The objective was to validate a semi-automated segmentation method for 3-dimensional (3D) reconstruction of the mandibular condyle from cone beam computed tomography (CBCT) data and illustrate its application in volumetric analysis of the condyle.Ten CBCT datasets were used to validate the proposed semi-automatic method for 3D rendering of mandibular condyles. First, a standardized orientation protocol of the skull was applied. After defining the volume of interest, a grey-scale cut-off value was selected to allow an automatic reconstruction of the condyle's surface. Subsequently, condylar contour was optimized manually. The whole process was repeated twice by 2 independent investigators. Volumetric measurements of the condyle were used as a measure of conformity between both investigators. The reproducibility of condylar volume reconstruction was excellent for intra-examiner measurements (CV = 3.65%, intraclass correlation coefficient = 0.97) and good for inter-examiner measurements (CV = 7.15%, ICC = 0.89). The overall mean time required for the segmentation process was 6.31 + 2.78 minutes. The proposed protocol provides an accurate and reproducible tool for 3D reconstruction of the mandibular condyle using CBCT data. Its implementation will enable adequate follow-up of morphological changes in bone tissue with a Hounsfield unit-based imaging segmentation method.