| 姚炜,田建明,赵宝珍,陈宁宁,钱国正.星形算法对心尖四腔切面左心室内膜全自动勾画的初步研究[J].中国医学影像技术,2004,20(8):1284~1288 |
| 星形算法对心尖四腔切面左心室内膜全自动勾画的初步研究 |
| Initial study of automatic left ventricular endocardial boundary detection in apical four chamber view using star algorithm |
| |
| DOI： |
| 中文关键词: 星形算法 全自动勾画 心内膜边界 超声心动检查描记术 |
| 英文关键词:Star algorithm Automatic detection Endocardial boundary Echocardiography |
| 基金项目: |
|
| 摘要点击次数: 1848 |
| 全文下载次数: 872 |
| 中文摘要: |
| 目的 初步测试一种叫做星形算法的全自动超声边界探测方法的实用价值。方法 星形算法采用了从左心室腔内向四周作放射形探测的心内膜探测方法，并且运用了以平均相邻距离和局部平均边界灰度为构件的代价函数来勾画边界。结果 本文用八组带有各种质量问题的舒张末期和收缩末期心尖四腔切面图像对星形算法进行测试，与手工勾画的心内膜相比，星形算法勾画边界的平均半径误差率为12.90％，所围平均面积误差率为10.93％。结论 星形算法能够在临床条件下实现心内膜的全自动勾画。 |
| 英文摘要: |
| Objective To testify the value of a new automatic endocardial boundary detector called star algorithm. Methods Star algorithm has adopted the strategy of radial detection which detects the endocardial boundary from inside the left ventricular chamber. A cost function, using mean neighborhood distance and sectional mean gray level of boundary points, has been applied to determine the boundary points from the candidates. Results Star algorithm was performed on eight pairs of end diastolic and end systolic apical four chamber view images which had various image quality problems. Compared with the manually traced boundary, the mean error radial length rate was 12.90%, and the mean error area rate of star algorithm was 10.93%. Conclusion Star algorithm is able to automatically detect the endocardial boundary under clinical condition. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
