| 基于 Inception-ResNet-V2 的乳腺癌病理图像识别研究 |
| |
| 引用本文： | 刘靖雯,黄理灿.基于 Inception-ResNet-V2 的乳腺癌病理图像识别研究[J].教育技术导刊,2020,19(5):225-229. |
| |
| 作者姓名： | 刘靖雯 黄理灿 |
| |
| 作者单位： | 浙江理工大学 信息学院，浙江 杭州 310018 |
| |
| 基金项目： | 深圳市教育科学规划项目（ybzz18005，zdzz18008，ybfz18030） |
| |
| 摘 要： | 乳腺癌严重威胁女性健康和生命，及时诊断并提供治疗方案给医生带来了挑战，病理图像分类结果是医生确诊的重要依据，实现乳腺癌病理图像识别分类具有重要意义及临床应用价值。近年来，大多数研究集中于良恶性分类，而不同类型的乳腺肿瘤本身具有不同病因及治疗方法。采用 Inception-ResNet-V2 深度卷积神经网络模型，实现对乳腺癌病理图像的八分类，利用数据增强和迁移学习方法，在 Matlab 上对数据集 BreaKHis进行实验。结果表明，该方法识别率基本达到 80%以上，比大部分已有研究成果效果更优。
|
| 关 键 词： | Inception-ResNet-V2 深度卷积神经网络 数据增强 迁移学习 乳腺癌病理图像 |
| 收稿时间： | 2019-07-30 |
Pathological Image Recognition of Breast Cancer Based on Inception-ResNet-V2 |
| |
| LIU Jing-wen,HUANG Li-can.Pathological Image Recognition of Breast Cancer Based on Inception-ResNet-V2[J].Introduction of Educational Technology,2020,19(5):225-229. |
| |
| Authors: | LIU Jing-wen HUANG Li-can |
| |
| Institution: | School of Informatics Science and Technology，Zhejiang Sci-tech University，Hangzhou 310018，China |
| |
| Abstract: | Breast cancer is a serious threat to women’s health and lives，and it is a challenge for doctors to make timely diagnosis and treatment options. At the same time，because the classification results of pathological images become an important basis for the diagnosis of doctors，it is of great significance and clinical application value to realize the classification and classification of breast cancer pathological images. In recent years，most of the research has focused on benign and malignant classification，but different types of breast tumors should have different etiology and treatment methods. This paper uses the Inception-ResNet-V2 deep convolutional neural network model to achieve eight classifications of breast cancer pathological images. Using data augmentation and transfer learning methods，experiments were performed on the dataset Breakhis on Matlab，and the results show that the recognition rate of this method
basically reaches over 80%，and the effect is better than most previous research results. |
| |
| Keywords: | Inception-ResNet-V2 deep convolutional neural network data enhancement transfer learning breast cancer pathology image |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
