THE EVALUATION OF MORPHOMETRY OF THREE-DIMENSIONAL RADIUS BONE MODELS USING OPEN-SOURCE MEDICAL PROGRAMS

Authors : Alperen Sarıtaş, Abdülkadir Bilir, Ozan Turamanlar, Özgür Verim, Çiğdem Özer Gökaslan, Gülan Albaş Kurt
Pages : 579-587
Doi:10.46519/ij3dptdi.1697749
View : 61 | Download : 147
Publication Date : 2025-12-28
Article Type : Research Paper
Abstract :The use of three-dimensional (3D) models, created using data obtained from radiological images, has significantly increased in recent years across the fields of medicine and health. The digitization of these models primarily utilizes open-source or commercial software. However, while the use of commercial software presents a significant economic burden, questions remain regarding model accuracy and output quality in open-source solutions. Therefore, this study aims to evaluate the morphometric accuracy of 3D radius models created using open-source medical software (InVesalius®, ITK-SNAP®, Seg3D®, and 3D Slicer®) by comparing them with gold standard (dry bone) measurements. Computed Tomography images of 15 dry human radius bones were used to generate the 3D digital models. These images were used to obtain 3D digital models via four different open-source software programs. Model lengths were calculated using MeshLab®, and volumes were calculated using Mimics® software. For gold standard comparison, the actual bone lengths were measured using digital calipers, and volumes were measured based on the Archimedes Principle.As a result, successful 3D digital radius models were created with all four programs. When the obtained measurements were compared with the gold standard values, no statistically significant difference was found between the programs (p > 0.05). Nevertheless, only the 3D Slicer® software demonstrated a high level of agreement in volume measurements (Cronbach\\\'s Alpha: 0.996; 95% CI: 0.988–0.999), standing out among the open-source medical software options.
Keywords : 3D Modeling, Bland-Altman Analysis, Morphometry, Software, Radius

ORIGINAL ARTICLE URL