GLANCE: continuous global-local exchange with consensus fusion for robust nodule segmentation - Takeaways - MDSpire

GLANCE: continuous global-local exchange with consensus fusion for robust nodule segmentation

  • By

  • Ruijie Ming

  • Fengpin Wang

  • Taotao Zheng

  • Zhongjian Yu

  • Xiaoping Huang

  • Shuangyan Huang

  • Han Tian

  • Wei Wang

  • Jinhai Deng

  • Huawen Liu

  • Yanfang Zheng

  • December 30, 2025

  • 0 min

Share

  • 1

    GLANCE is a novel dual-stream architecture designed for accurate pulmonary nodule segmentation from CT scans.

  • 2

    It integrates a global context transformer and a multi-receptive grouped atrous mixer to capture both long-range dependencies and fine local details.

  • 3

    The continuous cross-scale consensus fusion mechanism enhances feature integration, preventing representational clashes during segmentation.

  • 4

    GLANCE achieves state-of-the-art performance in nodule segmentation and detection across four public benchmarks.

  • 5

    An extensive ablation study confirms the critical role of each architectural component, particularly the continuous fusion strategy.

Original Source(s)

GLANCE: continuous global-local exchange with consensus fusion for robust nodule segmentation

  • by Ruijie Ming, Fengpin Wang, Taotao Zheng, Zhongjian Yu, Xiaoping Huang, Shuangyan Huang, Han Tian, Wei Wang, Jinhai Deng, Huawen Liu, Yanfang Zheng

  • December 30, 2025

Related Content

Hierarchical reasoning for lung cancer detection: from multi-scale perception to hypergraph inference with CR-YOLO

Evidential deep learning-based ALK-expression screening using H&E-stained histopathological images

Machine learning identifies TIME subtypes linking EGFR mutations and immune states in lung adenocarcinoma