Accuracy and reliability of lower dental arch reconstructions

Introduction: In the literature, there is information regarding the accuracy of models generated by conventional impression materials and intraoral scanners, but data on the accuracy of 3D models generated from cone-beam computed tomography (CBCT) is still lacking. Aim: The aim of this article is to investigate and compare the accuracy of tooth reconstructions made on 3D models generated from CBCT and intraoral scanning, as well as on plaster models from conventional impression materials, to the results of intraoral measurements using a digital caliper. Materials and Methods: The study included a total of 38 individuals (16 males and 22 females). After the initial examination, we scheduled appointments for the approved participants to undergo clinical procedures and imaging studies in the following sequence: 1. placement of composite markers; 2. physical measurements; 3. intraoral scanning; 4. CBCT; 5. capturing a conventional impression; 6. removal of composite markers. Following are the laboratory and measurement stages for the study: 7. casting gypsum models; 8. measurements on gypsum models; 9. converting the DICOM files from CBCT scans to STL files; 10. conducting measurements on digital models from CBCT and intraoral scanning. Results: Results from the reliability assessment of the researcher’s measurements for the studied modalities indicate a correlation coefficient ranging from moderate to excellent with very high statistical significance. Concerning accuracy, differences are observed between individual modalities. Conclusion: In summary, conventional methods and materials still outperform intraoral scanners in terms of the accuracy of the obtained reconstructions. 3D models generated from CBCT scans are generally the least satisfactory among the tested modalities, with deviations typically within clinically acceptable values.