Scholarlink

Background <p>This study aimed to perform the first multi-architecture comparison of pixel-level mandibular condyle segmentation on panoramic radiographs using transformer-based (RT-DETR), CNN-based (EfficientNet, Mask R-CNN, ConvNeXt), and YOLO-based (YOLOv9-Seg, YOLOv11-Seg) deep learning models.</p> Methods <p>A dataset of 1,300 panoramic radiographs (2,600 condyles) was retrospectively curated. Ground-truth masks were annotated by a primary radiologist and reviewed by a senior radiologist; inter-observer agreement was quantified on a blinded 10% subset (Dice: 0.92 ± 0.03). Six state-of-the-art architectures were trained and evaluated on a fixed test set. Performance was assessed using Intersection over Union (IoU), Dice Similarity Coefficient (DSC), precision, recall, and F1-score.</p> Results <p>All models achieved high segmentation accuracy, with DSC values ranging from 0.819 to 0.866. The transformer-based RT-DETR model showed the highest numerical DSC (0.866), IoU (0.764), and F1-score (0.866), indicating a balanced overall segmentation profile. Among the one-stage detectors, YOLOv9-Seg provided competitive results (DSC: 0.862) with high recall (0.902), outperforming CNN-based alternatives. YOLOv11-Seg showed high sensitivity but lower precision compared to other architectures.</p> Conclusions <p>Deep learning enables accurate and automated condylar segmentation on panoramic radiographs. While RT-DETR showed favorable anatomical fidelity for quantitative morphometry, YOLOv9-Seg presented a viable real-time alternative. This study establishes a benchmark for selecting segmentation architectures tailored to specific clinical needs in TMJ analysis.</p> Trial registration <p>Not applicable.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparative analysis of transformer, CNN, and YOLO architectures for mandibular condyle segmentation on panoramic radiographs: a deep learning benchmark

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月16日

查看全文
  • Serkan Yilmaz,
  • Hilal Peker Ozturk,
  • Hatice Seda Ozgedik,
  • Ismail Hakan Avsever,
  • Bugra Senel,
  • Murat Tasyurek,
  • Mehmet Hakan Kurt

摘要

Background

This study aimed to perform the first multi-architecture comparison of pixel-level mandibular condyle segmentation on panoramic radiographs using transformer-based (RT-DETR), CNN-based (EfficientNet, Mask R-CNN, ConvNeXt), and YOLO-based (YOLOv9-Seg, YOLOv11-Seg) deep learning models.

Methods

A dataset of 1,300 panoramic radiographs (2,600 condyles) was retrospectively curated. Ground-truth masks were annotated by a primary radiologist and reviewed by a senior radiologist; inter-observer agreement was quantified on a blinded 10% subset (Dice: 0.92 ± 0.03). Six state-of-the-art architectures were trained and evaluated on a fixed test set. Performance was assessed using Intersection over Union (IoU), Dice Similarity Coefficient (DSC), precision, recall, and F1-score.

Results

All models achieved high segmentation accuracy, with DSC values ranging from 0.819 to 0.866. The transformer-based RT-DETR model showed the highest numerical DSC (0.866), IoU (0.764), and F1-score (0.866), indicating a balanced overall segmentation profile. Among the one-stage detectors, YOLOv9-Seg provided competitive results (DSC: 0.862) with high recall (0.902), outperforming CNN-based alternatives. YOLOv11-Seg showed high sensitivity but lower precision compared to other architectures.

Conclusions

Deep learning enables accurate and automated condylar segmentation on panoramic radiographs. While RT-DETR showed favorable anatomical fidelity for quantitative morphometry, YOLOv9-Seg presented a viable real-time alternative. This study establishes a benchmark for selecting segmentation architectures tailored to specific clinical needs in TMJ analysis.

Trial registration

Not applicable.