A fast model independent method for automatic correction of intensity nonuniformity in MRI data

A novel nonparametric approach for correcting intensity nonuniformity in ma gnetic resonance (MR) images is described. This approach is based solely on the assumption that the various sources of nonuniformity in MR imaging giv e rise to smooth variations in image intensity, and that these variations c an be extracted and corrected for, The advantage of this computationally fa st method is that it can be applied early in quantitative analysis while be ing independent of pulse sequence and is insensitive to pathological proces ses. This algorithm has been tested on both simulated and real data. Applic ation to tissue segmentation and functional MR imaging has shown a marked i mprovement in quantitative analysis. (C) 1999 Wiley-Liss, Inc.