Tang Xiaofeng, Zhang Haoyan, Mao Rushuang, Zhang Yafang, Jiang Xinhua, Lin Min, Xiong Lang, Chen Haolin, Li Li, Wang Kun, Zhou Jianhua
Department of Ultrasound, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, Guangdong, China.
CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, Beijing, China; School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China.
Acad Radiol. 2025 Jan;32(1):1-11. doi: 10.1016/j.acra.2024.07.029. Epub 2024 Aug 5.
Deep learning can enhance the performance of multimodal image analysis, which is known for its noninvasive attributes and complementary efficacy, in predicting axillary lymph node (ALN) metastasis. Therefore, we established a multimodal deep learning model incorporating ultrasound (US) and magnetic resonance imaging (MRI) images to predict ALN metastasis in patients with breast cancer.
A retrospective cohort of patients with histologically confirmed breast cancer from two hospitals composed of the primary cohort (n = 465) and the external validation cohort (n = 123). All patients had undergone both preoperative US and MRI scans. After data preprocessing, three convolutional neural network models were used to analyze the US and MRI images, respectively. After integrating the US and MRI deep learning prediction results (DL and DL, respectively), a multimodal deep learning (DL+Clinical parameter) model was constructed. The predictive ability of the proposed model was compared to that of the DL, DL, combined bimodal (DL), and clinical parameter models. Evaluation was performed using the area under the receiver operating characteristic curves (AUCs) and decision curves.
A total of 588 patients with breast cancer participated in this study. The DL+Clinical parameter model outperformed the alternative models, achieving the highest AUCs of 0.819 (95% confidence interval [CI] 0.734-0.903) and 0.809 (95% CI 0.723-0.895) on the internal and external validation sets, respectively. The decision curve analysis confirmed its clinical usefulness.
The DL＋Clinical parameter model demonstrates the feasibility and reliability of its performance for ALN metastasis prediction in patients with breast cancer.
深度学习可以提高多模态图像分析在预测腋窝淋巴结（ALN）转移方面的性能，多模态图像分析以其非侵入性属性和互补功效而闻名。因此，我们建立了一个结合超声（US）和磁共振成像（MRI）图像的多模态深度学习模型，以预测乳腺癌患者的ALN转移。
一项回顾性队列研究，研究对象为来自两家医院的组织学确诊乳腺癌患者，包括初级队列（n = 465）和外部验证队列（n = 123）。所有患者均接受了术前超声和MRI扫描。经过数据预处理后，分别使用三个卷积神经网络模型分析超声和MRI图像。在整合超声和MRI深度学习预测结果（分别为DL和DL）后，构建了一个多模态深度学习（DL+临床参数）模型。将所提出模型的预测能力与DL、DL、联合双峰（DL）和临床参数模型的预测能力进行比较。使用受试者操作特征曲线下面积（AUC）和决策曲线进行评估。
共有588例乳腺癌患者参与本研究。DL+临床参数模型优于其他模型，在内部和外部验证集上分别实现了最高AUC，分别为0.819（95%置信区间[CI]0.734-0.903）和0.809（95%CI 0.723-0.895）。决策曲线分析证实了其临床实用性。
DL＋临床参数模型证明了其在预测乳腺癌患者ALN转移方面性能的可行性和可靠性。
