IADR Abstract Archives
Accuracy of Digital Impression for Full-arch Implant-supported Prostheses
Objectives: Although CAD/CAM technology offers uninterrupted digital workflow, concerns still remain about the accuracy of digital-impression procedure, particularly for full-arch implant rehabilitations. Inaccuracies may affect the long-term success of both prosthetic-rehabilitations and implants. The aim of this study was to determine the accuracy of an intraoral-scanner (TDS, 3M ESPE) creating a metrological model as reference system for digital-impressions.
Methods: A CT-scan of an edentulous mandible was randomly selected to obtain a SLM-titanium-model, which received 6 implants (Premium3.80; Sweden&Martina) in 3.6, 3.5, 3.3, 4.3, 4.5, 4.6 teeth positions. Each implant was connected to a metal scan-body, then their position was spatially determined using an opto-mechanical coordinate measuring machine (OCMM, CNC 300, Optical Gaging Products). The system integrates a video, a touch and a laser probe to determine the spatial coordinates (X-Y-Z axes) of each scan-body center-point. The data were processed with a 3D-modeling-software (Rhinoceros® 5.0) obtaining a reference system called “calibration-model”.
Four differently experienced operators took 5 digital-impressions each; then, the .stl files were processed with the modeling-software. Each impression was described by 6 scan-body center-points to be compared, for accuracy determination, with the calibration-model. The data were statistically analyzed with 1-way-ANOVA test, multiple comparison Scheffè test and Intraclass Correlation Coefficient (ICC), (p<.05).
Results: The mean error (accuracy) between the calibration-model and the 20 digital-impressions was 63±31μm, as determined combining the errors in the X-Y-Z axis coordinates. The operators experience degree did not influence the mean accuracy of the data obtained with the scanning-procedure (p>.05). The reproducibility of the scanning-procedure was assessed at a good level, as the ICC value was 0,593.
Conclusions: The accuracy determined on the digital-impressions was clinically acceptable, well within the maximum admitted mismatch reported for implant-supported prostheses fitting (150μm). The digital-impression performances were not influenced by the operator skill; this is probably due to error compensations made by the intraoral-scanner-software.
Methods: A CT-scan of an edentulous mandible was randomly selected to obtain a SLM-titanium-model, which received 6 implants (Premium3.80; Sweden&Martina) in 3.6, 3.5, 3.3, 4.3, 4.5, 4.6 teeth positions. Each implant was connected to a metal scan-body, then their position was spatially determined using an opto-mechanical coordinate measuring machine (OCMM, CNC 300, Optical Gaging Products). The system integrates a video, a touch and a laser probe to determine the spatial coordinates (X-Y-Z axes) of each scan-body center-point. The data were processed with a 3D-modeling-software (Rhinoceros® 5.0) obtaining a reference system called “calibration-model”.
Four differently experienced operators took 5 digital-impressions each; then, the .stl files were processed with the modeling-software. Each impression was described by 6 scan-body center-points to be compared, for accuracy determination, with the calibration-model. The data were statistically analyzed with 1-way-ANOVA test, multiple comparison Scheffè test and Intraclass Correlation Coefficient (ICC), (p<.05).
Results: The mean error (accuracy) between the calibration-model and the 20 digital-impressions was 63±31μm, as determined combining the errors in the X-Y-Z axis coordinates. The operators experience degree did not influence the mean accuracy of the data obtained with the scanning-procedure (p>.05). The reproducibility of the scanning-procedure was assessed at a good level, as the ICC value was 0,593.
Conclusions: The accuracy determined on the digital-impressions was clinically acceptable, well within the maximum admitted mismatch reported for implant-supported prostheses fitting (150μm). The digital-impression performances were not influenced by the operator skill; this is probably due to error compensations made by the intraoral-scanner-software.