动态磁共振成像中利用时空滤波的非笛卡尔稀疏数据重建新算法

One New Algorithm of Utilizing Spatio-Temporal Filtering for the Reconstruction of Non-Cartesian Sparse Data in Dynamic MRI

通过非笛卡尔稀疏采样,可以显著缩短磁共振成像数据采集时间,在动态磁共振中具有良好的应用前景。现有的动态图像重建算法主要利用信号在时间域上的相关性完成图像重建,对信号在空间域上的相关性探讨不足。本文提出利用时空滤波的非笛卡尔稀疏数据重建新算法,同时考虑采集信号在时间域和空间域上的相关性。对动态的磁共振数据在时间域上采用改进的Hanning滤波,以克服数据量的不足;在空间域上,引入高频增强约束以突出图像中的细节信息。仿真实验结果显示,该方法重建出时间分辨率良好、细节比较清晰的图像,实现了稀疏采集的非笛卡尔数据在k空间的重建。

Sparsely non-Cartesian sampling indicates a promising future for dynamic magnetic resonance imaging （MRI） because it can shorten acquisition time remarkably. Current reconstruction algorithms for dynamic images mainly utilize signal correlations in temporal domain, which insufficiently use the signal correlations in spatial domain. We proposed a new algorithm for the reconstruction of non-Cartesian sparse data in dynamic MRI, which exploited signal correlations in time to circumvent data sparsity and employed high-frequency enhancement in k-space to emphasize edge sharpness. Simulated experiments showed that this proposed approach allowed k-space based reconstruction results with good temporal resolution and spatial resolution of non-Cartesian sparse data in dynamic MRI.