Prediction of adequate alveolar ridge contours for dental implant placement using an artificial intelligence-based algorithm.
Authors
Affiliations (5)
Affiliations (5)
- Graduate Institute of Clinical Dentistry, School of Dentistry, College of Medicine, National Taiwan University, Taipei, Taiwan.
- Department of Dentistry, National Taiwan University Hospital, Taipei, Taiwan.
- Muen Biomedical and Optoelectronics Technologies Inc., Taipei, Taiwan.
- Cramille Digital Laboratory, Taipei, Taiwan.
- School of Dentistry, College of Dental Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan.
Abstract
Accurate prediction of ridge contour following ridge augmentation is an integral part of planning dental implant therapy. This study aimed to develop an artificial intelligence (AI)-based algorithm to predict adequate alveolar ridge contours for accommodating implants based on preoperative cone-beam computed tomography (CBCT) images using a deep learning (DL) approach. A total of 96 mandibular CBCT images with minor and major defects were used, virtual implants with adequate ridge contours were added, and a U-Net 2.5D model was used for DL training. The results revealed that the predicted ridge contours achieved Dice Similarity Coefficients of 0.928 for minor defects and 0.932 for major defects. To further validate clinical reliability, preoperative CBCT images from a 53-year-old patient presenting with a thin edentulous ridge were used, and ridge augmentation was performed by adapting a titanium mesh (T-mesh) prefabricated on the predicted ridge contour model with freeze-dried bone allografts. A ridge width of 4-4.5 mm was obtained, and the dental implants were successfully installed after 8 months. An AI-based algorithm for predicting adequate ridge contours for accommodating dental implants was successfully established with acceptable clinical feasibility. This algorithm offers a transformative solution for prefabricating T-mesh and can be integrated as a part of a.