抄録
NMR spectra are usually obtained from free-induction-decay signals by means of the computationally efficient fast Fourier transform (FFT). However, in spectral analysis by FFT, there are several inherent limitations such as low resolution for the truncated signal, or side lobes due to windowing. Recently, several least-squares alternative methods to the FFT have been proposed to reduce these problems.
A similar limitation can be also observed in the magnetic resonance (MR) image reconstruction. A simplest method would allow for reduced imaging times, through the collection of fewer phase encoding steps. But the limitation of available data leads to truncation artifacts in the FFT reconstruction. These artifacts, which are known as the Gibbs phenomenon in engineering, introduce uncertainty in the discrimination of anatomical details in the MR images. It is possible to reduce truncation artifacts by using low-pass filtering technique, but this produces a severe loss in resolution.
We propose a data extrapolation method for truncation artifact removal. This method is based on a linear prediction approach with singular value decomposition. Reconstruction results from phantom data and clinical data are presented to illustrate the performance of the proposed method.
