Analysis of linear measurements on 3D surface models using CBCT data segmentation obtained by automatic standard pre-set thresholds in two segmentation software programs: an in vitro study

• Published in: Clinical oral investigations Vol. 20; no. 1; pp. 179 - 185
• Main Authors: Poleti, Marcelo Lupion, Fernandes, Thais Maria Freire, Pagin, Otávio, Moretti, Marcela Rodrigues, Rubira-Bullen, Izabel Regina Fischer
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01-01-2016; Springer Nature B.V
• Subjects: Cone-Beam Computed Tomography; Dentistry; Humans; Imaging, Three-Dimensional; In Vitro Techniques; Mandible - diagnostic imaging; Medicine; Original Article; Radiographic Image Interpretation, Computer-Assisted; Reproducibility of Results; Software; Mandible; Dimensional measurement accuracy; Cone beam computed tomography; Three-dimensional image; Segmentation
• ISSN: 1432-6981; 1436-3771
• DOI: 10.1007/s00784-015-1485-5

Objective
The aim of this in vitro study was to evaluate the reliability and accuracy of linear measurements on three-dimensional (3D) surface models obtained by standard pre-set thresholds in two segmentation software programs.
Materials and methods
Ten mandibles with 17 silica markers were scanned for 0.3-mm voxels in the i-CAT Classic (Imaging Sciences International, Hatfield, PA, USA). Twenty linear measurements were carried out by two observers two times on the 3D surface models: the Dolphin Imaging 11.5 (Dolphin Imaging & Management Solutions, Chatsworth, CA, USA), using two filters(Translucent and Solid-1), and in the InVesalius 3.0.0 (Centre for Information Technology Renato Archer, Campinas, SP, Brazil). The physical measurements were made by another observer two times using a digital caliper on the dry mandibles.
Results
Excellent intra- and inter-observer reliability for the markers, physical measurements, and 3D surface models were found (intra-class correlation coefficient (ICC) and Pearson’s
r
 ≥ 0.91). The linear measurements on 3D surface models by Dolphin and InVesalius software programs were accurate (Dolphin Solid-1 > InVesalius > Dolphin Translucent). The highest absolute and percentage errors were obtained for the variable R1–R1 (1.37 mm) and MF–AC (2.53 %) in the Dolphin Translucent and InVesalius software, respectively.
Conclusion
Linear measurements on 3D surface models obtained by standard pre-set thresholds in the Dolphin and InVesalius software programs are reliable and accurate compared with physical measurements.
Clinical relevance
Studies that evaluate the reliability and accuracy of the 3D models are necessary to ensure error predictability and to establish diagnosis, treatment plan, and prognosis in a more realistic way.