Deep learning-based prediction of osseointegration for dental implant using plain radiography

Oh et al. BMC Oral Health (2023) 23:208 https://doi.org/10.1186/s12903-023-02921-3 BMC Oral Health Open Access RESEARCH Deep learning-based prediction of osseointegration for dental implant using plain radiography Seok Oh1, Young Jae Kim1, Jeseong Kim2, Joon Hyeok Jung2, Hun Jun Lim2, Bong Chul Kim2* and Kwang Gi Kim1* Abstract Background In this study, we investigated whether deep learning-based prediction of osseointegration of dental implants using plain radiography is possible. Methods Panoramic and periapical radiographs of 580 patients (1,206 dental implants) were used to train and test a deep learning model. Group 1 (338 patients, 591 dental implants) included implants that were radiographed immediately after implant placement, that is, when osseointegration had not yet occurred. Group 2 (242 patients, 615 dental implants) included implants radiographed after confirming successful osseointegration. A dataset was extracted using random sampling and was composed of training, validation, and test sets. For osseointegration prediction, we employed seven different deep learning models. Each deep-learning model was built by performing the experiment 10 times. For each experiment, the dataset was randomly separated in a 60:20:20 ratio. For model evaluation, the specificity, sensitivity, accuracy, and AUROC (Area under the receiver operating characteristic curve) of the models was calculated. Results The mean specificity, sensitivity, and accuracy of the deep learning models were 0.780–0.857, 0.811–0.833, and 0.799–0.836, respectively. Furthermore, the mean AUROC values ranged from to 0.890–0.922. The best model yields an accuracy of 0.896, and the worst model yields an accuracy of 0.702. Conclusion This study found that osseointegration of dental implants can be predicted to some extent through deep learning using plain radiography. This is expected to complement the evaluation methods of dental implant osseointegration that are currently widely used. Keywords Dental Digital radiography, Deep learning, Artificial Intelligence, Dental Implant, Osseointegration, Oral Surgical Procedures *Correspondence: Bong Chul Kim Kwang Gi Kim 1 Gil Medical Center, Department of Biomedical Engineering, Gachon University College of Medicine, Incheon 21565, Korea 2 Department of Oral and Maxillofacial Surgery, Daejeon Dental Hospital, Wonkwang University College of Dentistry, Daejeon 35233, Korea © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Oh et al. BMC Oral Health (2023) 23:208 Background Dental implants are widely used for the rehabilitation of edentulous spaces, and osseointegration is essential for the success of dental implants. Various methods have been used to evaluate osseointegration [1–3]. However, the most widely used methods have the disadvantage of being invasive. Recently, deep learning has been actively applied to dentistry and maxillofacial surgery. The prediction of extraction difficulty and postoperative paresthesia in relation to the mandibular third molar has been described [4, 5]. Studies related to dentofacial dysmorphosis have also been reported [6]. This modality can also help predict the need [7, 8] and outcomes [9] of orthognathic surgery. Prediction of implant osseointegration through plain radiography and based on deep learning is a potential noninvasive modality; however, it has not yet been studied. Therefore, in this study, we investigated whether plain radiography can help predict osseointegration of dental implants. Methods Datasets In this study, panoramic and periapical radiographs of 580 patients (311 men, 269 women; age range, 21–78 years) with 1,206 dental implants, who visited the Daejeon Dental Hospital, Wonkwang University, between January 2015 and December 2018 for dental implant treatment, were used for the training and testing of a deep learning model. In this study, only dental implants placed by a single surgeon after confirming adequate bone healing 3 or more months after tooth extraction were included. Periotest was used to evaluate implant osseointegration 3 or more months after implant placement. All dental implants used in this study satisfied the following conditions: (1) no tenderness on palpation, percussion, or function; (2) no horizontal and/or vertical mobility; (3) no uncontrolled progressive bone loss; (4) no uncontrolled exudate; and (5) no alveolar bone loss around the implant. Implants for which additional bone grafting, including sinus elevation, was performed were excluded. The reverse torque test was performed to confirm successful osseointegration at the time of abutment connection. In all cases, it was confirmed that there was no such problem by observing a lapse of three years or more. Panoramic and/or periapical radiography were performed immediately after implant placement and after successful osseointegration. Panoramic views were obtained using Promax (Planmeca, Helsinki, Finland; current, 12 mA; voltage, 72 kV; exposure time, 15.8 s) or PCH-2500 (Vatech, Hwaseong, Korea; current, 10 mA; voltage, 72 kV; exposure time, 13.5 s). Planmeca ProX Page 2 of 7 (Planmeca, Helsinki, Finland) was used for periapical radiography. Group 1 (338 patients, 591 dental implants) included patients who underwent radiography immediately after implant placement, that is, when osseointegration had not yet occurred. Group 2 (242 patients, 615 dental implants) included patients who underwent radiography after confirming successful osseointegration. Preprocessing In this study, the data-cleaning process was not performed. Therefore, deep learning models were trained by raw data. The validation and test processes were also conducted by using raw data. For segmentation, a semi-automatic process, automatic segmentation based on Otsu’s method [10], was initially used for each dental implant. After the initial segmentation, additional manual correction was performed to clearly define the region of interest (ROI (...truncated)