首页> 外国专利> The internal organs organ border selection manner null in vivo which uses the multiple image frames of the internal organs organ

The internal organs organ border selection manner null in vivo which uses the multiple image frames of the internal organs organ

机译:内脏器官边界选择方式在体内无效,它使用了内脏器官的多个图像帧

摘要

PURPOSE: To perform more efficiently and rapidly than others a work with the same level of accuracy as that of an image analysis which an expert for evaluation of the image can perform by fixing the allowable range of the movement of the contour between continuous frames among a series of image frames by using preliminarily obtd. boundary line dynamic data. ;CONSTITUTION: In a block 102, the initial region on the contour of the left ventricle of the heart in the left ventricle contracting image is measured by keeping a distance over one heart period or larger. To provide a shape to the initial estimation, picture elements are classified by the functions of position and density value of each picture element by referring to parameters determined by evaluating manually in advance the learning data by using Bayes classification method. In addition, each picture element of the image frame of a patient is classified whether it is inside or outside of the contour of the left ventricle of the heart and the estimated initial left ventricle region is obtd. by the block 104. In the block 106, the initial left ventricle region estimated in the block 104 is improved by applying dynamic restriction to the change of the contour of the region between a continuous image frames and the estimated left ventricle region is prepd. in the block 108.;COPYRIGHT: (C)1996,JPO
机译:目的:为了以比其他分析更快,更高效的方式执行与图像分析相同精度的工作,评估图像的专家可以通过固定轮廓线在连续帧之间的允许移动范围来执行图像分析系列图像帧通过初步使用。边界线动态数据。 ;组成:在方框102中,通过在一个心律周期或更大的时间内保持距离来测量左心室收缩图像中心脏左心室轮廓上的初始区域。为了给初始估计提供形状,通过参考使用贝叶斯分类方法预先通过手动评估学习数据而确定的参数,通过每个像素的位置和浓度值的功能对像素进行分类。另外,对患者的图像帧的每个图片元素进行分类,无论其位于心脏左心室轮廓的内部还是外部,并且估计的初始左心室区域均被覆盖。通过块104。在块106中,通过对连续图像帧之间的区域的轮廓的变化施加动态限制来改善在块104中估计的初始左心室区域,并且准备估计的左心室区域。在方框108中。;版权:(C)1996,日本特许厅

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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