Progressive curriculum learning with Scale-Enhanced U-Net for continuous airway segmentation

作者：Bingyu Yang, Qingyao Tian, Huai Liao, Xinyan Huang, Jinlin Wu, Jianda Hu, Hongbin Liu · 发表于：BMC Medical Imaging · 年份：2025 · DOI：10.1186/s12880-025-02066-5 · 被引用次数：2 · 研究领域：Lung Cancer Diagnosis and Treatment、COVID-19 diagnosis using AI、Advanced Radiotherapy Techniques

Continuous and accurate segmentation of airways in chest CT images is essential for preoperative planning and real-time bronchoscopy navigation. Despite advances in deep learning for medical image segmentation, maintaining airway continuity remains a challenge, particularly due to intra-class imbalance between large and small branches and blurred CT details. To address these challenges, we propose a progressive curriculum learning pipeline and a Scale-Enhanced U-Net (SE-UNet) to improve detail extraction, thereby enhancing segmentation continuity. Compared with previous connectivity-aware methods, our framework directly tackles the imbalance between large and small branches through end-to-end progressive learning, while balancing airway tree completeness and accuracy. Specifically, our curriculum learning pipeline comprises three stages. Stage 1 performs coarse learning to extract main airways. Stage 2 introduces a General Union Loss (GUL) to improve the identification of smaller airways. In Stage 3, we propose an Adaptive Topology-Responsive Loss (ATRL), which encourages the network to focus on preserving airway continuity. Throughout all stages, a crop sampling strategy is employed to reduce feature interference between airways of varying scales, effectively addressing the intra-class imbalance. The progressive training pipeline shares the same SE-UNet, integrating multi-scale inputs and Detail Information Enhancers (DIEs) to enhance information flow and effectively capture...