Table 9.

AI outcomes in oral and maxillofacial surgery.

Target	AI Model	Sample	Results	Study
Lower-third-molar treatment-planning decisions	Neural networks	Data from 119 patients	Sensitivity of 0.78, which was slightly lower than the oral surgeon’s (0.88), but the difference was not significant, and a specificity of 0.98, which was lower than the oral surgeon’s (0.99) (p = NS).	Brickley and Shepherd (1996) [192]
To predict postoperative facial swelling following impacted mandibular third molar extraction	ANN	Data from 400 patients	This AI-based algorithm was 98% reliable in forecasting facial swelling after extraction of impacted third molar teeth.	Zhang et al., (2018) [193]
Ameloblastoma and keratocystic odontogenic tumor diagnosis	CNN	400/100 panoramic X-rays	The CNN had 81.8% sensitivity, 83.3% specificity, 83.0% accuracy, and a diagnostic time of 38 s, respectively.	Poedjiastoeti and Suebnukarn (2018) [194]
Evaluation of maxillary sinusitis on panoramic radiography	CNN	Panoramic X-rays from 400 maxillary sinusitis patients/400 healthy subjects	Accuracy of 87.5%, sensitivity of 86.7%, specificity of 88.3%, and area under the curve of 0.875 were obtained by the model.	Murata et al., (2019) [195]
Periapical disease detection	ANN	2902 panoramic X-rays	The deep learning method outperformed 14 of the 24 surgeons in the sample, with an average accuracy of 0.60 (0.04).	Endres et al., (2020) [196]
Automated detection of cyst and tumors of the jaw	CNN	1602 lesions on panoramic X-rays	Comparable with expert dentists.	Yang et al., (2020) [197]

AI, artificial intelligence; ANN, artificial neural network; CNN, convolutional neural network; NS, non-significant.