南方医科大学学报

南方医科大学学报 ›› 2019, Vol. 39 ›› Issue (01): 88-.doi: 10.12122/j.issn.1673-4254.2019.01.14

• • 上一篇    下一篇

基于深度学习特征的乳腺肿瘤分类模型评估

梁翠霞，李明强，边兆英，吕闻冰，曾栋，马建华   

  • 出版日期:2019-01-20 发布日期:2019-01-20

Establishment of a deep feature-based classification model for distinguishing benign and malignant breast tumors on full-filed digital mammography

  • Online:2019-01-20 Published:2019-01-20

HTML

PDF

2210

摘要： 目的本文结合深度学习特征（DF）和传统图像特征（HCF）特点，利用多分类器融合的方法建立一个乳腺肿瘤分类模型，并 深入评估和分析不同深度学习网络特征的肿瘤分类性能。方法回顾性分析106例乳腺肿瘤患者的头尾位和内外倾斜位投影 的全数字乳腺成像数据。首先从肿瘤区域提取23维HCF（12维形态及11维纹理特征），用t检验进行显著性特征选择；然后分别 从3 个卷积神经网络模型提取不同维度DF，在实验中，3 个不同深度学习网络产生了相应DF，分别是AlexNet，VGG16 和 GoogLeNet；最后结合2个投影数据的DF和HCF，采用多分类器的融合模型对特征进行训练和测试，实验重点分析不同DF在 肿瘤分类上的性能。结果结合DF和HCF建立的分类模型比使用单独HCF的分类模型表现出更好的性能；相比于其它网络框 架，DFAlexNet和HCF的结合表现出更高精度的分类结果。结论结合DF和HCF的特征方法建立一个分类模型，对于良恶性乳腺 肿瘤具有优秀的鉴别能力，且泛化能力更强，能作为临床辅助诊断工具。

Abstract: Objective To develop a deep features-based model to classify benign and malignant breast lesions on full- filed digital mammography. Method The data of full-filed digital mammography in both craniocaudal view and mediolateral oblique view from 106 patients with breast neoplasms were analyzed. Twenty-three handcrafted features (HCF) were extracted from the images of the breast tumors and a suitable feature set of HCF was selected using t-test. The deep features (DF) were extracted from the 3 pre-trained deep learning models, namely AlexNet, VGG16 and GoogLeNet. With abundant breast tumor information from the craniocaudal view and mediolateral oblique view, we combined the two extracted features (DF and HCF) as the two-view features. A multi-classifier model was finally constructed based on the combined HCF and DF sets. The classification ability of different deep learning networks was evaluated. Results Quantitative evaluation results showed that the proposed HCF+DF model outperformed HCF model, and AlexNet produced the best performances among the 3 deep learning models. Conclusion The proposed model that combines DF and HCF sets of breast tumors can effectively distinguish benign and malignant breast lesions on full-filed digital mammography.