Table 10 Comparative analysis of caries segmentation performance in terms of accuracy, sensitivity, and specificity, as reported in referenced studies.
References | Algorithms | Avg. Acc | Avg. Pre | Avg. Sen | Avg. Spec |
|---|---|---|---|---|---|
Duong et al.9 | SVM | 92.30% | – | 88.10% | 96.60% |
Yu et al.10 | ResNet-FPN and FCN | 95.00% | – | 89.83% | 96.00% |
Geetha et al.11 | Morphological and BPNN | 97.10% | – | – | – |
Cantu et al.12 | U-Net | 80.00% | – | 75.00% | 83.00% |
Vinayahalingam et al.13 | CNN with MobileNet V2 | 87.00% | – | 86.00% | 88.00% |
Lee et al.14 | CNN using U-Net | – | 63.29%, | – | – |
Mao et al.15 | AlexNet model | 90.30% | – | – | – |
Kuhnisch et al.17 | CNN | – | – | 89.60% | 94.30% |
Vimalarani et al.18 | DG-LeNet | 98.74% | – | 91.37% | 98.92%, |
Zhu et al.19 | Faster-RCNN | – | 73.49% | – | – |
Ramana et al.20 | Neural networks | 93.67% | – | 94.66% | 92.73% |
Imak et al.21 | Deep CNN | 99.13% | – | – | – |
Park et al.22 | U-Net, ResNet-18, Faster RCNN | 81.30% | 86.80% | 86.50% | – |
Kim et al.23 | DeNTNet | – | – | 77.00% | 95.00% |
Hung et al.24 | SVM | 97.10% | 95.10% | 99.60% | 94.30% |
Abdulaziz et al.25 | CNN | 97.07% | – | – | – |
Bayraktar et al.16 | YOLO and CNN | 94.59% | – | 72.26% | 98.19% |
Our proposed | FCM-FRWS and MDOT | 91.62% | 90.89% | 97.78% | 91.26% |
