Instance Segmentation and Teeth Classification in Panoramic X-rays

2024-06-06

arXiv.org

AI Summary

Concurrent optimization of tooth segmentation and detection in panoramic radiographs remains challenging due to inherent discrepancies between bounding-box-based detection and pixel-level segmentation tasks. Method: We propose BB-UNet, a novel end-to-end architecture that pioneers YOLOv8-guided U-Net segmentation—integrating object detection with instance-aware pixel-level segmentation via joint mAP and Dice coefficient optimization. Contribution/Results: We introduce the first dental instance segmentation dataset comprising 425 panoramic images, each annotated with both bounding boxes and high-fidelity polygon masks. Experiments demonstrate a 3% improvement in classification mAP and a 10–15% gain in Dice score over standard U-Net, with markedly enhanced robustness for heterogeneous teeth (e.g., molars and incisors). BB-UNet establishes a unified, high-accuracy, and interpretable framework for AI-assisted dental diagnosis.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Solvers and Tools

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Teeth segmentation and recognition are critical in various dental applications and dental diagnosis. Automatic and accurate segmentation approaches have been made possible by integrating deep learning models. Although teeth segmentation has been studied in the past, only some techniques were able to effectively classify and segment teeth simultaneously. This article offers a pipeline of two deep learning models, U-Net and YOLOv8, which results in BB-UNet, a new architecture for the classification and segmentation of teeth on panoramic X-rays that is efficient and reliable. We have improved the quality and reliability of teeth segmentation by utilising the YOLOv8 and U-Net capabilities. The proposed networks have been evaluated using the mean average precision (mAP) and dice coefficient for YOLOv8 and BB-UNet, respectively. We have achieved a 3% increase in mAP score for teeth classification compared to existing methods, and a 10-15% increase in dice coefficient for teeth segmentation compared to U-Net across different categories of teeth. A new Dental dataset was created based on UFBA-UESC dataset with Bounding-Box and Polygon annotations of 425 dental panoramic X-rays. The findings of this research pave the way for a wider adoption of object detection models in the field of dental diagnosis.

Problem

• Simultaneous tooth segmentation and detection in panoramic X-rays
• Improving accuracy of tooth classification and segmentation
• Leveraging U-Net and YOLOv8 for robust dental diagnostics

Innovation

• Combines U-Net and YOLOv8 for segmentation and detection
• Uses bounding box priors to guide dental segmentation
• Achieves higher accuracy in teeth detection and segmentation