摘要
乳腺癌是女性最多见和死亡率最高的癌症之一,因此乳腺癌的早期筛查与诊断非常有必要,不仅能及时发现隐患,而且可以有效提高患者的存活率。乳腺癌病理图像的特征提取和分类已经成为医学图像处理研究领域的热点,如何准确、高效地检测乳腺癌也成为重要研究内容之一。按照是否需要人工提取乳腺癌病理图像特征,将乳腺癌病理图像特征提取算法分为两大类,分别是基于纹理、形态特征等的传统人工特征提取方法和基于深层神经网络的自动特征提取方法。介绍了几种乳腺癌病理图像相关数据集,然后总结归纳了近几年乳腺癌病理图像特征提取算法的研究进展,并分析了其优缺点,最后,得出乳腺癌病理图像特征提取研究的结论,并对乳腺癌病理图像特征提取的未来发展趋势进行了展望。
Breast cancer is one of the most ordinary cancers in female with the maximumdeath rate.Therefore,early detection and diagnosis of breast cancer is very indispensable,which can not only find latent dangersin good season,but also efficientlyaccelerate the survival rate of patients.Feature extraction and classification of breast cancer pathological images have become a hotspot in the research field of medical image processing,and how to accurately and efficiently detect breast cancer has also become one of the important research contents.According to the need of manual extraction of breast cancer pathological image features,breast cancer pathological image feature extraction algorithm is to be classified in traditional artificial feature extraction based on texture and morphological features and soon or mechanical extraction feature based on neural network.Several breast cancer pathology image data sets were introduced,and then the recent research progress of breast cancer pathology image feature extraction algorithms was summarized,and their merit and demerit were analyzed,finally,the conclusion of the research for breast cancer pathology image feature extraction was draw,and feature extraction for breast cancer pathology images of the future development trend was prospected.
作者
石静文
李嘉
Shi Jingwen;Li Jia(School of Intelligent Manufacturing,Wuyi University,Jiangmen,Guangdong 529020,China)
出处
《机电工程技术》
2022年第4期16-19,共4页
Mechanical & Electrical Engineering Technology
基金
2020年江门市基础与理论科学研究类科技计划项目(编号:2020JC01037)
2021年江门市基础与理论科学研究类科技计划项目(编号:2021030102120004848)。
