Science Has Alarming News for Dental Technicians
- Details
- Published: Tuesday, 30 September 2025 10:18
- Written by Peter Ingle
- Hits: 915
One professional group has gone against the trend of increasing numbers on the UK dental register, with nearly 10% fewer dental technicians listed than in 2020. A recent paper offers little prospect of a recovery in numbers here, but may also concern those technicians around the world who are carrying out the work once done in the UK.
Published in The Journal of Dentistry the researchers from Peking University School and Hospital of Stomatology conducted a comparative analysis of full crown morphology designed by artificial intelligence and dental technicians.
Digital files from 12 individuals were 3D printed and prepared by an experienced dentist using depth cut burs to achieve consistency. Following tooth preparation, the resin models were rescanned and the digital models were exported for subsequent design.
For each case two crowns were designed, one by an experienced dental technician (ET) and one by an AI system.
Morphological characteristics, including buccal and lingual cusp angle, buccal-lingual diameter, mesial-distal diameter, functional wear facets and occlusal contact point counts, were compared among natural teeth, ET crowns, and AI crowns.
Statistical analysis revealed that both the ET and AI crowns exhibited larger buccal cusp angles than the natural teeth, though there were no significant differences with the lingual cusps. Buccal-lingual dimensions of AI crowns closely matched those of the natural teeth, but the ET crowns were smaller. AI did not accurately reproduce functional wear facets, but in contrast these were overcompensated in the ET crowns.
Based upon the statistical analysis, the paper concluded that AI systems can reproduce overall crown morphology to a degree comparable with experienced technicians, but remains limited in replicating functional wear facets.
The authors referred to previous research which has indicated that errors in the Z-axis (depth) scanning of intraoral scanners are significantly higher compared to errors in the X and Y axes. As a result the larger deviations from the natural tooth tended to occur in areas such as deep fissures and grooves where there is a greater distance in the Z-direction. However, the observed deviations remained within clinically acceptable thresholds.
Looking to the future, the incorporation of functional occlusion data, such as patient-specific records or dynamic bite patterns, might improve the clinical fidelity of AI systems.
At present, AI can generally be categorized into two types: knowledge-based AI and data-driven AI.
Previous studies had found that data-driven AI performed better than knowledge-based AI in reproducing dental morphology.
Another earlier study conducted in 2022 was referenced, which found that the morphology of crowns designed by experienced technicians is closer to that of natural teeth than those designed by knowledge-based AI. It went on to recommend that the restoration design should involve human correction instead of relying solely on AI to complete the work.
The authors concluded that overall, AI demonstrates strong potential in dental restoration design.
(Image generated with CoPilot).
You need to be logged in to leave comments.
Report
My comments