arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.00744cs.CVcs.AI
收藏

原型记忆引导的无训练产前超声异常分类与定位

Prototype Memory-Guided Training-Free Anomaly Classification and Localization in Prenatal Ultrasound

Medical Ultrasound Image Computing (MUSIC) Lab, Shenzhen University(深圳大学医学超声图像计算实验室);School of Biomedical Engineering, Medical School, Shenzhen University(深圳大学医学部生物医学工程学院);School of Artificial Intelligence, Shenzhen University(深圳大学人工智能学院);School of Biomedical Engineering and Informatics, Nanjing Medical University(南京医科大学生物医学工程与信息学院);Centre for Artificial Intelligence and Robotics, Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences(中国科学院香港创新研究院人工智能与机器人创新中心);Shenzhen Luohu People’s Hospital (The Third Affiliated Hospital of Shenzhen University)(深圳市罗湖区人民医院(深圳大学第三附属医院));Center for Medical Ultrasound, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou Municipal Hospital, Nanjing Medical University(南京医科大学附属苏州医院(苏州市立医院)医学超声中心);Northwest Women’s and Children’s Hospital(西北妇女儿童医院);Qilu Hospital of Shandong University(山东大学齐鲁医院)

Huanwen Liang, Yuhao Huang, Xiliang Zhu, Yuanji Zhang, Xuedong Deng, Xinru Gao, Guowei Tao, Yuhan Zhang, Dong Ni

AI总结:

提出一种无需训练的多类产前超声异常分类与定位框架,利用多粒度原型记忆库、原型驱动软合并和类别感知细化策略,在少量参考图像下实现高效检测。

摘要:

产前异常分类与定位对胎儿健康和妊娠管理至关重要。尽管超声(US)是产前筛查的主要方式,但由于异常的低发生率和高度异质性,准确诊断仍具挑战。现有的产前任务深度学习方法依赖于大规模标注数据集,而实践中难以获取。虽然少样本学习缓解了数据稀缺问题,但通常需要针对新类别进行微调,限制了其在资源有限的临床环境中的实用性。为应对这些挑战,我们提出了一种无训练框架,用于多类产前US异常分类与定位,仅需每类少量参考图像即可运行,这是对该设置的首个探索。我们的框架包含三个关键组件:(1)一个具有多粒度原型的记忆库,显式建模类别级语义和异常特征;(2)一种原型驱动的软合并机制,聚合判别性特征以检测异常区域;(3)一种类别感知细化策略,利用原型一致性改进类别预测。在一个包含1,149例病例、共2,357张图像和9个类别的多中心产前US数据集上进行了广泛验证,所提方法优于竞争对手。

英文摘要:

Prenatal anomaly classification and localization is of critical importance for fetal health and pregnancy management. Although ultrasound (US) is the primary modality for prenatal screening, accurate diagnosis remains challenging due to the low prevalence and high heterogeneity of anomalies. Existing deep learning methods for prenatal tasks rely on large-scale annotated datasets, which are difficult to obtain in practice. Although few-shot learning alleviates data scarcity, it typically requires fine-tuning for new categories, limiting its practicality in resource-limited clinical settings. To address these challenges, we propose a training-free framework for multi-class prenatal US anomaly classification and localization that operates with only a few reference images per class, representing the first exploration of this setting. Our framework comprises three key components: (1) a memory bank with multi-granular prototypes that explicitly models both class-level semantics and anomaly characteristics; (2) a prototype-driven soft merging mechanism that aggregates discriminative features to detect the anomaly region; and (3) a class-aware refinement strategy that leverages prototype consistency to improve category prediction. Extensively validated on a multi-center prenatal US dataset containing 1,149 cases, with a total of 2,357 images and 9 categories, our proposed method outperforms the competitors.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑