Valentinitsch A, et al. - Authors assessed 154 candidates using 3D texture features and regional volumetric bone mineral density (vBMD) using multi-detector computed tomography (CT) images to estimate the opportunistic osteoporosis screening. They dichotomized the individuals considering prevalent vertebral osteoporotic fractures. They obtained gray-level co-occurrence matrix Haralick features (HAR), histogram of gradients (HoG), local binary patterns (LBP), and wavelets (WL) for 3D texture analysis. They observed that the random forest classifier conferred a high discriminatory power (AUC=0.88). They found a significant contribution of parameters of all vertebral levels to this classification. They concluded the algorithm significantly higher as compared to that of volumetric global BMD alone (AUC=0.64).