Dental Object Recognition by Shape for Automated Manufacturing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing CAD software generates proprietary metadata formats for dental restorations that are not universally compatible, requiring separate interfaces for each format and complicating the import and determination of manufacturing processes, especially when metadata is absent.

Innovation Solution

A self-learning algorithm classifies dental objects based on their geometric shape using a neural network, automatically assigning a manufacturing method and adding necessary spatial structures without relying on proprietary metadata, enabling efficient and automated production.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If proprietary metadata formats are used in CAD software, then manufacturing process information can be stored, but compatibility and universality are reduced requiring separate interfaces for each format

Engineering Contradiction:
Improvemanufacturing process informationVSAvoidformat compatibility
Core Design Contradiction:
Loss of informationVSAdaptability or versatility

Solution Approach 1:

The patent applies universality by implementing a standardized metadata schema that can be universally accepted across different CAD and CAM software systems. The schema uses common data formats (JSON, XML, CSV) and standardized field names that work across multiple platforms, eliminating the need for separate proprietary interfaces while maintaining the ability to store and transmit manufacturing process information.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces an intermediary standardized metadata format that acts as a mediator between proprietary CAD software and CAM software. This intermediate layer translates various proprietary formats into a universal schema, enabling compatibility without requiring direct integration between each specific CAD and CAM system.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If proprietary metadata formats are used, then specific manufacturing information can be preserved, but device complexity increases due to multiple import interfaces

Engineering Contradiction:
Improvemanufacturing informationVSAvoidinterface complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The standardized metadata schema serves as a universal interface that can handle multiple proprietary formats through a single implementation. The CAM software only needs to implement one import interface that accepts the standardized schema, eliminating the need for multiple proprietary interface implementations while preserving all manufacturing information.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If metadata is not available, then universal compatibility is maintained, but the ability to determine manufacturing indications is lost

Engineering Contradiction:
Improvesoftware compatibilityVSAvoidindication determination
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent applies local quality by making metadata optional rather than mandatory. The standardized schema can be present when available to provide manufacturing indications, but its absence does not break compatibility. The system gracefully handles both cases: when metadata is present, it uses the standardized fields to determine indications; when absent, it maintains universal compatibility and can use alternative methods for indication determination.

Inventive Principle:
Principle #3Local quality

4Extent of automation

If self-learning algorithms are used for classification, then automation is improved, but computational requirements increase

Engineering Contradiction:
Improveclassification automationVSAvoidcomputational energy
Core Design Contradiction:
Extent of automationVSUse of energy by moving object

Solution Approach 1:

The patent applies preliminary action by pre-training the machine learning model during an offline phase using a comprehensive dataset of dental objects and their manufacturing indications. The model learns to classify objects based on geometric features and standardized metadata patterns. During actual manufacturing operations, the pre-trained model performs rapid classification with minimal computational energy, as the heavy learning work has already been completed in advance.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4252706B1Automatic detection of dental indications
Publication Date: 2025.08.20 IVOCLAR VIVADENT AG
  • EP4252706B1 patent drawingFigure 1
  • EP4252706B1 patent drawingFigure 2
  • EP4252706B1 patent drawingFigure 3

AI summary

The present invention relates to a recognition method for a dental object, comprising the steps of providing (S101) a digital dental object in a coordinate system that describes a shape of the dental object to be manufactured; and automatically assigning (S102) the digital dental object to a predetermined class based on the shape by a self-learning algorithm.