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

The study demonstrates that Deep learning-based prediction of osseointegration of dental implants is possible using plain radiography. Key takeaways: Seven different deep learning models were employed and evaluated for osseointegration prediction, with mean specificity, sensitivity, accuracy, and AUROC values ranging from 0.780 to 0.857, 0.811 to 0.833, 0.799 to 0.836, and 0.890 to 0.922, respectively. The best-performing model achieved an accuracy of 0.896, while the worst-performing model achieved an accuracy of 0.702. This study suggests that deep learning has the potential to be a valuable tool in predicting dental implant osseointegration, which may aid in improving clinical decision-making and patient outcomes in implant dentistry.