Artificial Intelligence in Fractured Dental Implant Detection and Classification: Evaluation Using Dataset from Two Dental Hospitals

Abstract

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.

Deep learning improves implant classification by dental professionals: a multi-center evaluation of accuracy and efficiency

“…In a recent study, an automated DL model showed excellent accuracy performance in the detection and classification of DISs using dental radiographic images [ 21 24 ]. The automated DL model using periapical images showed more reliable accuracy performance in the detection (AUC=0.984; 95% CI, 0.900–1.000) and classification (AUC=0.869; 95% CI, 0.778–0.929) of fractured implants compared to those of pre-trained and fine-tuned VGGNet-19 and GoogLeNet models [ 24 ].…”

“…In a recent study, an automated DL model showed excellent accuracy performance in the detection and classification of DISs using dental radiographic images [ 21 24 ]. The automated DL model using periapical images showed more reliable accuracy performance in the detection (AUC=0.984; 95% CI, 0.900–1.000) and classification (AUC=0.869; 95% CI, 0.778–0.929) of fractured implants compared to those of pre-trained and fine-tuned VGGNet-19 and GoogLeNet models [ 24 ]. In addition, the automated DL model for panoramic and periapical images has shown excellent accuracy performance (AUC=0.954; 95% CI, 0.933–0.970), and results comparable to or better than those of dental professionals including board-certified periodontists, periodontology residents, and dentists not specialized in implantology ( P <0.05) [ 21 ].…”

Deep learning in periodontology and oral implantology: A scoping review

“…Third, deep learning models were also used to detect dental implant complications, including bone loss 52 and implant fracture. 53 Routine detection of early bone loss on radiographs can support early intervention, and current models showed high accuracy 54 when measured against clinicians. 52 The same-high accuracywas found for the detection of fractured implants, which may lead to post-traumatic and inflammatory responses and, consequently, severe bone loss around the implant.…”

“…Third, deep learning models were also used to detect dental implant complications, including bone loss 52 and implant fracture 53 . Routine detection of early bone loss on radiographs can support early intervention, and current models showed high accuracy 54 when measured against clinicians 52 .…”

Aplicación de la inteligencia artificial en Odontología

“…Así mismo, para detectar periimplantitis Cha et al 15 utilizaron CNN, no encontrando diferencia significativa entre el modelo CNN y el de periodoncistas. También se ha analizado la detección y clasificación de fracturas de los implantes utilizando 3 DCNN (convolutional neural network (DCNN) architectures (VGGNet-19, GoogLeNet, Inception.V3 and automated DCNN) 16 con una precisión de 0,80. A diferencia de los estudios anteriores, Ha et al 17 utilizaron SVM con LOOCV (Support vector machine with LOOCV) para establecer los factores que influyen en el pronóstico del implante estableciéndose que la posición mesiodistal del mismo es determinante.…”