Development of Arterial Spin Labeling Techniques forQuantitative Perfusion Measurements at 3 Tesla

2014 
Microvascular perfusion is an important parameter of high clinical value as it provides important information on tissue viability and function. In contrast to dynamic contrastenhanced magnetic resonance imaging (DCE-MRI), arterial spin labeling (ASL) is a method to assess quantitative perfusion values non-invasively by MRI. The main goal of this work was the development of a 3D ASL technique that allows the quantitative perfusion measurement within a whole volume. Further, the diagnostic signifcance of ASL was tested by applying it in dedicated renal disease models. ASL yielded significantly (P < 0.01) different values in healthy kidneys ((500+-91) ml/100g/min) compared to kidneys with acute kidney injury ((287+-83) ml/100/min) as well as for acutely rejected transplanted kidneys compared to chronically rejected grafts. A comparison to DCE-MRI showed no signifcant differences. While literature reports on 3D ASL sequences show a prevalence for spin-echo based data acquisitions, the presented sequence employs a 3D balanced steady-state free precession (bSSFP) readout that has the ability to overcome some of the drawbacks of spin-echo based 3D sequences. The developed 3D ASL technique includes several features, each of which has been optimized separately. Special emphasis was put on design, simulation, and implementation of a slice-selective adiabatic inversion pulse. In an initial measurement, the whole-brain perfusion of a healthy volunteer was assessed with an isotropic resolution of 3mm. Mean perfusion values were f_GM = (51+-17) ml/100g/min and f_WM = (24+-8) ml/100/min for gray and white matter, respectively, which are in very good agreement with findings from the "gold-standard" method of 15O-positron emission tomography. The diagnostic value of ASL in combination with the possibility to assess absolute perfusion values in a whole volume makes it a promising technique for future clinical diagnosis.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []