...
首页> 外文期刊>Multimedia Tools and Applications >Automatic computer-aided caries detection from dental x-ray images using intelligent level set
【24h】

Automatic computer-aided caries detection from dental x-ray images using intelligent level set

机译:使用智能水准仪从牙齿X射线图像自动检测计算机辅助龋齿

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

Dental diseases have high risk of affection across the globe and mostly in adult population. The analysis of dental X-ray images has some difficulties in comparison to other medical images, which makes segmentation a more challenging process. One of the most important and yet largely unsolved issues in the level set method framework is the efficiency of signed force, speed function and initial contour (IC) generation. In this paper, a new segmentation method based on level set (LS) is proposed in two phases; IC generation using morphological information of image and intelligent level set segmentation utilizing motion filtering and back propagation neural network. The segmentation results are efficient and accurate as compared to other studies. The new approach to isolate each segmented teeth image is proposed by employing integral projection technique and feature map designed for each tooth to extract the local information and therefore to detect caries area. The achieved overall performance of the proposed segmentation method was evaluated at 120 periapical dental radiograph (X-ray), with images at 90% and the detection accuracy of 98%.
机译:牙科疾病在全球范围内以及大多数成年人口中都有很高的患病风险。与其他医学图像相比,牙科X射线图像的分析存在一些困难,这使分割过程更具挑战性。水平集方法框架中最重要但尚未解决的问题之一是有符号力,速度函数和初始轮廓(IC)生成的效率。本文分两个阶段提出了一种基于水平集的最小分割方法。利用图像形态信息生成IC,并利用运动滤波和反向传播神经网络进行智能水平集分割。与其他研究相比,分割结果是有效且准确的。通过采用整体投影技术和针对每个牙齿设计的特征图,提出了一种分离每个分割牙齿图像的新方法,以提取局部信息,从而检测出龋齿区域。在120根根尖牙X光片(X射线)上评估了所提出的分割方法的总体性能,图像质量为90%,检测精度为98%。

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