Table 8.

CAD–CAM selected studies.

Authors Name and Year	Methods	Results	Authors Suggestions/Conclusions
Adel et al., (2018) [111]	SVM, ORB	Accuracy = 92.8%	Regarding the detection of oral epithelial dysplasia, this approach had the highest success rates.
Chatterjee et al., (2018) [112]	SVM, k nearest neighbor, random forest.	Accuracy = 90%	Predictive classifiers are better able to distinguish between illness and control groups when statistical and cytomorphometric features are combined.
Xu et al., (2018) [114]	Customized CNN	Accuracy = 99.06%	It directly satisfies the industrial clinical treatment demands and is also robust to any possible foreign matters on dental model surface.
Tian et al., (2019) [115]	Sparse voxel octree and 3D CNN	Accuracy = 95.96%	the proposed method has great application potential in the computer-assisted orthodontic treatment diagnosis.