• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>World congress on medical physics and biomedical engineering;International congress of the IUPESM >Computerized method for classification between dementia with Lewy bodies and Alzheimer's disease by use of texture analysis on brain MR I
【24h】

Computerized method for classification between dementia with Lewy bodies and Alzheimer's disease by use of texture analysis on brain MR I

机译：利用脑部MR I的纹理分析对路易体痴呆症和阿尔茨海默氏病进行分类的计算机化方法

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Dementia with Lewy bodies (LBD) is the second most frequent dementia after Alzheimer's disease (AD). It is very important to diagnose LBD and AD distinguishably as early as possible, because of adequate treatment. However, it is very difficult to diagnose them. Therefore, the purpose of this study is to develop a computer-aided diagnosis system for classification of LBD and AD on brain MRI. We obtained 71 cases, which are all cases taken in a hospital from 2008: 10 cases with LBD (2 male, 8 female; mean age, 65.2 years), 36 cases with AD (10 male, 26 female; mean age, 66.3 years), and 25 control cases (8 male, 17 female; mean age, 62.2 years). There are no statistical significance with age and gender among groups. For distinguishing AD and LBD, the cerebral parenchyma was automatically extracted from MRI coronal section, and evaluated by use of a texture analysis. We calculated 14 feature values from a co-occurrence matrix and five feature values from a run-length matrix, and selected seven feature values by use of a step-wise feature selection method. A discriminant analysis was applied to distinguish AD and LBD. The performance of our computerized method was evaluated by a round-robin test. 87.3% (62/71) of cases, which were 91.7% (33/36) of AD cases, 70.0% (7/10) of LBD cases, and 88.0% (22/25) of control cases, were correctly detected by our computerized method. We developed a computerized scheme for classification of LBD and AD. Our results indicate that LBD and AD can be distinguished by computer.
机译：路易体痴呆症（LBD）是仅次于阿尔茨海默氏病（AD）的第二常见的痴呆症。由于有足够的治疗方法，因此尽早区分出LBD和AD非常重要。但是，诊断它们非常困难。因此，本研究的目的是开发一种计算机辅助诊断系统，用于在脑MRI上对LBD和AD进行分类。我们获得了71例病例，这些病例都是从2008年开始在医院就诊的：LBD病例10例（男性2例，女性8例;平均年龄65.2岁），AD病例36例（10例男性，26例女性;平均年龄66.3岁） ）和25例对照病例（男8例，女17例;平均年龄62.2岁）。各组之间的年龄和性别均无统计学意义。为了区分AD和LBD，从MRI冠状面自动提取脑实质，并通过纹理分析进行评估。我们从同现矩阵计算了14个特征值，从游程长度矩阵计算了5个特征值，并使用逐步特征选择方法选择了7个特征值。判别分析用于区分AD和LBD。我们的计算机化方法的性能通过循环测试进行了评估。正确检测出87.3％（62/71）的病例，其中AD病例为91.7％（33/36），LBD病例为70.0％（7/10），对照为88.0％（22/25）。我们的计算机化方法。我们开发了一种用于分类LBD和AD的计算机化方案。我们的结果表明，LBD和AD可以通过计算机加以区分。

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