Our objectives were to evaluate and compare the digital dental models generated from 2 commercial intraoral scanners with manual measurements when performing 3-dimensional surface measurements along a curved line (curvilinear).
Dry mandibles (n = 61) with intact dentition were used. The mandibles were digitized using 2 chair-side intraoral scanners: Cadent iTero (Align Technology, San Jose, Calif) and Lythos Digital Impression system (Ormco, Orange, Calif). Digitized 3-dimensional models were converted to individual stereolithography files and used with commercial software to obtain the curvilinear measurements. Manual measurements were carried out directly on the mandibular teeth. Measurements were made on different locations on the dental arch in various directions. One-sample t tests and linear regression analyses were performed. To further graphically examine the accuracy between the different methods, Bland-Altman plots were computed. The level of significance was set at P <0.05.
There were no significant differences between any of the paired methods; this indicated a certain level of agreement between the methods tested ( P >0.05). Bland-Altman analysis showed no fixed bias of 1 approach vs the other, and random errors were detected in all comparisons. Although the mean biases of the digital models obtained by the iTero and Lythos scanners, when compared with direct caliper measurements, were low, the comparison of the 2 intraoral scanners yielded the lowest mean bias. No comparison displayed statistical significance for the t scores; this indicated the absence of proportional bias in these comparisons.
The intraoral scanners tested in this study produced digital dental models that were comparatively accurate when performing direct surface measurements along a curved line in 3 dimensions.
Curvilinear accuracy of Cadent iTero and Lythos Digital Impressions was tested.
Accuracy was sufficient to reproduce geometric information with both 3-D model systems.
Algorithms used to produce digital models can transform physical information from a curved surface into the digital platform.
Due to the recent advances in dental technology, many appliances such as retainers, expanders, and clear retainers can now be produced directly from digital dental models. By incorporating this feature into the clinical practice, the time and cost of making impressions and sending them to the laboratories are minimized. Even indirect bracket setups can be performed on the digital platform with high precision, helping to decrease the need for bracket repositioning later in treatment, thus reducing overall treatment time.
Although many practitioners have already embraced the use of digital cameras and digital radiography units, universal acceptance of intraoral scanners has been slower to make the transition considering the number of methods available to digitize the dental models. Even though impression-free digital models obtained with intraoral scanners offer high accuracy compared with other alternatives, it was reported that alginate impressions are still the preferred model acquisition method with regard to chair time and patient acceptance. However, as digital technology keeps improving, so do the intraoral scanners regarding enhancements in the scanner body and tip, portability, and scanning time. When the time spent for laboratory processing is considered, obtaining digital dental models directly from the patient without the need for dental impressions may offer a faster protocol. Soon chair-side scanners may be the standard of care as long as they can provide clinicians with the most accurate and efficient system in producing digital dental models.
Recent studies have evaluated the interarch and intra-arch measurements derived from digital models generated from intraoral scans and compared those obtained from conventional gypsum models. Direct digital acquisition of the dental arches with a chair-side scanner appeared to be reliable and accurate in these reports. Al Mortadi et al demonstrated successful fabrication of a Hawley retainer using an intraoral scanner. Anh et al reported that, from a clinical standpoint, intraoral scanners were highly accurate regardless of the degree of tooth irregularity. However, scanning sequence and careful application of the scanning procedure were shown to affect the precision of the end result. The largest biases were mostly reported for the posterior part of the dental arch, which may relate to the scanning technique and the difficulty of access to these areas. Therefore, it is important to evaluate the precision of intraoral scanners using complex measurements on various aspects of the dentoalveolar anatomy.
To date, no studies have assessed the accuracy of different intraoral scanners using surface measurements made along a curved line (curvilinear), which should offer clinically more relevant information as opposed to linear measurements constructed on only 2 points. The aim of this study was to investigate the curvilinear accuracy of 2 commercial intraoral scanners in comparison with digital caliper measurements.
Material and methods
The study sample comprised 61 dry skulls. The mandibular arch from each skull was scanned using the Lythos Digital Impression system (Ormco, Orange, Calif) and the Cadent iTer scanner (Align Technologies, San Jose, Calif). Once the scans were completed, the raw images were converted to stereolithography files. These files were used with commercial software (3-matic Research 9.0, x64; Materialise, Leuven, Belgium) for the measurements. The measurements were carried out on unsectioned, shaded digital models of the mandibles with the software’s builtin ruler tool. The ruler tool was set to measure the distance over a surface with a curve creation method set on the true shortest path of a curve using the World Coordinate System. Nylon monofilament and digital calipers were used to make the curvilinear measurements directly on the tooth surface.
Measurements were performed in different aspects of the mandible in various directions. The following curvilinear measurements were made: buccal surface of the mandibular right canine along the long axis of the tooth from the cusp tip to the crestal bone ( Fig 1 , A ); the uppermost surface of the crestal bone along the mandibular right first molar below the most lingual points of the marginal ridges ( Fig 1 , B ); and the uppermost occlusal surface along the lingual cusp of the mandibular left second premolar, starting from the center of the mesial marginal ridge and ending in the center of the distal marginal ridge ( Fig 1 , C ). All measurements were carried out by the same operator (S.M.) to the nearest 0.01 mm.
SPSS software (version 24; IBM, Armonk, NY) was used for statistical analysis. Measurements performed in 3 directions for each digital model were paired with corresponding direct caliper measurements. A 1-sample t test was used to test the hypothesis that there would be no difference between the paired sets of measurements. The test value was set at 0. No significant differences were found for any of these comparisons. Therefore, the 3 measurements obtained from the same model were combined to represent the iTero, Lythos, and direct measurements for further analysis. The Bland-Altman analysis was performed using XLSTAT Mac (version OS X; Addinsoft, New York, NY), and Bland-Altman plots were computed for the paired comparisons of the 3 methods. The analysis was used to visually demonstrate the agreement for measurement values between the manual caliper measurements, the Lythos scan measurements, and the iTero scan measurements. The within-observer repeatability was evaluated using intraclass correlation analysis by repeating all measurements from 10 randomly selected models at a 1-month interval. The error study was performed using Dahlberg’s formula. Furthermore, linear regression analyses were used to investigate whether there was a proportional bias in the data. The level of significance was set at P <0.05 for all tests.
Repeatability for each paired measurement set was excellent with intraclass correlation coefficients between 0.96 and 0.98. Operator error measurements varied between 0.04 and 0.27 mm. Mean biases, standard deviations, confidence intervals, and P values for the paired method comparisons are given in Table I . According to the 1-sample t test, there were no significant differences between any paired method; this indicated a certain level of agreement between the methods tested. The Bland-Altman plots of the method comparisons are shown in Figures 2 through 4 . In essence, the Bland-Altman analysis showed no fixed bias of 1 approach vs the other, and random errors were detected in all comparisons. Mean biases of iTero and Lythos scanner measurements, when compared with direct measurements, were −0.17 and −0.13 mm, respectively. The lowest minimum mean bias occurred for the comparison of the 2 intraoral scanners (−0.03 mm; 95% confidence interval and agreement limits of −0.8 and 0.7).