Simple recipe for accurate T(2) quantification with multi spin-echo acquisitions

MAGMA. 2014 Dec;27(6):567-77. doi: 10.1007/s10334-014-0438-3. Epub 2014 Mar 19.

Abstract

Objective: The quantification of magnetic resonance relaxation parameters T 1 and T 2 have the potential for improved disease detection and classification over standard clinical weighted imaging. Performing a mono-exponential fit on multi spin-echo (MSE) data provides quantitative T 2 values in a clinically acceptable scan-time. However, due to technical imperfections of refocusing pulses, stimulated echo contributions to the signals lead to significant deviations in the resulting T 2 values. In this work, a simple auto-calibrating correction procedure is presented, allowing the accurate estimation of T 2 from MSE acquisitions.

Materials and methods: Correction factors for T 2 values obtained from MSE acquisitions with a mono-exponential fit are derived from simulations following the extended phase graph formulation. A closed formula is given for the calculation of the required correction factors directly from the measured data itself.

Results: Simulations and phantom experiments show high accuracy of corrected T 2 values for a wide range of clinically relevant T 2 values and for different nominal refocusing flip angles. In addition, corrected T 2 maps of the human brain are presented.

Conclusion: A simple recipe is provided to correct T 2 values obtained from MSE acquisitions via a mono-exponential fit for the influence of stimulated echoes. Since all required parameters are extracted from the data themselves, no additional acquisitions are required.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms*
  • Artifacts*
  • Brain / anatomy & histology*
  • Echo-Planar Imaging / instrumentation
  • Echo-Planar Imaging / methods*
  • Humans
  • Image Enhancement / methods*
  • Image Interpretation, Computer-Assisted / methods*
  • Numerical Analysis, Computer-Assisted
  • Phantoms, Imaging
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Signal Processing, Computer-Assisted*