联合多视角Transformer编码与在线融合互学习的乳腺癌病理图像分类模型  

Breast Cancer Pathological Image Classification Model via CombiningMulti-View Transformer Coding and Online Fusion Mutual Learning

作　　者：李广丽[1] 叶艺源 吴光庭 李传秀 吕敬钦 张红斌[2] LI Guang-li;YE Yi-yuan;WU Guang-ting;LI Chuan-xiu;LÜJing-qin;ZHANG Hong-bin(School of Information Engineering,East China Jiaotong University,Nanchang,Jiangxi 330013,China;School of Software,East China Jiaotong University,Nanchang,Jiangxi 330013,China)

机构地区：[1]华东交通大学信息工程学院,江西南昌330013 [2]华东交通大学软件学院,江西南昌330013

出　　处：《电子学报》2024年第7期2369-2381,共13页Acta Electronica Sinica

基　　金：国家自然科学基金(No.62161011);江西省重点研发计划重点项目(揭榜挂帅)(No.20223BBE51036);江西省自然科学基金(No.20212BAB202006);江西省社科规划项目(No.22TQ01);江西省教育厅科技项目(No.GJJ200628,No.GJJ2200639);江西省研究生创新基金(No.YC2022-s546)~~。

摘　　要：乳腺癌是女性最常见的癌症.单一网络在乳腺癌病理图像分类中存在缺陷,卷积神经网络无法提取全局上下文,而Transformer不能准确描述局部细节.本文提出联合多视角Transformer编码与在线融合互学习的乳腺癌病理图像分类模型(Multi-View Transformer Online Fusion Mutual Learning,MVT-OFML).采用ResNet-50(Residual Network-50)提取图像局部特征,设计多视角Transformer编码模块,捕获图像中全局上下文;联合Logits和中间特征层构建OFML框架,实现ResNet-50与多视角Transformer编码模块间双向传递知识,使2个网络优势互补以完成乳腺癌病理图像分类.实验表明,在BreakHis和BACH数据集上,MVT-OFML的准确率比最强基线分别提升0.90%和2.26%,F1均值比最强基线分别提升4.75%和3.21%.Breast cancer is the most common cancer in women.The single neural network used in breast cancer path⁃ological image classification has the following defects:the convolutional neural network(CNN)lacks the ability to extract global context information while the Transformer lacks the ability to depict local lesion details.To alleviate the problem,a novel model,named multi-view Transformer coding and online fusion mutual learning(MVT-OFML),is proposed for breast cancer pathological image classification.First,ResNet-50 is employed to extract local features in images.Then,a new multi-view Transformer(MVT)coding module is designed to capture the global context information.Finally,a novel online fusion mutual learning(OFML)framework based on the Logits and middle feature layers is designed to implement the bi-directional knowledge transfer between ResNet-50 and the MVT coding module.This makes the two networks com⁃plement each other to complete breast cancer pathological image classification.Experiments validated on BreakHis and BACH show that compared to the best baseline,the performance improvements of accuracy are 0.90%and 2.26%,respec⁃tively,whereas the corresponding improvements of average F1 score are 4.75%and 3.21%,respectively.

关 键 词：乳腺癌 病理图像分类 多视角Transformer 卷积神经网络 在线融合互学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]