Journal of Emerging Trends in Engineering and Applied Sciences: Research Areas

Article Title: A Quick MRI Method Premised on Compressed Sensing and K-space Denoising by Apodization
by Henry Kiragu, Elijah Mwangi, and George Kamucha

Abstract:
Magnetic Resonance Imaging (MRI) is a medical diagnostic tool that has very significant advantages over other medical imaging modalities. However, it is characterized by long imaging times, reconstruction artifacts as well as high cost. In this paper, a proposed MRI method that addresses these problems is presented. The method is based on Compressed Sensing (CS), and the sparsity of k-space coefficients the wavelet transform domain. It commences with sub-Nyquist acquisition of the k-space data of the image followed by the full k-space reconstruction in the wavelet transform domain using the Orthogonal Matching Pursuit (OMP) algorithm. The wavelet coefficients are then transformed into the Discrete Fourier Transform (DFT) domain and then re-arranged into a vector to reveal the concomitant artifacts. The artifacts are then suppressed using a denoising apodization function. The denoised coefficients are then transformed into the reconstructed image. The Structural SIMilarity (SSIM) and the Peak Signal to Noise Ratio (PSNR) quality measures are used to assess the quality of the resulting images. The results show that this method yields an average PSNR improvement of more than 1.2 dB over the Stagewise Orthogonal Matching Pursuit (StOMP) method at 20% or more measurements. __________________________________________________________________________________________ Keywords: Compressed Sensing, MRI, Orthogonal Matching Pursuit, Apodization, Optimization

Keywords: Compressed Sensing, MRI, Orthogonal Matching Pursuit, Apodization, Optimization

Download full paper