Skip to main content
. 2022 Mar 15;19(6):3449. doi: 10.3390/ijerph19063449

Table 3.

Baseline characteristics of the studies included in the review by studying neural networks in orthodontics.

Study [Ref.] Year of Publication Type of Data Type of Neural Network Number of Database Accuracy of Neural Network
Auconi [33] 2015 Cephalometric records Fuzzy clustering repartition 54 cephalograms 83.3%
Peilini [34] 2019 Medical records ANN 302 patients 94.0% (extraction pattens); 92.8 % (anchorage patterns)
Bianchi [36] 2020 CBCT
blood serum saliva
clinical investigation
Logistic Regression, Random Forest, LightGBM, XGBoost 52 patients 82.3%
Muraev [39] 2020 Cephalometric records ANN 330 cephalograms 99.9%
Kök [40] 2021 Cephalometric and hand-wrist radiographs ANN 419 patients 94.27%