Removal of nuisance signals from limited and sparse 1H MRSI data using a union-of-subspaces model.

摘要: Purpose To remove nuisance signals (e.g., water and lipid signals) for 1H MRSI data collected from the brain with limited and/or sparse (k, t)-space coverage. Methods A union-of-subspace model is proposed removing signals. The exploits partial separability of both metabolite signal, decomposes an dataset into several sets generalized voxels that share same spectral distributions. This enables estimation has coverage. Results The method been evaluated using in vivo data. For conventional chemical shift imaging k-space coverage, produced “lipid-free” spectra without suppression during acquisition at 130 ms echo time. acquired pulses suppression, was also able to remaining negligible residuals. Conclusion Nuisance reside low-dimensional subspaces. property can be utilized removal even when they have coverage t)-space. should prove useful especially accelerated high-resolution brain. Magn Reson Med, 2015. © 2015 Wiley Periodicals, Inc.