多模态影像组学与深度学习在乳腺癌腋窝淋巴结状态预测中的概述
综述总结了影像组学与深度学习在多模态影像(乳腺X线、超声、MRI、PET/CT)中无创预测乳腺癌腋窝淋巴结状态的研究进展。文章系统分析了影像组学工作流程与模型构建策略,指出该领域已取得显著进展但仍存在方法学与技术挑战,旨在为精准诊断评估提供参考。
要点
- 多模态影像组学与深度学习在无创预测乳腺癌腋窝淋巴结状态方面显示出可观潜力。
- 综述系统梳理了影像组学工作流程与模型构建策略。
- 该领域仍存在若干方法学与技术挑战有待解决。
核心结论
- 影像组学与深度学习在多模态影像中预测乳腺癌腋窝淋巴结状态已取得显著进展,但仍有若干方法学与技术挑战有待解决。
研究设计与样本
- 综述纳入影像组学与深度学习在多模态(乳腺X线、超声、MRI、PET/CT)预测乳腺癌腋窝淋巴结状态的研究,并详细分析影像组学工作流程与模型构建策略。
局限性
- 原文指出影像组学预测乳腺癌腋窝淋巴结转移虽已取得显著进展,但仍有若干方法学与技术挑战有待解决。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
Breast cancer is the most prevalent malignancy in women, with the status of axillary lymph nodes being a pivotal factor in treatment decision-making and prognostic evaluation. With the integration of deep learning algorithms, radiomics has become a transformative tool with increasingly extensive applications across multimodality, particularly in oncological imaging. Recent studies of radiomics and deep learning have demonstrated considerable potential for noninvasive diagnosis and prediction in breast cancer through multimodalities (mammography, ultrasonography, MRI and PET/CT), specifically for predicting axillary lymph node status. Although significant progress has been achieved in radiomics-based prediction of axillary lymph node metastasis in breast cancer, several methodological and technical challenges remain to be addressed. The comprehensive review incorporates a detailed analysis of radiomics workflow and model construction strategies. The objective of this review is to synthesize and evaluate current research findings, thereby providing valuable references for precision diagnosis and assessment of axillary lymph node metastasis in breast cancer, while promoting development and advancement in this evolving field.
引用
- Zhao X, Wang M, Wei Y, et al. Overview of Multimodal Radiomics and Deep Learning in the Prediction of Axillary Lymph Node Status in Breast Cancer. Acad Radiol. 2025. PMID: 40830005. doi:10.1016/j.acra.2025.07.017
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。