Scholarlink

<p>Early and reliable detection of subcentimeter pulmonary nodules remains a major bottleneck in low-dose CT–based lung cancer screening due to high miss rates, vascular-adhesion–induced false positives, and insufficient multi-scale feature fusion. To address these limitations, we propose DA<sup>3</sup>-LungNet, a multi-stage deep framework with adaptive attention that integrates dynamic candidate generation, attention-guided fine segmentation, and 3D contextual verification. The first stage introduces dynamic focal RetinaNet with cross-scale feature interaction to maximize recall under extreme class imbalance. The second stage employs an attention-guided U-Net++ augmented with a dense attention bridging module (DABM) for enhanced edge representation and gradient propagation. The final stage integrates a 3D contextual pyramid module (3D-CPM) to model inter-slice spatial continuity and suppress vascular false positives. Extensive experiments on LIDC-IDRI and DSB2017 datasets demonstrate that DA<sup>3</sup>-LungNet achieves state-of-the-art performance with 92.7% Dice, 93.4% sensitivity for &lt; 6&#xa0;mm nodules, and an FP/scan rate of 1.4, outperforming existing models (e.g., nnU-Net, DeepLabV3 + , TransUNet) by up to 21.1% in Dice improvement and 87.3% in FP reduction. The model generalizes robustly across multi-center datasets with &lt; 3% performance variance, while maintaining real-time inference (2.3&#xa0;s/case). These findings indicate that DA<sup>3</sup>-LungNet effectively redefines the sensitivity–specificity trade-off in early lung cancer screening by unifying dynamic loss optimization, dense attention refinement, and contextual 3D reasoning—offering a clinically viable paradigm for precision pulmonary nodule detection.</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

DA3-LUNGNET: a multi-stage deep framework with adaptive attention for early detection of subcentimeter pulmonary nodules

  • 期刊论文

  • 发表日期：

    2025年12月17日

查看全文
  • Bin Zhong,
  • Runan Zhang,
  • Shuai Yu,
  • Chunhua Liu,
  • Huilan Wen

摘要

Early and reliable detection of subcentimeter pulmonary nodules remains a major bottleneck in low-dose CT–based lung cancer screening due to high miss rates, vascular-adhesion–induced false positives, and insufficient multi-scale feature fusion. To address these limitations, we propose DA3-LungNet, a multi-stage deep framework with adaptive attention that integrates dynamic candidate generation, attention-guided fine segmentation, and 3D contextual verification. The first stage introduces dynamic focal RetinaNet with cross-scale feature interaction to maximize recall under extreme class imbalance. The second stage employs an attention-guided U-Net++ augmented with a dense attention bridging module (DABM) for enhanced edge representation and gradient propagation. The final stage integrates a 3D contextual pyramid module (3D-CPM) to model inter-slice spatial continuity and suppress vascular false positives. Extensive experiments on LIDC-IDRI and DSB2017 datasets demonstrate that DA3-LungNet achieves state-of-the-art performance with 92.7% Dice, 93.4% sensitivity for < 6 mm nodules, and an FP/scan rate of 1.4, outperforming existing models (e.g., nnU-Net, DeepLabV3 + , TransUNet) by up to 21.1% in Dice improvement and 87.3% in FP reduction. The model generalizes robustly across multi-center datasets with < 3% performance variance, while maintaining real-time inference (2.3 s/case). These findings indicate that DA3-LungNet effectively redefines the sensitivity–specificity trade-off in early lung cancer screening by unifying dynamic loss optimization, dense attention refinement, and contextual 3D reasoning—offering a clinically viable paradigm for precision pulmonary nodule detection.