• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International journal of image and data fusion >Bone density assessment of oral implant sites using texture parameters
【24h】

Bone density assessment of oral implant sites using texture parameters

机译：使用纹理参数评估口腔种植部位的骨密度

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

The primary objective of this article is to compare the performance of various textural features, computed from dental CT in detecting bone quality (density) at local implant recipient sites and to investigate the correlation between local bone density measured using CT machine and texture features. These findings may provide the clinician with guidelines for dental implant surgical procedures (i.e. to determine whether a one-stage or a two-stage protocol is required). Fifty-five patients were subjected to clinical CT to obtain the CT number and texture-based architectural parameters, respectively. In each site, texture features were extracted using grey level co-occurrence matrices, run-length matrices, histogram and ridgelet-based statistical co-occurrence analysis. The findings are compared with Hounsfield Units measured from the CT machine at local implant sites, which is a standard reference. Then Bayes classifier is used to build the mapping relationships between these features and quality of the bone, respectively. A very difficult problem in classification techniques is the choice of features to distinguish between classes. However, the performance of any classifier is not optimised when all features are used. The feature optimisation problem is addressed using classical sequential methods and floating search algorithms in terms of the best recognition rate and the optimal number of features. Testing this on a series of 90-image sections of trabecular bone with normal, partial and total edentulous patients correctly classified over 90% of the porous bone group with an overall accuracy of 92%. Classical sequential methods and floating search algorithms are compared in terms of the best recognition rate achieved and optimum number of features. The classification performance of 93% is achieved in sequential floating backward search and sequential floating forward search with optimal features compared with sequential methods.
机译：本文的主要目的是比较牙科CT计算的各种纹理特征在检测局部植入物接受者部位的骨质量（密度）方面的性能，并研究使用CT机测量的局部骨密度与纹理特征之间的相关性。这些发现可以为临床医生提供用于牙种植体手术程序的指南（即，确定是否需要一阶段或两阶段方案）。 55例患者接受了临床CT检查，分别获得了CT数和基于纹理的建筑参数。在每个站点中，使用灰度共现矩阵，游程矩阵，直方图和基于山脊的统计共现分析提取纹理特征。将这些发现与通过CT机在局部植入部位测量的Hounsfield单位进行比较，这是标准参考。然后使用贝叶斯分类器分别建立这些特征与骨骼质量之间的映射关系。分类技术中的一个非常困难的问题是选择特征以区分类。但是，使用所有功能时，没有优化任何分类器的性能。在最佳识别率和最佳特征数量方面，使用经典顺序方法和浮动搜索算法解决了特征优化问题。在正常，部分和全部无牙的患者的一系列小梁骨的90幅图像切片上进行测试，正确分类了90％以上的多孔骨组，总体准确率为92％。根据获得的最佳识别率和最佳特征数量对经典顺序方法和浮动搜索算法进行了比较。与顺序方法相比，在具有最佳功能的顺序浮动向后搜索和顺序浮动正向搜索中，可以实现93％的分类性能。

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