Elimination of T2 artifacts caused by RARE sequence in MRI

This paper proposes a new method which removes artifacts due to T2 relaxation in Rapid Acquisition with Relaxation Enhancement (RARE) images. The method removes the artifacts by dividing the average of two Hermitian symmetric data points in k-space by a hyperbolic cosine function prior to inverse Fourier transformation. In order to evaluate the effectiveness of the method, computer simulations were done in which T2 distribution was spatially variant. The results showed that the proposed method worked well for spatially variant T2 decay and was less sensitive to noise than the conventional method using the geometric mean of two symmetric points and that using an exponential filter. In addition, the method yielded good results for a wide range of T2 values.