期刊文献+

基于频域对称法的磁共振数据残余噪声消除方法 被引量:2

Residual noise cancellation of MRS signal based on frequency-domain symmetry
下载PDF
导出
摘要 根据磁共振(MRS)信号在频域呈现的特点,提出一种频域对称法,用作传统消噪流程的补充步骤以消除残余噪声。首先,对含噪信号进行频域解调,分解为实时域分量和虚时域分量;然后,利用2个分量的频谱在工频噪声频率处相同、负频率处相反的特点,估算与MRS探测信号混叠的工频噪声实部分量,与虚时域分量同步剔除,达到信噪分离的目的。研究结果表明:采用频域对称法消除残余噪声后,数据信噪比由15.29 dB提升至19.33 d B,初始振幅提取相对误差由2.03%减小为0.86%,弛豫时间提取相对误差由5.53%减小为2.27%。相较于传统MRS消噪方法,采用频域对称法可以进一步消除残余噪声,提高信号信噪比。 Based on the characteristics of magnetic resonance sounding(MRS) signal presented in frequency domain, a frequency-domain symmetry method used as a supplementary step of the traditional MRS-processing workflow was proposed to remove residual noise. Firstly, the noisy signal was demodulated in the frequency domain and decomposed into real and imaginary components in the time domain. Then, the real component of the powerline harmonic noise which was aliased with MRS signal in the time domain was estimated by using the distinctions of the frequency spectrum of the two components to be the same at the harmonic frequency and the opposite at the negative harmonic frequency. The real component of the powerline harmonic noise was discarded together with the imaginary time domain component to separate noise and signal. The results show that after eliminating residual noise using the frequency-domain symmetry method, the data signal-to-noise ratio increases from 15.29 dB to 19.33 dB, the initial amplitude extraction error decreases from 2.03% to 0.86%, and the relaxation time extraction error decreases from 5.53% to 2.27%. Compared with the traditional MRS-processing workflow,the frequency-domain symmetry method can further eliminate the residual noise and improve the signal SNR.
作者 林婷婷 李玥 刘大震 万玲 LIN Tingting;LI Yue;LIU Dazhen;WAN Ling(College of Instrumentation&Electrical Engineering,Jilin University,Changchun 130026,China;Key Laboratory of Geophysical Exploration Equipment,Ministry of Education,Changchun 130026,China)
出处 《中南大学学报(自然科学版)》 EI CAS CSCD 北大核心 2021年第10期3494-3504,共11页 Journal of Central South University:Science and Technology
基金 国家优秀青年基金资助项目(41722405) 国家自然科学基金资助项目(41974208) 吉林省科技厅重点科技研发项目(20180201017GX) 吉林省科技发展计划项目(20180520183JH)。
关键词 磁共振 频域对称法 残余噪声 数据处理 magnetic resonance sounding frequency-domain symmetry method residual noise data processing
  • 相关文献

参考文献3

二级参考文献17

共引文献22

同被引文献32

引证文献2

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部