Improved k-t PCA Algorithm Using Artificial Sparsity in Dynamic MRI.

Yiran Wang,Lixia Yuan,Ling Xia,Zhifeng Chen,Feng Liu,Jing Wang

doi:10.1155/2017/4816024

Yiran Wang, Lixia Yuan + Show 4 more

Open Access

https://doi.org/10.1155/2017/4816024

Copy DOI

Abstract

The k-t principal component analysis (k-t PCA) is an effective approach for high spatiotemporal resolution dynamic magnetic resonance (MR) imaging. However, it suffers from larger residual aliasing artifacts and noise amplification when the reduction factor goes higher. To further enhance the performance of this technique, we propose a new method called sparse k-t PCA that combines the k-t PCA algorithm with an artificial sparsity constraint. It is a self-calibrated procedure that is based on the traditional k-t PCA method by further eliminating the reconstruction error derived from complex subtraction of the sampled k-t space from the original reconstructed k-t space. The proposed method is tested through both simulations and in vivo datasets with different reduction factors. Compared to the standard k-t PCA algorithm, the sparse k-t PCA can improve the normalized root-mean-square error performance and the accuracy of temporal resolution. It is thus useful for rapid dynamic MR imaging.

Highlights

High spatial and temporal resolutions are very important in dynamic magnetic resonance imaging clinical applications, such as functional MRI, cardiac cine imaging, and perfusion imaging among others
We propose an approach to combine k-t principal component analysis (k-t principal component analysis (PCA)) with artificial sparsity, termed sparse k-t PCA, for further improving the image reconstruction quality of dynamic magnetic resonance imaging (dMRI)
Its procedure is summarized as follows: (1) firstly, the inverse Fast Fourier Transform (IFFT) is conducted on the time-averaged k-space to obtain the direct-current (DC) image; (2) this time-averaged k-t space is subtracted from each frame of the sampled k-t data to get the residual k-t space; (3) after the k-t PCA reconstruction, the final image is constructed by adding the DC image and residual image together

Summary

Introduction

High spatial and temporal resolutions are very important in dynamic magnetic resonance imaging (dMRI) clinical applications, such as functional MRI, cardiac cine imaging, and perfusion imaging among others. For k-t BLAST/SENSE and k-t GRAPPA with the residual k-space method, in which the temporal invariant terms were calculated separately and subtracted from each frame before reconstruction, artificial sparsity was successfully developed for dMRI applications [5, 6]. In another artificial-sparsity-based work [20], a high-pass filter was used to suppress the low frequency parts while preserving the high frequency information for GRAPPA reconstruction, which corresponded to the image details and edge information. Comparison between sparse k-t PCA, standard k-t PCA, and residual k-t PCA is presented

Theory

Materials and Methods

Results

Discussion

Conclusions