基于多源异构数据的淋巴结转移状态分类

Prediction of Lymph Node Metastasis Status Based on Multi-Source Heterogeneous Data

提出一种基于多源异构数据的深度学习分类模型,对早期乳腺癌患者腋窝淋巴结转移状态进行评估。模型分别提取传统影像组学特征、B型超声图像及剪切波弹性超声图像数据的共性特征,构建特征空间并进行分类预测。该模型对淋巴结转移状态的预测准确率达到86.2%,实验结果表明,多源异构数据融合能有效提高腋窝淋巴结转移状态的诊断效能。

A deep learning classification model based on multi-source heterogeneous data is proposed to evaluate the status of axillary lymph node metastasis in patients with early-stage breast cancer.The model extracts the features of traditional radiomics features,Bmode ultrasound images and shear wave elastic ultrasound image data,constructs a feature space and performs classification predic⁃tion.The classification accuracy rate of this model for lymph node metastasis status reached 86.2%.Experimental results show that multi-source heterogeneous data fusion can effectively improve the diagnostic performance of axillary lymph node metastasis status.