Image reconstruction algorithms in electromagnetic tomography

Image reconstruction from partial k-space acquisition plays a key role in magnetic resonance imaging (MRI). This paper presents a novel method for reconstructing spatial images from k-space data composed of the positive half k-space data together with a few negative spatial frequencies. The method is based on the notion of analytic image. It consists in splitting the given k-space into two parts, the positive half k-space and negative spatial frequencies. The two parts are then considered as the spectral representations of two analytic images. The final image is reconstructed by taking the magnitude of the sum of the two analytic images. The method is evaluated via experiments on both simulated and real human brain images, and compared with existing techniques