A standardization model based on image recognition for performance evaluation of an oral scanner

Area of Science:

• Dental technology
• Medical imaging
• Prosthodontics

Background:

• Accurate 3D data is vital for modern prosthodontic treatment using optical medical equipment.
• Oral scanners are essential tools, but their performance can be limited by image recognition errors.

Purpose of the Study:

• To evaluate oral scanners by examining a standardized model with a focus on image recognition errors.
• To design an improved standardized model incorporating variables referenced in ISO 12836:2015.

Main Methods:

• A basic model was created with variations in tooth profile (chamfer, groove, curve, square) and bottom surface.
• Photo-type and video-type scanners were used to capture and analyze 3D images of the models.
• Scanning was repeated to assess model distinguishability and reliability.

Main Results:

• The initial basic model failed to produce a 3D shape upon scanning.
• Image recognition rates improved with the introduction of variable factors.
• The effectiveness of different tooth profiles and floor surface patterns on recognition was observed.

Conclusions:

• Similar patterns within a model reduce the image recognition rate, as indicated by linear discriminant analysis (LDA) errors.
• Distinct class differences are necessary when developing standardized models for accurate 3D data acquisition.
• This research informs the development of more reliable standardized models for oral scanner evaluation in dentistry.