Fracture of a dental implant (DI) is a rare mechanical complication that is a critical cause of DI failure and explantation. The purpose of this study was to evaluate the reliability and validity of a three different deep convolutional neural network (DCNN) architectures (VGGNet-19, GoogLeNet Inception-v3, and automated DCNN) for the detection and classification of fractured DI using panoramic and periapical radiographic images. A total of 21,398 DIs were reviewed at two dental hospitals, and 251 intact and 194 fractured DI radiographic images were identified and included as the dataset in this study. All three DCNN architectures achieved a fractured DI detection and classification accuracy of over 0.80 AUC. In particular, automated DCNN architecture using periapical images showed the highest and most reliable detection (AUC = 0.984, 95% CI = 0.900–1.000) and classification (AUC = 0.869, 95% CI = 0.778–0.929) accuracy performance compared to fine-tuned and pre-trained VGGNet-19 and GoogLeNet Inception-v3 architectures. The three DCNN architectures showed acceptable accuracy in the detection and classification of fractured DIs, with the best accuracy performance achieved by the automated DCNN architecture using only periapical images.
Background: Various approaches are used for alveolar ridge preservation (ARP); however, there is no standard method or material. Purpose: To investigate the effect of ARP with a dense polytetrafluoroethylene (d‐PTFE) membrane and freeze‐dried irradiated allogenic bone for sockets with bone deficiency. Materials and Methods: Thirty‐four patients (with sockets exhibiting ≥3 mm hard tissue loss in ≥1 walls) were randomized to undergo natural socket healing (control) or ARP with a d‐PTFE membrane and freeze‐dried irradiated allogenic bone (test group). After 4 months, horizontal and vertical ridge changes were measured using cone beam computed tomography. Results: Ridge width at l mm below the ridge crest demonstrated significantly less change in the test group (median =2.3; Q1 = 0.6; Q3 = 4.3 mm) than in the control group (median =3.9; Q1 =2.6; Q3 = 7.8 mm; P = .021). There was no significant difference between the two groups in horizontal ridge changes at 3 and 5 mm below the crest or vertical changes ( P > .05). Requirement for bone augmentation at implant placement was significantly reduced in the test group compared to the control group ( P < .001). Conclusion: ARP with a d‐PTFE membrane and freeze‐dried irradiated allogenic bone substitute reduced horizontal bone resorption in sockets with bone deficiency.