Related Experiment Video
Updated: Sep 3, 2025
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.9K
Collaborative deep learning model for tooth segmentation and identification using panoramic radiographs
Geetha Chandrashekar1, Saeed AlQarni2, Erin Ealba Bumann3
1Department of Computer Science Electrical Engineering, University of Missouri(2), Kansas City, MO, USA.
Computers in Biology and Medicine
|July 22, 2022
Summary
This study introduces a collaborative deep learning model for analyzing panoramic dental radiographs. The novel approach significantly improves tooth segmentation and identification accuracy, aiding dental treatment planning.
Area of Science:
- Dentistry
- Artificial Intelligence
- Medical Imaging
Background:
- Panoramic radiographs are crucial for dental treatment planning, but manual analysis can lead to diagnostic inconsistencies.
- Deep learning models offer advancements in dental anomaly detection, yet collaborative approaches are underexplored.
Purpose of the Study:
- To develop a novel collaborative deep learning technique for enhanced tooth segmentation and identification from panoramic radiographs.
- To aggregate independently developed models to improve learning performance and diagnostic accuracy.
Main Methods:
- Developed independent tooth segmentation and identification models for panoramic radiographs.
- Created a collaborative learning framework to integrate these individual models.
- Evaluated the model's performance in recognizing and numbering up to 32 teeth.
Main Results:
- The collaborative model achieved significantly higher accuracy (98.77% for segmentation, 98.44% for recognition) compared to individual models (96%, 91%).
- Outperformed existing state-of-the-art methods in tooth segmentation and identification.
- Demonstrated effectiveness across diverse dental scenarios, including complex cases with missing teeth or implants.
Conclusions:
- Collaborative learning enhances the performance of deep learning models for dental radiographic analysis.
- The proposed technique offers a more reliable and accurate method for tooth segmentation and identification, supporting better dental treatment planning.
