跳到主要导航 跳到搜索 跳到主要内容

Genap: Generalizing Across the Augmentation Gap in Medical Image Segmentation Using Single-Source Domain

  • Jianyu Chen
  • , Haojin Li
  • , Zenan Chen
  • , Heng Li*
  • , Yijie Pan
  • , Xun Zhang
  • , Jiang Liu
  • *此作品的通讯作者
  • Southern University of Science and Technology
  • Shenzhen University of Advanced Technology
  • Eastern Institute of Technology, Ningbo
  • Tsinghua University
  • Ltd.

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

摘要

Due to variations in imaging devices and pathological differences, traditional deep learning segmentation models often encounter domain shifts between training and clinical test data, leading to significant performance degradation. Given the high cost of acquiring multi-source target domain data in medical scenarios, Single-Source Domain Generalization (SDG) has emerged as a promising approach, enabling models trained on a single-source dataset to generalize to unseen target domains. However, existing SDG methods based on augmentation techniques suffer from the augmentation gap, where augmented data fail to fully encompass the variations present in clinical test scenarios, potentially resulting in model failure. To address this challenge, we propose a novel SDG algorithm that leverages frequency spectrum analysis and band-specific perturbations to bridge the augmentation gap and enhance domain generalization. Specifically, we identify augmentation regions in the frequency spectrum where conventional augmentations introduce minimal perturbations and apply perturbations to enhance data diversity. Considering model stability, we incorporate frequency band attention mechanisms and an adversarial training framework, ensuring semantic consistency during the augmentation and feature extraction processes. Extensive experiments on medical image datasets demonstrate that our method significantly improves segmentation performance across unseen domains, outperforming state-of-the-art SDG approaches. These findings highlight the potential of our method for real-world clinical deployment, where model robustness to domain shifts is critical.

源语言英语
主期刊名Advanced Intelligent Computing Technology and Applications - 21st International Conference, ICIC 2025, Proceedings
编辑De-Shuang Huang, Qinhu Zhang, Chuanlei Zhang, Wei Chen
出版商Springer Science and Business Media Deutschland GmbH
页282-293
页数12
ISBN(印刷版)9789819500352
DOI
出版状态已出版 - 2025
已对外发布是
活动21st International Conference on Intelligent Computing, ICIC 2025 - Ningbo, 中国
期限: 26 7月 2025 → 29 7月 2025

丛书

姓名Lecture Notes in Computer Science
卷15869 LNBI
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议21st International Conference on Intelligent Computing, ICIC 2025
国家/地区中国
市Ningbo
时期26/07/25 → 29/07/25

学术指纹

探究 'Genap: Generalizing Across the Augmentation Gap in Medical Image Segmentation Using Single-Source Domain' 的科研主题。它们共同构成独一无二的学术指纹。

引用此