Validation of model analysis according to the severity of crowding of digital models obtained by intraoral and model scanners

Abstract

In this study, the correspondence between linear measurements taken from cast models, cast-scan models, and intraoral-scan models in patients who had various degrees of tooth crowding was assessed. Fifty-eight patients who had undergone both intraoral scans for diagnosis and for whom cast models were made for indirect bonding were included. Cast models were divided into two groups depending on the amount of crowding, as determined by initial caliper-based measurements. Cast models with crowding of less than 3.0 mm were included in the mild crowding (MC) group, and those with crowding of more than 4.5 mm were included in the moderate to severe crowding (SC) group. Twenty maxillary models and 20 mandibular models were randomly chosen from each of the groups and digital scans were performed. For the three types of models, i.e., intraoral scan digital model (IS), cast model (C), and cast-scan digital model (CS), linear measurement was performed twice by one graduate student. First, the reproducibility of the repeated linear measurements from each model was estimated. The mean values of repeated linear measurements from each model were compared to evaluate the precision of the measurements. We found that linear measurements made using digital calipers on a cast model and on the relevant software were reproducible. The mean differences ranged from -0.01 mm to 0.12 mm on the mesio-distal width of teeth, -0.30 mm to 0.45 mm in the arch width, and 0.14 to 1.35 mm in the arch length. In the MC group, there was a significant difference in the amount of crowding between cast and intraoral scans of the maxilla (mean difference: 0.78 mm; p < 0.05). In the SC group, there was a significant difference in the amount of crowding on both the maxilla and the mandible between the cast and intraoral scans (mean difference: 0.86 mm and 0.51 mm, respectively; p < 0.05). There was no interaction between the amount of crowding and the precision of the linear measurements in the three models (p > 0.05). Thus, there was no significant difference in most linear measurements between digital models and the cast model. There were significant differences in crowding, although these were not clinically meaningful, and there was no relationship to the severity of crowding. Thus, as with cast models, digital models can also be used for measuring crowding for orthodontic diagnoses in both patients with mild and those with severe crowding. However, the tendency for detection of less crowding in digital models should be considered