Purpose <p>To comparatively evaluate mesh quality metrics in PLY, STL, and OBJ files derived from native intraoral scans and exported using different mesh processing software, with emphasis on computational performance, geometric fidelity, and mesh resolution.</p> Methods <p>Fifty maxillary and mandibular full-arch intraoral scan datasets were acquired using a Trios 4 intraoral scanner under clinical conditions. Native datasets were exported directly in PLY format without decimation, smoothing, or mesh optimization and designated as reference files. Each reference dataset was converted into PLY, STL, and OBJ formats using four mesh processing software packages (Meshmixer, Blender, MeshLab, and CloudCompare) under default export settings without mesh-altering procedures. Quantitative mesh evaluation was performed in MeshLab, including computational performance (loading time, export time, and file size), geometric accuracy (bounding box dimensions and surface area), and mesh resolution (total edge count and average edge length). Statistical analyses were conducted to compare differences among file formats and software platforms.</p> Results <p>Computational performance differed significantly across file formats (<i>p</i> &lt; 0.001). Native PLY files demonstrated the fastest loading (median 293 MS) and export times (median 627 MS), whereas OBJ files exhibited the longest export durations (median 51,639 MS) and largest file sizes. STL files showed intermediate performance. In contrast, bounding box dimensions and surface area measurements showed no significant differences between reference and converted files (<i>p</i> = 1.000), indicating preservation of scale and anatomical coverage. Edge-based metrics also revealed no significant differences in total edge count or average edge length (<i>p</i> = 1.000), confirming maintenance of mesh resolution and tessellation density.</p> Conclusion <p>File format significantly influences computational efficiency but does not affect geometric accuracy or mesh resolution when default export parameters are applied. PLY format demonstrated superior workflow performance while maintaining equivalent geometric and topological integrity across software platforms.</p> Clinical significance <p>Under the specific conditions of this study, PLY, STL, and OBJ formats preserved global geometric properties during conversion, while PLY demonstrated superior computational efficiency and smaller file size. Within these constraints, the PLY format may be considered a preferred export option for intraoral scan data, as it can help streamline digital workflows in prosthodontics, orthodontics, and restorative dentistry, reducing technical bottlenecks and improving efficiency. However, clinicians and dental technicians may still exchange PLY, STL, and OBJ formats without concern for scaling distortion, and caution is warranted when generalizing these findings to other systems, non-default settings, or workflows involving mesh modification.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Mesh quality comparison of common three file formats derived from intraoral scans

  • Mahmoud E. Elbashti,
  • Gülce Çakmak,
  • Marta Macarena Paz-Cortes,
  • Nadin Al-Haj Husain,
  • Samir Abou-Ayash,
  • Andrea Roccuzzo,
  • Pedro Molinero-Mourelle

摘要

Purpose

To comparatively evaluate mesh quality metrics in PLY, STL, and OBJ files derived from native intraoral scans and exported using different mesh processing software, with emphasis on computational performance, geometric fidelity, and mesh resolution.

Methods

Fifty maxillary and mandibular full-arch intraoral scan datasets were acquired using a Trios 4 intraoral scanner under clinical conditions. Native datasets were exported directly in PLY format without decimation, smoothing, or mesh optimization and designated as reference files. Each reference dataset was converted into PLY, STL, and OBJ formats using four mesh processing software packages (Meshmixer, Blender, MeshLab, and CloudCompare) under default export settings without mesh-altering procedures. Quantitative mesh evaluation was performed in MeshLab, including computational performance (loading time, export time, and file size), geometric accuracy (bounding box dimensions and surface area), and mesh resolution (total edge count and average edge length). Statistical analyses were conducted to compare differences among file formats and software platforms.

Results

Computational performance differed significantly across file formats (p < 0.001). Native PLY files demonstrated the fastest loading (median 293 MS) and export times (median 627 MS), whereas OBJ files exhibited the longest export durations (median 51,639 MS) and largest file sizes. STL files showed intermediate performance. In contrast, bounding box dimensions and surface area measurements showed no significant differences between reference and converted files (p = 1.000), indicating preservation of scale and anatomical coverage. Edge-based metrics also revealed no significant differences in total edge count or average edge length (p = 1.000), confirming maintenance of mesh resolution and tessellation density.

Conclusion

File format significantly influences computational efficiency but does not affect geometric accuracy or mesh resolution when default export parameters are applied. PLY format demonstrated superior workflow performance while maintaining equivalent geometric and topological integrity across software platforms.

Clinical significance

Under the specific conditions of this study, PLY, STL, and OBJ formats preserved global geometric properties during conversion, while PLY demonstrated superior computational efficiency and smaller file size. Within these constraints, the PLY format may be considered a preferred export option for intraoral scan data, as it can help streamline digital workflows in prosthodontics, orthodontics, and restorative dentistry, reducing technical bottlenecks and improving efficiency. However, clinicians and dental technicians may still exchange PLY, STL, and OBJ formats without concern for scaling distortion, and caution is warranted when generalizing these findings to other systems, non-default settings, or workflows involving mesh modification.