Table 6.
AI outcomes in periodontics.
| Target | AI Model | Sample | Results | Study |
|---|---|---|---|---|
| To classify periodontitis by immune response profile to aggressive periodontitis or chronic periodontitis class. | MLP ANN | Data from 29 subjects | 90–98% accuracy | Papantonopoulos et al., (2014) [165] |
| Diagnosis of periodontal diseases. | ANNs, decision trees, and support vector machine | Data from 150 patients | Performance was 98%. The poorest correlation between input and output variables was found in ANN, and its performance was assessed to be 46%. | Ozden et al., (2015) [166] |
| To identify and predict periodontally compromised teeth. | CNN encoder + three dense layers | 1740 periapical X-rays | AUC of 73.4–82.6 [95% CI, 60.9–91.1] in predicting hopeless teeth. | J. H. Lee et al., (2018) [162] |
| To detect periodontal bone loss (PBL) on panoramic dental radiographs. | CNN + three dense layers | 85 panoramic X-rays | Predictive accuracy was determined to be 81%, which is similar to the examiners. | Krois et al., (2019) [159] |
| Pre-emptive detection and diagnosis of periodontal disease and gingivitis by using intraoral images. | Faster R-CNN | 134 photographs | Tooth detection accuracy of 100% to determine region of interest and 77.12% accuracy to detect inflammation. | Alalharith et al., (2020) [167] |
| Predicting periodontitis stage. | CNN | 340 periapical X-rays | Accuracy of 68.3% | Danks et al., (2021) [168] |
| Predicting immunosuppression genes in periodontitis. | DisGeNet, HisgAtlas | Saliva | Accuracy of 92.78% | Ning et al., (2021) [169] |
| Clinical, immune, and microbial profiling of peri-implantitis patients against health. | CNN FARDEEP | Metabolites | Successfully used in logistic regression of plaque samples. | Wang et al., (2021) [170] |
| Research trialing different methods of segmentation to assess plaque on photographs of tooth surfaces (including ‘dye labelling’). | CNN OCNet, Anet |
2884 photographs | AUC prediction of 87.11% for gingivitis and 80.11% for calculus. | Li et al., (2021) [171] |
AI, artificial intelligence; ANN, artificial neural network; AUC, area under the curve; CBCT, cone-beam computed tomography; CI, confidence interval; CNN, convolutional neural network.