Improved time series reconstruction for dynamic magnetic resonance imaging

IEEE Trans Med Imaging. 2009 Jul;28(7):1093-104. doi: 10.1109/TMI.2008.2012030. Epub 2009 Jan 13.

Abstract

Time series of in vivo magnetic resonance images exhibit high levels of temporal correlation. Higher temporal resolution reconstructions are obtained by acquiring data at a fraction of the Nyquist rate and resolving the resulting aliasing using the correlation information. The dynamic imaging experiment is modeled as a linear dynamical system. A Kalman filter based unaliasing reconstruction is described for accelerated dynamic magnetic resonance imaging (MRI). The algorithm handles arbitrary readout trajectories naturally. The reconstruction is causal and very fast, making it applicable to real-time imaging. In vivo results are presented for cardiac MRI of healthy volunteers.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Algorithms*
  • Fourier Analysis
  • Heart / anatomy & histology
  • Humans
  • Image Processing, Computer-Assisted / methods*
  • Magnetic Resonance Imaging / methods*
  • Time Factors