期刊文献+
共找到4篇文章
< 1 >
每页显示 20 50 100
基于3D U-Net和DTI模型约束的扩散张量场估计方法 被引量:1
1
作者 麦兆华 李嘉龙 +1 位作者 冯衍秋 张鑫媛 《南方医科大学学报》 CAS CSCD 北大核心 2023年第7期1224-1232,共9页
目的为从少量的低信噪比扩散加权(DW)图像中估计得到准确的扩散张量成像(DTI)量化参数,本文提出一种基于3D U-Net和DTI模型约束的扩散张量场估计网络(3D DTI-Unet)。方法3D DTI-Unet的输入为有噪声的扩散磁共振成像(dMRI)数据(包含1幅... 目的为从少量的低信噪比扩散加权(DW)图像中估计得到准确的扩散张量成像(DTI)量化参数,本文提出一种基于3D U-Net和DTI模型约束的扩散张量场估计网络(3D DTI-Unet)。方法3D DTI-Unet的输入为有噪声的扩散磁共振成像(dMRI)数据(包含1幅非扩散加权图像与6幅不同扩散编码方向的DW图像),通过3D U-Net网络预测得到降噪后的非扩散加权图像以及准确的扩散张量场,并通过DTI模型重建得到dMRI数据,将其与dMRI数据的真实值进行比较来优化网络,从而保证dMRI数据与扩散张量场的物理模型一致性。为验证所提方法的有效性,与Marchenko-Pastur主成分分析(MP-PCA)和基于全局指导下的局部高阶奇异值分解(GL-HOSVD)这两种扩散加权图像去噪算法进行实验对比。结果从DW图像、扩散张量场以及DTI量化参数的定量分析结果以及视觉效果来看,所提方法均优于MP-PCA与GL-HOSVD。结论本文所提方法能够从1幅非扩散加权图像和6幅DW图像得到准确的DTI量化参数,可减少临床采集时间,提高临床量化诊断的可靠性。 展开更多
关键词 扩散张量成像 张量估计 3D U-Net Rician噪声 图像去噪
下载PDF
DTI扩散张量的一种稳健估计方法 被引量:5
2
作者 白衡 高玉蕊 +1 位作者 王世杰 罗立民 《计算机研究与发展》 EI CSCD 北大核心 2008年第7期1232-1238,共7页
为了获得更精确的DTI扩散张量场,提出了一种基于约束M估计子的稳健估计方法.首先对扩散加权图像序列进行双树复数小波降噪预处理,以减少热噪声影响.然后通过试探法找到一个合适的回归起始点,并通过Cholesky分解对扩散张量进行正定约束.... 为了获得更精确的DTI扩散张量场,提出了一种基于约束M估计子的稳健估计方法.首先对扩散加权图像序列进行双树复数小波降噪预处理,以减少热噪声影响.然后通过试探法找到一个合适的回归起始点,并通过Cholesky分解对扩散张量进行正定约束.最后寻找局部最小获得DTI扩散张量的约束M估计,并在模拟二阶张量场和真实DTI数据集上进行了实验.与最小二乘法和M估计子回归模型相比,该方法可以更有效地排除热噪声和生理性离群点影响,对DTI扩散张量估计很有价值. 展开更多
关键词 扩散加权成像 扩散张量成像 张量估计 约束M估计 生理性噪声
下载PDF
Estimation of surface tension of organic compounds using quantitative structure-property relationship 被引量:2
3
作者 戴益民 刘又年 +3 位作者 李浔 曹忠 朱志平 杨道武 《Journal of Central South University》 SCIE EI CAS 2012年第1期93-100,共8页
A novel quantitative structure-property relationship (QSPR) model for estimating the solution surface tension of 92 organic compounds at 20℃ was developed based on newly introduced atom-type topological indices. Th... A novel quantitative structure-property relationship (QSPR) model for estimating the solution surface tension of 92 organic compounds at 20℃ was developed based on newly introduced atom-type topological indices. The data set contained non-polar and polar liquids, and saturated and unsaturated compounds. The regression analysis shows that excellent result is obtained with multiple linear regression. The predictive power of the proposed model was discussed using the leave-one-out (LOO) cross-validated (CV) method. The correlation coefficient (R) and the leave-one-out cross-validation correlation coefficient (Rcv) of multiple linear regression model are 0.991 4 and 0.991 3, respectively. The new model gives the average absolute relative deviation of 1.81% for 92 substances. The result demonstrates that novel topological indices based on the equilibrium electro-negativity of atom and the relative bond length are useful model parameters for QSPR analysis of compounds. 展开更多
关键词 surface tension quantitative structure-property relationship (QSPR) topological indice organic compound
下载PDF
Estimation of multipath parameters in wireless communications using multi-way compressive sensing 被引量:3
4
作者 Fangqing Wen Gong Zhang De Ben 《Journal of Systems Science & Complexity》 SCIE EI CSCD 2015年第5期908-915,共8页
This paper addresses the problem of joint angle and delay estimation(JADE) in a multipath communication scenario. A low-complexity multi-way compressive sensing(MCS) estimation algorithm is proposed. The received data... This paper addresses the problem of joint angle and delay estimation(JADE) in a multipath communication scenario. A low-complexity multi-way compressive sensing(MCS) estimation algorithm is proposed. The received data are firstly stacked up to a trilinear tensor model. To reduce the computational complexity,three random compression matrices are individually used to reduce each tensor to a much smaller one. JADE then is linked to a low-dimensional trilinear model. Our algorithm has an estimation performance very close to that of the parallel factor analysis(PARAFAC) algorithm and automatic pairing of the two parameter sets. Compared with other methods, such as multiple signal classification(MUSIC), the estimation of signal parameters via rotational invariance techniques(ESPRIT), the MCS algorithm requires neither eigenvalue decomposition of the received signal covariance matrix nor spectral peak searching. It also does not require the channel fading information, which means the proposed algorithm is blind and robust, therefore it has a higher working efficiency.Simulation results indicate the proposed algorithm have a bright future in wireless communications. 展开更多
关键词 multipath channel delay estimation direction-of-arrival (DOA) estimation multi-way compressive sensing (MCS)
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部