Photogrammetry as an alternative for acquiring digital dental models: A proof of concept

• Published in: Medical hypotheses Vol. 128; pp. 43 - 49
• Main Authors: Stuani, Vitor T., Ferreira, Rafael, Manfredi, Gustavo G.P., Cardoso, Matheus V., Sant'Ana, Adriana C.P.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.07.2019
• Subjects: Calibration; Dental Models; Dental photography; Humans; Image Processing, Computer-Assisted; Imaging, Three-Dimensional; Incidence; Molar - anatomy & histology; Patient-specific modeling; Photogrammetry; Photogrammetry - methods; Proof of Concept Study; Reproducibility of Results; Software; Technology; Tooth - anatomy & histology; Dental photography; Patient-specific modeling; Technology; Photogrammetry
• DOI: 10.1016/j.mehy.2019.05.015
• ISSN: 0306-9877

Photogrammetry is a mathematical technique that generates three-dimensional coordinates of specific points identified from multiple images of the same object obtained at different angles. This technique may be a low-cost alternative for traditional scanning. The objective of this proof of concept was to evaluate the accuracy and precision in obtaining digital models (DM) from a plaster model (PM) using photogrammetry. Five DM were generated from 50 photographs taken surrounding the PM. The photographs were taken by a single operator on five consecutive days using natural light. The images obtained were processed on 3DF Zephyr Free software. The height and width of all teeth were recorded on both PM and DM, as well as the distance between the canine cusps (C–C) and between the mesiobuccal cusps of the first molars (1 M–1 M). For the PM the measurements were taken with a digital caliper, whereas the DM was measured using the software Blender. The DM and PM measurements presented a limit of agreement between −0.433 and 0.611 mm. The accuracy of DM measurements showed a SD of ±0.171 mm and a repeatability coefficient of 0.474. In the superimposition of all DM, it was possible to notice a greater discrepancy in the posterior region of the arch and palate, but this difference decreased when the region was segmented. It can be concluded that photogrammetry appears to be a viable technique for the digitization of dental models. Further studies need to be performed to evaluate its clinical application.