Local Mesh Density of Maxillary First Molar Occlusal Surfaces in an Intraoral Scanning and ROI-Processing Workflow

Abstract

Background/Objectives: Local mesh density characterizes the polygonal representation of digital surface models; however, reference values and clinically relevant thresholds for posterior occlusal meshes generated through intraoral scanning and subsequent software processing have not been established. This study aimed to characterize the occlusal mesh density of maxillary first molars using V/A as the primary descriptor and F/A as a topology-derived secondary descriptor, and to evaluate mean bilateral differences between teeth 16 and 26. Methods: Twenty participants with intact bilateral maxillary first molars were included in this prospective within-subject observational study. Full-arch scans were acquired with a Medit i900 intraoral scanner and exported as STL files. Occlusal regions of interest were isolated in Medit Design, exported as PLY files, and analyzed for surface area, vertex count, face count, vertices per unit area (V/A), and faces per unit area (F/A). Bilateral differences were evaluated using paired-sample t-tests. Results: Mean participant-level occlusal mesh density was 37.62 ± 2.13 vertices/mm2 and 73.13 ± 4.22 faces/mm2. No significant bilateral differences were found for surface area, vertex or face counts, V/A, or F/A (all p > 0.05). For V/A, the mean right–left difference was 0.48 vertices/mm2, with Bland–Altman 95% limits of agreement from −3.87 to 4.82 vertices/mm2. In a descriptive cross-study comparison involving independent participant samples, the occlusal mesh-density values observed in the present study were higher than the buccal-surface values reported previously. Conclusions: Under the specified Medit i900 acquisition and Medit Design ROI-processing workflow, no statistically significant mean right–left differences in local mesh density were detected. The reported V/A and F/A values provide workflow-specific baseline descriptors of the final exported PLY mesh structure and should not be interpreted as measures of scanning accuracy. Their relationship to downstream CAD, AI-based design, or clinical outcomes requires direct investigation.