[1]武明虎,马旭茹,刘 敏.基于改进的AlexNet的乳腺肿瘤智能诊断方法[J].湖北工业大学学报,2020,(4):1-5.
　School of Electronics and Electrical Engineering,Hubei Univ. of Tech.,Wuhan 008,et al.Breast Tumor Intelligent Diagnosis Method Based on AlexNet with Batch Normalization[J].,2020,(4):1-5.
点击复制

基于改进的AlexNet的乳腺肿瘤智能诊断方法()
分享到：

《湖北工业大学学报》[ISSN:1003-4684/CN:42-1752/Z]

卷:
期数:
2020年第4期
页码:
1-5
栏目:
湖北工业大学学报
出版日期:
2020-08-31

文章信息/Info

Title:
Breast Tumor Intelligent Diagnosis Method Based on AlexNet with Batch Normalization
文章编号:
1003-4684(2020)04-0001-05
作者:
武明虎;  马旭茹;  刘  敏
湖北工业大学电子与电气工程学院， 湖北 武汉 430068
Author(s):
School of Electronics and Electrical Engineering;  Hubei Univ. of Tech.;  Wuhan 430068;  China
School of Electronics and Electrical Engineering, Hubei Univ. of Tech., Wuhan 430068, China
关键词:
计算机辅助诊断;  乳腺癌病理图像;  AlexNet;  批量归一化
Keywords:
computer-aided diagnosis ;  breast cancer;  AlexNet;  batch normalization
分类号:
TP520.2060
文献标志码:
A
摘要:
由于医学病理图像标注成本高、获取困难，采用数据增强方法对小样本数据集进行扩充。针对传统分类方法对特征工程过度依赖的问题，采用以AlexNet为架构的卷积网络对良性、恶性肿瘤进行全自动分类。为了解决AlexNet网络中间层改变原始数据分布的问题，引入批量归一化算法以减小网络中间层产生的偏移量，加速网络收敛，提高分类准确率。同时采用Dropout避免过拟合。实验结果表明，改进后的AlexNet网络分类准确率较原方法提高了6.09%，具有较好的鲁棒性和泛化性。
Abstract:
Breast cancer has become a globally recognized tumor disease with high incidence and high mortality. It is particularly necessary to use machine learning to assist the diagnosis of breast pathology. In view of the problems of high cost and difficulty in obtaining medical pathological images, data augmentation methods are used to expand small sample data sets. In order to solve the problem of over-reliance on feature engineering by traditional classification methods, a convolutional network based on AlexNet is used to fully classify benign and malignant tumors. In order to solve the problem of changing the original data distribution of the middle layer of the AlexNet, a batch normalization algorithm is introduced to reduce the offset generated by the middle layer of the network, accelerate network convergence, and improve classification accuracy. At the same time, Dropout is used to avoid overfitting. Experimental results show that the improved AlexNet classification accuracy is improved by 6.09% compared with the original method, and it has better robustness and generalization.

参考文献/References:

[1] 中国抗癌协会乳腺癌专业委员会.中国抗癌协会乳腺癌诊治指南与规范(2019版)[J].中国癌症杂志,2019,29(8):609-680.
[2] Almilaji Z, Ersoy I, Hafiane A, et al. Integrating segmentation with deep learning for enhanced classification of epithelial and stromal tissues in H&E images[J]. Pattern Recognition Letters, 2017，119（1）: 214-221.
[3] Kowal M, Filipczuk P, Obuchowicz A, et al. Computer-aided diagnosis of breast cancer based on fine needle biopsy microscopic images[J]. Computers in Biology and Medicine, 2013, 43(10):1563-1572.
[4] George Y, Zayed H H, Roushdy M, et al. Remote computer-aided breast cancer detection and diagnosis system based on cytological images[J]. IEEE Systems Journal, 2014, 8(3): 949-964.
[5] Spanhol F A, Oliveira L S, Petitjean C, et al. A dataset for breast cancer histopathological image classification[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(7): 1455-1462.
[6] Aresta G, Araujo T, Kwok S, et al. BACH: Grand challenge on breast cancer histology images[J]. arXiv: Computer Vision and Pattern Recognition, 2018，56：122-139.
[7] Baris G, Selim A, Ezgi M, et al. Detection and classification of cancer in whole slide breast histopathology images using deep convolutional networks[J]. Pattern Recognition, 2018，84：345-456.
[8] Spanhol F A, Oliveira L S, Petitjean C, et al. Breast cancer histopathological image classification using convolutional neural Networks[C]// International Joint Conference On Neural Network, 2016: 2560-2567.
[9] Spanhol F A, Oliveira L S, Cavalin P R, et al. Deep features for breast cancer histopathological image classification[C]// 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2017:1868-1873.
[10] 郑群花,段慧芳,沈尧,等. 基于卷积神经网络和迁移学习的乳腺癌病理图像分类[J]. 计算机应用与软件, 2018, 35(7):237-242.
[11] Deniz E, Abdulkadir engür, Kadirolu Z, et al. Transfer learning based histopathologic image classification for breast cancer detection[J]. Health Information Science & Systems, 2018, 6(1):1-7.
[12] Faris P D, Ghali W A, Brant R, et al. Multiple imputation versus data enhancement for dealing with missing data in observational health care outcome analyses[J]. Journal of Clinical Epidemiology, 2002, 55(2):184-191.
[13] Ioffe S, Szegedy C. Batch Normalization: Accelerating deep network training by reducing internal covariate shift[C]// International Conference On Machine Learning, 2015: 448-456.

备注/Memo

备注/Memo:
［收稿日期］ 2020-02-01
［基金项目］ 湖北省教育厅科研计划（T201805）
［第一作者］ 武明虎(1975-)， 男，  湖北巴东人，湖北工业大学教授，研究方向为信号与信息处理
［通信作者］ 刘   敏(1970-)，  女，  湖北武汉人，湖北工业大学教授，研究方向为机器视觉
更新日期/Last Update: 2020-08-16