Summary: | Several open-source 3D modeling software tools have recently been featured to reconstruct a 3D imagebased model using Structure from Motion (SfM) algorithms such as +IDonBlender. It is, therefore, necessary to evaluate the accuracy of the models extracted from this tool to prove effectiveness for different applications. This paper aims to evaluate the +IDonBlender methodology tool in rehabilitating facial deformations. +IDonBlender is an add-on tool programmed within Blender software and designed for medical applications. In this research, two individuals from different genders have volunteered to contribute to this study, one with severe facial deformation and the other is healthy with no defects. To assess the robustness of the derived digital model from the +IDonBlender methodology, Agisoft Metashape was used to process the data and create a reference model for comparison purposes. Various scenarios were followed for evaluation: (i) scaling the model based on reference ground truth points, (ii) scaling the model based on known real distance between two reference points, and (iii) cloud-to-mesh (C2M) analysis was applied to individual scenarios to assess the accuracy of the +IDonBlender model through cloud compare analysis. Next, the density analysis of the model was also used to investigate the model resolution for further post-processing and prototyping. The results show that the variation between the tested and the reference models in terms of coordinates and distance measurements was (0.470 mm) and (0.471mm), respectively. Despite the fact that the +IDonBlender model resolution was lower than the Metashape model, the overall results delivered acceptable geometric error values. Therefore, following visual and statistical point clouds data analysis, one can claim the robustness of the +IDonBlender 3D modeling tool for maxillofacial rehabilitation, surgical planning and other clinical applications. © 2021 University of Kerbala.
|