首页> 外文会议>International Conference on Computer Communication and Informatics >Extreme learning machine based classification for detecting micro-calcification in mammogram using multi scale features
【24h】

Extreme learning machine based classification for detecting micro-calcification in mammogram using multi scale features

机译:基于极端学习机的分类,用于使用多尺度特征检测乳房X线图中的微钙化

获取原文
获取外文期刊封面目录资料

摘要

In the human body, there are some genes that are lead to the growth of the cells. The mutation of these genes are called cancer. Breast cancer is higher in women, and which will causes largest number of cancer related deaths among women. Breast cancer rates are higher among women in many countries. To increase the results of breast cancer and survival, early diagnosis is crucial. There are two early screening plans for breast cancer: early detection and screening. Limited resource parameters with low health systems where most women are diagnosed in the late stages and should organize early diagnosis programs based on knowledge of the first signs and symptoms. Many methods are used to test women to identify cancer before all symptoms appear. Mammography is one of the methods in which an image of the breast used to detect and diagnose breast cancer tumors. Micro-calcification can be found in mammogram and it will indicate the presence of breast cancer. Preprocessing, feature extraction and classification are the three important steps to detect the micro calcification in mammogram. And there are different classifiers used for the classification of micro calcification. In this paper we analyze the performance of different classifiers and find out the best one for the classification using multi scale features.
机译:在人体中,有一些基因是导致细胞的生长。这些基因的突变被称为癌。乳腺癌是女性高,而这将导致妇女癌症相关死亡的人数最多。乳腺癌的发病率在许多国家妇女高。为了增加乳腺癌和生存的结果,早期诊断是至关重要的。有乳腺癌两种癌症的早期筛查计划:及早发现和筛查。与大多数妇女被诊断在后期阶段,并应组织基础上的第一个迹象和症状的知识早期诊断方案低卫生系统有限的资源参数。可采用多种方法来测试所有的女性在症状出现之前识别癌症。乳房X射线摄影是,其中乳房的图像用于检测和诊断乳腺癌肿瘤的方法之一。微钙化可在乳房X光检查发现,这将预示乳腺癌的存在。预处理,特征提取和分类可用于检测在乳房X线照片的微钙化的三个重要步骤。并有用于微钙化的分类不同的分类。在本文中,我们分析了不同分类器的性能,找出使用多尺度特征的分类最好的之一。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号