Related Experiment Video
Updated: May 28, 2025
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.7K
Deep learning-based prediction of possibility for immediate implant placement using panoramic radiography
Sae Byeol Mun1, Hun Jun Lim2, Young Jae Kim3
1Department of Health Sciences and Technology, GAIHST, Gachon University, Incheon, 21999, Republic of Korea.
Scientific Reports
|February 12, 2025
Summary
Deep learning accurately predicts immediate dental implant placement success using panoramic radiographs. This AI approach aids clinicians in determining implant suitability after tooth extraction.
Area of Science:
- Dentistry
- Artificial Intelligence
- Medical Imaging
Background:
- Immediate implant placement offers benefits but requires careful case selection.
- Predicting successful immediate implant placement from pre-extraction imaging is crucial.
Purpose of the Study:
- To evaluate the feasibility of deep learning models for predicting immediate implant placement.
- To assess the accuracy of various deep learning architectures in this task.
Main Methods:
- Trained and tested six deep learning models (DenseNet121, ResNet18/101, ResNeXt101, InceptionNetV3/ResNetV2) on 874 panoramic radiographs.
- Utilized a dataset divided into immediate implant placement possible and difficult groups.
- Preprocessed dental images for model training and validation.
Main Results:
- All evaluated deep learning models achieved high performance metrics.
- Sensitivity, precision, accuracy, balanced accuracy, and F1-scores exceeded 0.90 for all models.
- Demonstrated strong predictive capabilities for immediate implant placement feasibility.
Conclusions:
- Deep learning models can accurately predict the possibility of immediate implant placement.
- AI-driven analysis of panoramic radiographs shows promise for improving dental implant planning.
