Reconstruction of MRI Images based on Compressive Sensing

Magnetic Resonance Imaging (MRI) is a medical imaging technique for diagnosing patients suffering from torn ligaments to cancerous tumors. MRI is also very useful for examining the brain and spinal cord related issues. The size of the image decides how many times the Radio Frequency pulse needs to be applied, hence making the MRI scan a time consuming process. There is a growing requirement for reducing the diagnosing time in MRI necessitating investigation to accelerate this non linear optimization problem. Compressed Sensing (CS) aims to reconstruct images from considerably smaller measurements than were thought before. Applying CS to MRI offers potentially significant scan time reductions. A method to MRI reconstruction from under sampled measurements is proposed in this paper. Discrete Wavelet Transform is used to sparsely characterize MRI. CS reconstruction is performed using l
₁
norm by applying Basis pursuit primal-dual interior point method. Experimental results reveal that the proposed method reconstructs images with good quality.