RESUMO
OBJECTIVES: Software-based dental planning requires digital casts and oftentimes cone-beam computed tomography (CBCT) radiography. However, buying a dedicated model digitizing device can be expensive and might not be required. The present study aimed to assess whether digital models derived from CBCT and models digitized using a dedicated optical device are of comparable accuracy. MATERIAL AND METHODS: A total of 20 plaster casts were digitized with eight CBCT and five optical model digitizers. Corresponding models were superimposed using six control points and subsequent iterative closest point matching. Median distances were calculated among all registered models. Data were pooled per scanner and model. Boxplots were generated, and the paired t test, a Friedman test, and a post-hoc Nemenyi test were employed for statistical comparison. Results were found significant at p < 0.05. RESULTS: All CBCT devices allowed the digitization of plaster casts, but failed to reach the accuracy of the dedicated model digitizers (p < 0.001). Median distances between CBCT and optically digitized casts were 0.064 + - 0.005 mm. Qualitative differences among the CBCT systems were detected (χ 2 = 78.07, p < 0.001), and one CBCT providing a special plaster cast digitization mode was found superior to the competitors (p < 0.05). CONCLUSION: CBCT systems failed to reach the accuracy from optical digitizers, but within the limits of the study, accuracy appeared to be sufficient for digital planning and forensic purposes. CLINICAL RELEVANCE: Most CBCT systems enabled digitization of plaster casts, and accuracy was found sufficient for digital planning and storage purposes.