Multi-Phase Automated Segmentation of Dental Structures in CBCT Using a Lightweight Auto3DSeg and SegResNet Implementation

Computer Vision & MultiModal AI
Published: arXiv: 2508.12962v1
Authors

Dominic LaBella Keshav Jha Jared Robbins Esther Yu

Abstract

Cone-beam computed tomography (CBCT) has become an invaluable imaging modality in dentistry, enabling 3D visualization of teeth and surrounding structures for diagnosis and treatment planning. Automated segmentation of dental structures in CBCT can efficiently assist in identifying pathology (e.g., pulpal or periapical lesions) and facilitate radiation therapy planning in head and neck cancer patients. We describe the DLaBella29 team's approach for the MICCAI 2025 ToothFairy3 Challenge, which involves a deep learning pipeline for multi-class tooth segmentation. We utilized the MONAI Auto3DSeg framework with a 3D SegResNet architecture, trained on a subset of the ToothFairy3 dataset (63 CBCT scans) with 5-fold cross-validation. Key preprocessing steps included image resampling to 0.6 mm isotropic resolution and intensity clipping. We applied an ensemble fusion using Multi-Label STAPLE on the 5-fold predictions to infer a Phase 1 segmentation and then conducted tight cropping around the easily segmented Phase 1 mandible to perform Phase 2 segmentation on the smaller nerve structures. Our method achieved an average Dice of 0.87 on the ToothFairy3 challenge out-of-sample validation set. This paper details the clinical context, data preparation, model development, results of our approach, and discusses the relevance of automated dental segmentation for improving patient care in radiation oncology.

Paper Summary

Problem
The main problem addressed by this research is the need for efficient and accurate automated segmentation of dental structures in cone-beam computed tomography (CBCT) images. This is particularly important in radiation oncology, where accurate diagnosis and treatment planning are crucial for patients with head and neck cancer.
Key Innovation
The key innovation of this work is the development of a lightweight deep learning pipeline using the MONAI Auto3DSeg framework and a 3D SegResNet architecture. The pipeline is designed to be computationally efficient while achieving high accuracy in segmenting dental structures. The approach also involves preprocessing steps, such as image resampling and intensity clipping, to improve model performance.
Practical Impact
The practical impact of this research is the potential for automating dental segmentation in CBCT images, which can streamline clinical workflows and improve patient care. Specifically, the algorithm can be used to identify high-dose teeth and quantify patient-specific risk factors for osteoradionecrosis (ORN), a severe complication that can occur after radiation therapy. The goal is to integrate this technology into the clinical workflow for head and neck oncology, enabling automatic dental reports that flag high-dose teeth and inform personalized supportive care.
Analogy / Intuitive Explanation
Imagine trying to find a specific toy in a messy playroom. You need to look at the room as a whole, then zoom in on smaller areas until you find what you're looking for. That's similar to what this algorithm does - it looks at the entire CBCT image, then focuses on specific dental structures like teeth and nerves to segment them accurately. The analogy also highlights the importance of preprocessing (cleaning up the playroom) before trying to find the toy (segmenting the dental structures).
Paper Information
Categories:
cs.CV cs.AI
Published Date:

arXiv ID:

2508.12962v1

Quick Actions