用课程学习提升肺气道分割在少量标注数据下的泛化能力
Curriculum Learning for Few-Shot Domain Adaptation in CT-based Airway Tree Segmentation
- 按扫描复杂度分批训练，逐步提升模型难度
- 在两个大型公开数据集上实现高精度分割，少样本微调效果显著
- 适合标注成本高的医学图像分割场景
尽管深度学习取得进展，基于胸部CT的自动化气道分割仍面临分割质量与跨队列泛化能力不足的问题。为此，我们提出将课程学习（Curriculum Learning, CL）引入气道分割网络，依据CT扫描和对应真实分割树特征生成的复杂度评分，将训练集划分为不同批次。重点研究少样本域适应场景，即手动标注完整微调数据集成本过高的情况。在两个大型公开队列（ATM22 和 AIIB23）上的结果表明，使用CL进行全量训练（源域）和少样本微调（目标域）均表现优异，但若采用经典自举评分函数或未合理排序扫描，则可能产生负面效果。
原文摘要 · Abstract (English)
Despite advances with deep learning (DL), automated airway segmentation from chest CT scans continues to face challenges in segmentation quality and generalization across cohorts. To address these, we propose integrating Curriculum Learning (CL) into airway segmentation networks, distributing the training set into batches according to ad-hoc complexity scores derived from CT scans and corresponding ground-truth tree features. We specifically investigate few-shot domain adaptation, targeting scenarios where manual annotation of a full fine-tuning dataset is prohibitively expensive. Results are reported on two large open-cohorts (ATM22 and AIIB23) with high performance using CL for full training (Source domain) and few-shot fine-tuning (Target domain), but with also some insights on potential detrimental effects if using a classic Bootstrapping scoring function or if not using proper scan sequencing.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。