Related Experiment Video
Updated: Aug 5, 2026
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
ImplantPlanNet: A deep learning framework for automatic implant planning from preoperative CBCT images
Juan Yang1, Qihang Liu2, Hongjie Yang3
1National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University, Chengdu, Sichuan 610065, China.
Journal of Dentistry
|July 27, 2026
Summary
This study introduces ImplantPlanNet, an AI tool for automatic dental implant planning using cone-beam computed tomography (CBCT) scans. The system shows promise in streamlining the planning process for single-tooth replacements.
Area of Science:
- Artificial Intelligence in Dentistry
- Medical Imaging Analysis
- Surgical Planning Technologies
Background:
- Preoperative implant planning using cone-beam computed tomography (CBCT) is crucial for prosthetically driven treatment.
- Current planning methods are time-consuming and require significant expertise.
- Automating initial implant planning can improve efficiency and accessibility.
Purpose of the Study:
- To develop and evaluate ImplantPlanNet, an automated framework for initial dental implant planning.
- To assess the accuracy and feasibility of AI-driven implant planning for single-tooth missing scenarios.
- To compare AI-generated plans with specialist-created plans.
Main Methods:
- ImplantPlanNet utilizes candidate localization, patch extraction, and pose estimation from CBCT images.
- The framework estimates implant position, long-axis direction, length, and diameter.
- A dataset of 144 CBCT scans was used for training and testing, with comparisons based on 3D deviations and classification accuracy.
Main Results:
- Internal and external 3D coronal deviations were approximately 1.5-1.7 mm, and 3D apical deviations were 1.8-2.2 mm.
- Angular deviation was around 5.6-6.9 degrees.
- Length classification accuracy was 65%, and diameter accuracy ranged from 75-100%, with comparable safety distances to specialist plans.
Conclusions:
- ImplantPlanNet demonstrates feasibility for generating automatic initial implant plans from CBCT data.
- The AI framework can assist clinicians by providing initial planning proposals for single-tooth replacements.
- This technology has the potential to support clinician-supervised planning workflows.
