k-space weighted least-squares algorithm for accurate and fast motion extraction from magnetic resonance navigator echoes

Navigator echoes provide an effective method to monitor physiological motio n, allowing motion artifacts to be suppressed by modifying the data acquisi tion accordingly. The displacement can be measured directly from the naviga tor phase using an algorithm suggested by Ahn and Cho. Although computation ally efficient, it is susceptible to noise, particularly contributions from points at the edges of k-space. A k-space weighted least-squares (kLS) alg orithm is proposed which fits a straight line to the motion-induced phase s hift. The linear fit is weighted strongly to high SNR points near the k-spa ce center and only weakly to the poor SNR points at the edges of k-space. T his algorithm was found to be as efficient as Ahn's algorithm but more robu st against noise. (C) 2001 Wiley-Liss, Inc.