CARDIAC CINE MRI USING COMPRESSIVE SENSING PRINCIPLES

MR images can be reconstructed from undersampled k-t space data to increase image acquisition speed. We propose a new method to undersample the k-space and reconstruct images based on Compressive Sensing (CS) theory. To this end, statistical features extracted from each trajectory are clustered by the fuzzy c-means (FCM) method. The resulting class labels are considered as the states of a Markov chain. A hidden Markov model (HMM) is then trained to find the transition matrix. Trajectories having more non-diagonal transition matrices arechosen to sample data along them. An iterative thresholding algorithm is then used for reconstruction of the image. The proposed method outperforms two other methods in reconstructing half sampled Cardiac Cine MRI data. The use of fuzzy clustering as an intermediate tool to study complicated phenomena by HMM, applicability to non-dynamic MRI data, robustness to noise, faster and more accurate reconstruction describe specifications of the proposed method.

Keywords:

Cine Cardiac MRI , fuzzy c-Mean , HMM , Compressive Sensing , Inverse Problems