摘要
早期筛查和及时治疗是控制乳腺癌死亡率最为有效的方法.乳腺X线摄影检查作为医学界公认的最有效的早期乳腺癌筛检工具,可以很好地反映出乳腺存在的异常情况.在临床应用中,乳腺癌的X线摄影直接征象为钙化和肿块,对乳腺X线摄影中钙化点的检测技术已经相当的成熟,但对肿块区域的检测和分类依旧是一项具有挑战性的任务.因此,本文对近几年提出的基于全乳腺X线摄影的肿块检测方法进行简要综述,分别从基于传统的乳腺肿块检测与分割方法和基于深度学习的乳腺肿块检测方法进行介绍,并讨论了乳腺X线摄影中肿块检测未来研究的发展趋势.
Early screening and timely treatment are the most effective ways to control the mortality of breast cancer.Mammography is recognized as the most effective early breast cancer screening tool in the medical field nowadays,and it can well reflect abnormalities in the breast.In clinical application,the direct signs of breast cancer in mammography are calcification and mass.The detection technique of calcification in mammography is quite mature,but the detection and classification of the breast mass is still a challenging task.Therefore,this paper briefly reviews the methods of mass detection based on full mammography in recent years.We introduce the mass detection methods based on traditional mass detection and segmentation methods and deep learning mass detection methods,respectively.Finally,we summarize the proposed mass detection methods,and we also discuss the development trend of future research on mass detection in mammography.
作者
王俊茜
徐勇
孙利雷
蒲祖辉
WANG Jun-Qian;XU Yong;SUN Li-Lei;PU Zu-Hui(Bio-Computing Research Center,College of Computer Sci-ence and Technology,Harbin Institute of Technology,Shenzhen,Shenzhen 518055;Peng Cheng Laboratory,Shenzhe 518000;College of Computer Science and Technology,Guizhou Uni-versity,Guiyang 550025;Guizhou Provincial Key Labora-tory of Public Big Data,Guizhou University,Guiyang 550025;Shenzhen Second People0s Hospital(The First A±liated Hos-pital of Shenzhen University),Shenzhen 518035)
出处
《自动化学报》
EI
CAS
CSCD
北大核心
2021年第4期747-764,共18页
Acta Automatica Sinica
基金
深圳市科技创新委员会(GJHZ20180419190732022)
贵州省公共大数据重点实验室开放课题基金(2018BDKFJJ001)
黔科合重大专项字[2018]3001。