Impact of 'Data Exchange by Overscanning' and Different 'Cut- Out Rescanning' Procedures on Trueness and Scanning...

To investigate the effect of over-scanning and different cut-out/rescanning procedures on the local trueness of 3D-datasets.

A typodont with a prepared tooth (#26) was digitized using an intraoral-scanner (Cerec Primescan, v5.2) to create a master-die. Full-arch scans with a retraction-cord covering the preparation margin were performed and duplicated to obtain four identical scans to be used for each test group: Die-cut Re-scan (Die-RS), Cord-cut Re-scan (Cord-RS), Segmental-cut Re-scan (Seg-RS), and Over-scan (Over-S). After removing the cord, group-specific cut-out-rescanning procedures were performed on each duplicate. For the Over-S group, no cut-out modifications were made; the scanning procedure was continued until the cord's image disappeared. The time-expenditure of rescinding procedures were recorded in seconds for each group. The trueness between the reference master-die and experimental scans (N=15 per-group) was evaluated by the absolute-mean-deviation, root-mean-square-error (RMSE), and the top 0.5% percentile calculations. Data were analyzed with the Kruskal-Wallis test with the Bonferroni corrections, one-way-ANOVA with Tukey HSD tests (α =0.05).

The Seg-RS showed lower RMSE (6.26±1.13µm) and 0.5% percentile values (20.20±6.18µm) than the other groups (P<0.05). The absolute-mean-deviation of Seg-RS (5.04±0.85µm) was lower than the mean deviations of Die-RS (7.97±2.91µm) and of Cord-RS (5.81±1.61µm) (P<0.05), but not significantly different from Over-S (4.99±2.19µm) (P=1.000). Concerning time expenditure, the Die-RS (34±9s) showed significantly lower values among the groups (Seg-RS:44±5s, Cord-RS:51±8s and, Over-S:52±7s) (P<0.05).

The Seg-RS demonstrated the highest trueness values among the groups. While Die-RS was the most time-efficient approach, the trueness compared to Seg-RS was significantly lower.