Generate button 440 output area 450 Local processing (GDPR compliant) P Protrusion path L -- Laterotrusion path n - Surface normal (micro-normal) N -- Joint parameter normal (macro-normal) VIII. Brief description of the drawings

[0017] The invention is explained in more detail below with reference to the accompanying drawings. These show: Fig. 1 -- System block diagram of the data processing system (100) according to claim 1, comprising input interface (110) for dental geometries (10, 11, 12) and jaw joint parameters (13), space definition module (120), normal propagation module (130) with antagonist inversion module (135), mesh synthesis module (140), validation module (160) and output interface (150) for providing the synthesized crown geometry (20). Fig. 2 -- Flowchart of the CrownForge workflow with processing steps S1 (data import) to S6 (output), including normal propagation: representation of the conversion of macro-normals (temporomandibular joint parameters N) to micro-normals (occlusal surface normals n) via the normals map (210) and the iterative mesh synthesis from the start mesh (220) to the synthesized occlusal surface (230). Fig.3 - Browser-based arrangement according to claim 12 with client computer (400), 3D viewport (410) for displaying stump, antagonist and generated occlusal surface, parameter panel (420) with CondylaMap values, generate button (430), output area (440) and display of local processing (450) without server-side data upload. IX. Summary

[0018] Disclosed are a computer-implemented data processing system (claims 1--8), a synthesized dental crown geometry (claims 9--11) and a browser-based arrangement (claims 12--13) for the automatic generation of patient-specific occlusal occlusal shapes from individual temporomandibular joint parameters and three-dimensional dental geometries.

[0019] The system receives dental geometries and temporomandibular joint parameters, determines the available target volume for the crown, calculates the required surface normals of the occlusal surface based on simulated jaw movements using normal propagation, and synthesizes the occlusal occlusal surface geometry through iterative mesh synthesis. The invention reverses the conventional design approach: the occlusal surface is not loaded from a library and checked against articulation, but rather mathematically derived directly from the individual joint parameters and the available space. The resulting occlusal surfaces are dynamically compatible with all jaw movements of the patient. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited non-patent literature

[0000] Hanau (1926), Gysi and Lundeen & Wirth (1973)

[0002] (Lundeen & Wirth, 1973). Oancea et al. (2018

[0002]

Claims

Computer-implemented data processing system for the automatic generation of patient-specific occlusal crown geometries, comprising: a) an input interface (110) configured for receiving data; b) a space definition module (120) configured for determining a space definition module; c) a normal propagation module (130) configured for calculating a normal propagation module; d) a mesh synthesis module (140) configured for generating a mesh synthesis module; e) an output interface (150) configured for providing the synthesized crown geometry as a three-dimensional model for downstream manufacturing, simulation, or diagnostic systems. Data processing system according to claim 1, wherein the temporomandibular joint parameters include at least condylar path inclination (CPI), Bennett angle, immediate side shift (ISS) and / or intercondylar distance (ICD), in particular as a CondylaMap parameter set from an inverse parameter determination system. Data processing system according to claim 1 or 2, wherein the normal propagation module (130) calculates the antagonist path for each simulated jaw movement -- in particular protrusion, left laterotrusion and right laterotrusion - using 4x4 transformation matrices based on the temporomandibular joint parameters and determines for each surface element along the path the surface normal at which the antagonist can glide over the occlusal surface without collision. Data processing system according to one of claims 1 to 3, wherein the mesh synthesis module (140) generates the occlusal surface geometry by rule-based iterative vertex displacement, wherein each surface element is configured as an agent that follows at least the following local rules: normal consistency with neighboring surface elements, antagonist freedom in all simulated jaw movements, geometric compatibility with neighboring teeth and / or compliance with minimum wall thickness requirements. Data processing system according to one of claims 1 to 4, wherein the space definition module (120) further determines the target volume taking into account a parameterizable cement gap between the preparation stump and the inner surface of the crown, a minimum wall thickness, a Spee curve derived from the dental geometries and / or a Wilson curve. Data processing system according to one of claims 1 to 5, wherein the system further comprises an antagonist inversion module (135) configured to derive the occlusal surface topology from the geometry of the existing antagonist, wherein cusps of the antagonist are used as a template for fossae of the occlusal surface to be generated and fossae of the antagonist are used as a template for cusps of the occlusal surface to be generated. Data processing system according to one of claims 1 to 6, wherein the system further comprises a validation module (160) configured to check the synthesized crown geometry by simulating all jaw movements based on the temporomandibular joint parameters and calculating a collision and contact index as a quality measure, wherein a low collision index indicates the dynamic compatibility of the generated occlusal surface with the individual temporomandibular joint parameters. Data processing system according to one of claims 1 to 7, wherein the system is further configured to generate the crown geometry even without an existing antagonist scan, wherein the normal propagation module (130) derives the normal map exclusively from the temporomandibular joint parameters, the occlusal plane and the position of the tooth in the dental arch. Synthesized dental crown geometry comprising: a) a machine-readable data set stored on a b) wherein the occlusal surface geometry is derived from individual c) wherein the synthesized occlusal surface is dynamically compatible with the individual jaw movements of the patient, in particular protrusion and bilateral laterotrusion. Synthesized crown geometry according to claim 9, wherein the data set further includes the temporomandibular joint parameters used for generation, in particular as a CondylaMap parameter set comprising HKN, Bennett angle, ISS and IKA, as well as a validation parameter of dynamic compatibility. Synthesized crown geometry according to claim 9 or 10, wherein the occlusal geometry is configured as input for at least one of the following systems: a CAD / CAM manufacturing system for subtractive fabrication from ceramic, zirconia or plastic; an additive manufacturing system for 3D printing; a milling path generation system; or a virtual articulation and occlusion testing system. Browser-based or locally executed arrangement for the automatic generation of patient-specific occlusal crown geometries, comprising: a) a client computer (400) with a three-dimensional b) a space definition module executed in the browser or locally with c) a normal propagation module with real-time visualization d) a mesh synthesis module triggered by user input e) an output interface for providing the synthesized crown geometry in an open exchange format, in particular STL. Arrangement according to claim 12, wherein the arrangement is further configured to receive temporomandibular joint parameters from an upstream inverse parameter determination system, in particular a CondylaMap system according to IPC A61C 9 / 00, so that the crown geometry can be generated without manual input of joint parameters and without Axiograph or Facebow.

Images