Gaussian-Radial Under-Sampling Based CSMRI Reconstruction using a Modified Interpolation Approach

Abstract

Magnetic Resonance Imaging (MRI) is used to produce detailed images of body tissues and organs using strong magnets and radio waves, but with a very slow acquisition process. Compressed Sensing (CS) has efficiently accelerated the MRI acquisition process by employing different reconstruction strategies using a fraction of the Nyquist samples. This scan time can be further reduced using a new technique called interpolated compressed sensing (iCS) by exploiting the inter-slice correlation of multi-slice MRI. In this paper, a modified fast interpolated compressed sensing (Mod-FiCS) technique is proposed using the Gaussian-Radial under-sampling scheme. The Gaussian-Radial under-sampling approach adopted by Mod-FiCS has an edge that it neither shows any streaking artifacts like Radial nor blurred edges like Gaussian. The new interpolation approach used in Mod-FiCS technique uses three consecutive slices to estimate the missing samples. Six evaluation metrics are used to analyze the performance of the proposed technique such as structural similarity index measurement (SSIM), feature similarity index measurement (FSIM), mean square error (MSE), peak signal to noise ratio (PSNR), correlation (CORR), and sharpness index (SI), and compared with recent sampling and interpolation techniques. The simulation result shows that the proposed technique has improvement both quantitatively and qualitatively.