Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Boundary feature fusion network for tooth image segmentation

Dongping Zhang, Zheng Li, Fangao Zeng, Yutong Wei

TL;DR

This work tackles boundary blur in tooth image segmentation by introducing BFFNet, a boundary feature fusion network that explicitly models tooth boundaries. It combines a ResNet-based backbone with three Boundary Feature Extraction Modules and three Feature Cross-Fusion Modules to fuse boundary detail with global semantics across multiple stages, supervised by a combined IoU and BCE loss. On the STS dataset, BFFNet outperforms four established baselines, achieving a high composite Score and strong boundary delineation, with ablation confirming the value of both BFEM and FCFM. The approach offers improved accuracy for clinical dental analysis and shows potential for broader biomedical image segmentation tasks where boundary precision is critical.

Abstract

Tooth segmentation is a critical technology in the field of medical image segmentation, with applications ranging from orthodontic treatment to human body identification and dental pathology assessment. Despite the development of numerous tooth image segmentation models by researchers, a common shortcoming is the failure to account for the challenges of blurred tooth boundaries. Dental diagnostics require precise delineation of tooth boundaries. This paper introduces an innovative tooth segmentation network that integrates boundary information to address the issue of indistinct boundaries between teeth and adjacent tissues. This network's core is its boundary feature extraction module, which is designed to extract detailed boundary information from high-level features. Concurrently, the feature cross-fusion module merges detailed boundary and global semantic information in a synergistic way, allowing for stepwise layer transfer of feature information. This method results in precise tooth segmentation. In the most recent STS Data Challenge, our methodology was rigorously tested and received a commendable overall score of 0.91. When compared to other existing approaches, this score demonstrates our method's significant superiority in segmenting tooth boundaries.

Boundary feature fusion network for tooth image segmentation

TL;DR

This work tackles boundary blur in tooth image segmentation by introducing BFFNet, a boundary feature fusion network that explicitly models tooth boundaries. It combines a ResNet-based backbone with three Boundary Feature Extraction Modules and three Feature Cross-Fusion Modules to fuse boundary detail with global semantics across multiple stages, supervised by a combined IoU and BCE loss. On the STS dataset, BFFNet outperforms four established baselines, achieving a high composite Score and strong boundary delineation, with ablation confirming the value of both BFEM and FCFM. The approach offers improved accuracy for clinical dental analysis and shows potential for broader biomedical image segmentation tasks where boundary precision is critical.

Abstract

Tooth segmentation is a critical technology in the field of medical image segmentation, with applications ranging from orthodontic treatment to human body identification and dental pathology assessment. Despite the development of numerous tooth image segmentation models by researchers, a common shortcoming is the failure to account for the challenges of blurred tooth boundaries. Dental diagnostics require precise delineation of tooth boundaries. This paper introduces an innovative tooth segmentation network that integrates boundary information to address the issue of indistinct boundaries between teeth and adjacent tissues. This network's core is its boundary feature extraction module, which is designed to extract detailed boundary information from high-level features. Concurrently, the feature cross-fusion module merges detailed boundary and global semantic information in a synergistic way, allowing for stepwise layer transfer of feature information. This method results in precise tooth segmentation. In the most recent STS Data Challenge, our methodology was rigorously tested and received a commendable overall score of 0.91. When compared to other existing approaches, this score demonstrates our method's significant superiority in segmenting tooth boundaries.
Paper Structure (16 sections, 11 equations, 6 figures, 4 tables)

This paper contains 16 sections, 11 equations, 6 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Proposed Method
  4. Boundary Feature Extraction Module
  5. Feature Cross-Fusion Module
  6. Loss Function
  7. Experiments
  8. Datasets and Comparative Models
  9. Evaluation metrics
  10. Implementation details
  11. Ablation study
  12. Results and discussion
  13. Quantitative results
  14. Qualitative results
  15. Limitation and Future Work
  16. ...and 1 more sections

Figures (6)

  • Figure 1: Overview of the proposed BFFNet architecture. It mainly consists of three boundary feature extraction modules and three feature cross-fusion modules.
  • Figure 2: Internal structure of BFEM.
  • Figure 3: Internal structure of FCFM.
  • Figure 4: Box plot under different module combinations.
  • Figure 5: Precision-Recall curves of BFFNet and other four models.
  • ...and 1 more figures