Digitalization and Surfacing - Influence on the Accuracy in Dental CAD/CAM-technology

Methods: Real metal models of a canine and a molar and their corresponding identical CAD-surfaces built the basis for this experiment. The prepared teeth were digitized directly (metal teeth) and indirectly (gypsum duplicate dies) with three different measuring systems. As reference, a non-measured point cloud was generated directly from the CAD-surfaces. For each point cloud different surfaces were made. From these surfaces, copies were milled out of polyurethane. The copy teeth were digitized again and the manufacturing error and the over-all error compared to the original CAD-surfaces were determined. The statistical analysis was performed with one-way analysis of variance.

Results: In average, the manufacturing error of the canine's copies was between 36µm and -26µm. The over-all error amounted to average values from 65µm to 58µm. For the molar, mean values of +/-22µm (manufacturing error) and from 33µm to -34µm (over-all error) were found. The qualitative, quantitative and statistical analysis showed that the achievable precision is influenced by the kind of digitalization. An idealized mathematically generated point cloud can be used as gold-standard. The kind of surface generated did not significantly influence the duplicate's error. The tooth shape was found to be the dominant factor for the achievable manufacturing precision. Qualitative analysis revealed the largest deviations in areas with strong changes of curvature.

Conclusion: In order to improve the fabrication precision, milling strategies which take the tooth shape into consideration have to be developed. The experimental set-up evolved allows a differentiated analysis of a process chain as well as comparative studies between different CAD/CAM-systems.