将医学知识融入持续自监督学习，提升肺部CT图像的模型泛化能力。
Continual Self-supervised Learning Considering Medical Domain Knowledge in Chest CT Images
- 引入增强版DER机制，优化记忆缓冲区的多样性与代表性。
- 在两种成像条件下均优于现有方法，显著减少数据干扰。
- 适合医疗影像持续学习场景，尤其关注肺部CT分析的研究者。
我们提出一种新的持续自监督学习方法（CSSL），用于胸部CT图像并融合医学领域知识。该方法通过有效捕捉不同阶段已学知识与新信息之间的关系，解决序列学习挑战。通过在CSSL中引入增强版动态经验回放（DER），并保持其记忆缓冲区中数据的多样性和代表性，降低了预训练阶段的数据干扰风险，使模型能学习到更丰富、更鲁棒的特征表示。此外，结合Mixup策略与特征蒸馏，进一步提升了模型提取有意义表征的能力。我们在两种不同成像条件下的胸部CT图像上验证了该方法，结果表明其性能优于当前最优方法。
原文摘要 · Abstract (English)
We propose a novel continual self-supervised learning method (CSSL) considering medical domain knowledge in chest CT images. Our approach addresses the challenge of sequential learning by effectively capturing the relationship between previously learned knowledge and new information at different stages. By incorporating an enhanced DER into CSSL and maintaining both diversity and representativeness within the rehearsal buffer of DER, the risk of data interference during pretraining is reduced, enabling the model to learn more richer and robust feature representations. In addition, we incorporate a mixup strategy and feature distillation to further enhance the model's ability to learn meaningful representations. We validate our method using chest CT images obtained under two different imaging conditions, demonstrating superior performance compared to state-of-the-art methods.
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。