Table 6 Comparisons are the made in the multi-classification mode, evaluation metrics such as accuracy, precision, recall/sensitivity, specificity, and F1 score.
Method | Accuracy | Average Precision | Average Recall | Average Specificity | Average F1-Score |
|---|---|---|---|---|---|
Wei Lu51 Using 101-layers Resnet | 0.9520 | NA | NA | 0.9730 | NA |
Leyuan Fang52 Lesion-Aware Convolutional Neural Network (LACNN) | 0.9750 | 0.9690 | NA | 0.9830 | NA |
Kermany53 Using (DL) Framework | 0.9610 | 0.9610 | 0.9613 | 0.9870 | 0.9610 |
Samra Naz45 Using (SMV) And (KNN) | 0.7925 | NA | NA | 0.933 | NA |
Nugroho. (DNN) Using ResNet50 Model | 0.8926 | 0.91 | 0.89 | NA | 0.89 |
Nugroho. (DNN) Using DenseNet-169 Model | 0.8802 | 0.90 | 0.88 | NA | 0.88 |
Najeeb. Using (CNN) | 0.9566 | 0.9592 | 0.9566 | 0.9855 | 0.9563 |
Saja Mahdi Hussein41 Using CNN | 0.9821 | 0.9800 | 0.9910 | NA | 0.9860 |
Xuan Huang43 Using Global Attention Block (GABNet) | 0.9650 | NA | 0.9650 | 0.9883 | 0.9650 |
Ali Serener and Sertan Serte19 Deep Learning (AlexNet) | 0.938 | NA | 0.804 | 0.983 | NA |
Parsa Riazi36 (DNN) Using CNN Architecture | 0.9871 | 0.9577 | 0.9855 | 0.9876 | 0.9714 |
Proposed Ensemble Model(RBLTL + InceptionV3) | 0.9875 | 0.9885 | 0.9877 | 0.9875 | 0.9875 |
Proposed Ensemble Model(RBLTL + DenseNet201) | 0.9890 | 0.9932 | 0.9906 | 0.9909 | 0.9890 |
Proposed Ensemble Model(RBLTL + InceptionResNetV2) | 0.9920 | 0.9960 | 0.9946 | 0.9934 | 0.9920 |
