ALTERNATIVES TO THE USE OF THE DFT IN MRI AND SPECTROSCOPIC RECONSTRUCTIONS

Standard and functional magnetic resonance imaging (MRI and f MRI] mak e of use the two-dimensional (2D) discrete Fourier transform (DFT). Ma ny MR spectroscopic techniques use the 1D DFT. Experimental or time co nstraints frequently require that the DFT be applied to finite-length (truncated) data sequences. Truncation is essentially a windowing of t he data and introduces artifacts and resolution loss. in images or spe ctra, A number of alternative reconstruction algorithms have been prop osed to counteract these problems. These algorithms attempt to model t he known data and use the modeling information to implicitly or explic itly extrapolate the data to overcome the windowing. One modeling appr oach,;he Transient Error Reconstruction Algorithm (TERA), uses an auto regressive moving average method to recover the missing data, In this article, we briefly discuss variants of the TERA algorithm and develop ment of neural networks to take better account of the differing data p roperties of MR data sets, Our success with neural networks in fMRI re construction has led us to challenge some of the standard approaches t o validating MR algorithms and develop our own, These new approaches i nclude k-space phantom generation and automated computer observer (ROC analysis) to evaluate algorithms in terms of their clinical relevance . We have also developed an upgraded image quality measure based on Da ly's Visual Differences Predictor, This models the ability of the huma n visual system to detect significant differences between images produ ced by different MRI reconstruction algorithms. We also present. a pro tocol for generating Shepp-Logan phantoms which avoids the introductio n of the high-frequency k-space data distortion present in the existin g approach. (C) 1997 John Wiley & Sons, Inc.