“…Dataset Approach Performance (%) et al [20] BMD & Dental panoramic radiography data CNN, VGG16, VGG16-TR, VGG16-TR-FT 0.840 Yamamoto et al [25] Hip radiography image data ResNet (18, 34, 50, 101, 152) 0.809 Faysal et al [22] Clinical attributes data Cross-sectional study Not available Ali et al [9] Clinical attributes data Cross-sectional study Not available Wani et al [27] Knee x-ray image & Clinical attributes dataset AlexNet, VGGNet-16, ResNet, VGGNet-19 90.91 Iliou et al [28] Clinical attributes dataset RBFNetwork, NBTree, REPTree, LWL 71.74 Shim et al [23] KNHANES-V1 and V2 KNN, DT, RF, GBM, SVM, ANN, LR 74.9 Lin et al [29] Wan Fang Hospital, Taiwan clinical data ANN, LR, SVM, RF 75.00 Huang et al [30] CT scan images LR, RF, SVM, XGBoost, GNB, GBM 0.81 Bui et al [31] Health Information System Clinical database LR, SVM, NN, RF, OSTA 0.85 Kim et al [24] KNHANES-V1 SVM, RF, ANN, LR, OST 76.70 Devikanniga et al [26] Lumber spine & Femoral neck image data MBO-ANN for Lumber spine 97.9 Engels et al [21] Clinical attributes data LR, RUS Boost, Super learner, XGBoost 70.4 Bold indicates the results of the proposed method.…”