Reliability and Validity of OrthoInsight^® and iTero^® Sanners

Objectives: To quantify the reliability and establish the validity of extraoral and intraoral scanners.
Methods: The sample included 15 human mandibles that had complete dentitions except 3^rd molars. Each mandible was evaluated in four ways. Each mandible was scanned twice one week apart with the OrthoInsight extraoral scanner; each mandible was also scanned twice one week apart with the iTero intraoral scanner. All files were converted to STL format. In order to establish validity, one set of impression were taken with alginate and immediately poured up and trimmed, and intercanine width, intermolar width, arch length, mesiodistal width of the left first premolar and the height of the right canine were measured with digital calipers on mandibles. The actual measurements were taken on the mandibles. The STL files were imported into 3D tool and the same 3D measurements were taken on the scanned images.
Results: Only intermolar width, measured from the OI scans showed statistically significant (p=.043) systematic differences. Method errors, which ranged from 0.032-0.247 mm, with the IT measurements being slightly more (.02-.16 mm) reliable than OI measurements. Intraclass correlations were very high (>.926), and consistently higher for IT than OI. The cast and actual mandibular measurement showed no significant (p>.05) differences. Premolar width measured on the cast were significantly smaller than those measured from the OI (.13 mm) and IT (0.9 mm)scans. Height of the canine was measures on the cast was also larger than the OI scan (.32 mm) measurement. Arch length (.16 mm) and canine height (.37 mm) measure from the OI scan were significantly smaller than those measured on the actual mandible.
Conclusions: Both the OI and IT scans produce reliable dentoalveolar measures, with IT being slightly more reliable than OI. Measurements taken from OI scans may also be biased.