...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Journal of Innovative Computing Information and Control >ULTRASONIC LIVER TISSUE CHARACTERIZATION BY MULTIRESOLUTION FEATURE VECTOR AND AN ENSEMBLE OF CLASSIFIERS
【24h】

ULTRASONIC LIVER TISSUE CHARACTERIZATION BY MULTIRESOLUTION FEATURE VECTOR AND AN ENSEMBLE OF CLASSIFIERS

机译：多分辨率特征向量的超声肝组织表征及分类器

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This study describes the feasibility of characterizing liver tissue through mul-tiresolution feature vector and combined classifiers in the interpretation of ultrasonic imagery. This approach comprised three main stages. The first stage utilized multiresolution analysis to extract intrinsic features from ultrasonic liver images. As a result of spatial-frequency decomposition, feature representation for each subimage was obtained. The second stage tested various classification algorithms based on respective texture measurements and filter banks. The final stage involved the aggregation of individual classifiers based on fuzzy integral. The proposed approach was applied to discriminate among ultrasonic images of liver tissue in three states: normal liver, liver with cirrhosis and liver with hepatoma. The classification results showed that the approach yielded performance superior to that obtained through the use of individual classifiers. Furthermore, clinicians could use the quantitative index of the classification results to decide whether to proceed with an advanced medical examination, thereby improving the quality of medical care.
机译：这项研究描述了通过多分辨率特征向量和组合分类器表征肝脏组织在超声图像解释中的可行性。该方法包括三个主要阶段。第一阶段利用多分辨率分析从超声肝图像中提取固有特征。作为空间频率分解的结果，获得了每个子图像的特征表示。第二阶段基于各自的纹理测量和滤波器组测试了各种分类算法。最后阶段涉及基于模糊积分的各个分类器的汇总。所提出的方法用于区分三种状态的肝组织超声图像：正常肝，肝硬化肝和肝癌肝。分类结果表明，该方法产生的性能优于通过使用单个分类器获得的性能。此外，临床医生可以使用分类结果的定量指标来决定是否进行高级医学检查，从而提高医疗质量。

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