首页 | 本学科首页   官方微博 | 高级检索  
     检索      

超声断层扫描数据压缩的时间窗构造及数据提取方法
引用本文:秦博,马晓川,Michael Zapf,鄢社锋,Nicole V.Ruiter.超声断层扫描数据压缩的时间窗构造及数据提取方法[J].声学学报,2018,43(2):189-201.
作者姓名:秦博  马晓川  Michael Zapf  鄢社锋  Nicole V.Ruiter
作者单位:1. 中国电子科技集团公司电子科学研究院 北京 100041;
摘    要:介绍了一种针对三维超声断层扫描系统的数据压缩方法,该方法通过检测最早的目标反射重构目标表面位置;依据表面位置,计算目标回波信号的起止时间;以此为依据构造时间窗,标明成像算法所需数据在采样序列中存在的时间范围;最终提取有用数据,完成数据压缩。临床数据被用于算法的效果评价。结果表明,该方法可达到平均2.27的压缩率和0.21的压缩率标准差,以及2.25的数据传输加速比。较传统数据压缩方法,该方法只利用回波时域信息,可作为预压缩方法和传统方法结合使用以获得更高的压缩率。 

关 键 词:数据压缩    数据缩减    时域压缩    超声信号数据压缩    三维超声断层扫描技术
收稿时间:2017-06-05

The time-window construction and data extraction for the data reduction of ultrasound computer tomography
Institution:1. Academy of Electronics and Information Technology, China Electronics Technology Group Corporation Beijing 100041;2. Institute of Acoustics, Chinese Academy of Sciences Beijing 100190;3. Institute for Data Processing and Electronics, Karlsruhe Institute of Technology Karlsruhe 76021;4. University of Chinese Academy of Sciences Beijing 100049
Abstract:A data reduction method is presented for three-dimensional ultrasound computer tomography. This method identifies the earliest surface reflection of the breast, reconstructs the breast surface, calculates the begin and end time of the echo signals from the breast, builds time windows to mark the range of the valid data for imaging algorithms in the sequences of samples and extracts useful data to complete the data reduction. 15 breasts with different shapes and mislocalizations are used to evaluate the limitations of the technique. Based on clinical data, the presented method achieves a mean compression ratio of 2.27, a standard deviation of 0.21 and a speedup factor of 2.25 for the data transmission. This method is only based on the time-domain information of echoes; therefore, it can be applied as a pre-compression method and combined with conventional methods for a higher compression ratio. 
Keywords:
本文献已被 CNKI 等数据库收录!
点击此处可从《声学学报》浏览原始摘要信息
点击此处可从《声学学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号