Evaluation of Algebraic Iterative Algorithms for Reconstruction of Electron Magnetic Resonance Images

Electron Magnetic Resonance Imaging (EMRI) is emerging as a potential tool for noninvasive imaging of free radicals in biological systems. The EMR images are generally reconstructed using the filtered back projection (FBP) method, because of its speed and simplicity. But, the FBP method fails when reconstruction is attempted with limited number of projections or with noisy projections. In addition, it also suffers due to star artifact. Iterative algorithms such as additive algebraic reconstruction technique (AART), multiplicative algebraic reconstruction technique (MART) form another major class of tomographic reconstruction methods. These methods are known to give artifact free imaging with minimal number of projections in computerized tomography (CT), positron emission tomography (PET) and single photon emission tomography (SPECT). In this paper, the application of algebraic iterative reconstruction methods for reconstruction of EMR image is critically evaluated. Both phantom and in vivoEMR images are reconstructed using projection data collected from a radio frequency (RF) continuous wave (CW) EMR imager. Both visualization and computation of signal to noise ratio (SNR) are used to evaluate the performance of the methods. Nine different reconstruction methods viz. Brooks, Mayinger, Gordon, Gilbert and Anderson AART methods and four MART methods are compared. Among these methods, Mayinger method, a variant of AART performs better than the other methods.