Abstract: In order to speed up the Magnetic Resonance Imaging(MRI) and to obtain the magnitude and phase information with high reliability, an algorithm for simultaneous reconstructing magnitude and phase images of MRI is proposed based on dual tree complex wavelet transform. Under the compressed sensing framework, the proposed algorithm employs the dual tree complex wavelet transform as the sparse representation for magnitude and phase parts, and therefore benefits from the multi-directional selectivity and shift invariance of the dual tree complex wavelet transform. Experimental results show that, for different datasets, the algorithm can improve the quality of the reconstructed phase and magnitude images to some extent.