作者: Antonino Lo Giudice , Vincenzo Quinzi , Vincenzo Ronsivalle , Marco Farronato , Carmelo Nicotra
关键词:
摘要: The aim of this study was to assess the accuracy 3D rendering mandibular condylar region obtained from different semi-automatic segmentation methodology. A total 10 Cone beam computed tomography (CBCT) were selected perform condyles by using three free-source software (Invesalius, version 3.0.0, Centro de Tecnologia da Informacao Renato Archer, Campinas, SP, Brazil; ITK-Snap, version2.2.0; Slicer 3D, 4.10.2) and one commercially available Dolphin (Dolphin Imaging, 11.0, Chatsworth, CA, USA). same models also manually segmented (Mimics, 17.01, Materialise, Leuven, Belgium) set as ground truth. evaluated (1) comparing volume each rendered model with that manual segmentation, (2) deviation analysis those segmentation. No significant differences found in volumetric dimensions among tested (p > 0.05). However, color-coded map showed underestimation ITK-Snap overestimation Invesalius. Excellent reliability for both intra-observer inter-observer readings. Despite excellent reliability, present findings suggest data morphology should be taken caution when an accurate definition boundaries is required.