TY - JOUR
T1 - T1ρ MRI contrast in the human brain
T2 - Modulation of the longitudinal rotating frame relaxation shutter-speed during an adiabatic RF pulse
AU - Michaeli, Shalom
AU - Sorce, Dennis J.
AU - Springer, Charles S.
AU - Ugurbil, Kamil
AU - Garwood, Michael
N1 - Funding Information:
This work was partially supported by NIH Grants CA92004, RR08079, NS40801, and EB00422, the Keck Foundation, and the MIND Institute. We thank Drs. Xin Li, William Rooney, and Amy Springer for stimulating discussions.
PY - 2006/7
Y1 - 2006/7
N2 - Longitudinal relaxation in the rotating frame (T1ρ) is the dominant mechanism during a train of adiabatic full passage (AFP) RF pulses with no interpulse intervals, placed prior to an excitation pulse. Asymptotic apparent time constants (T1 ρ′) were measured for human occipital lobe 1H2O at 4 T using brief imaging readouts following such pulse trains. Two members of the hyperbolic secant (HSn) AFP pulse family (n = 1 or 4; i.e., arising from different amplitude- and frequency-modulation functions) were used. These produced two different non-monoexponential signal decays during the pulse trains. Thus, there are differing contrasts in asymptotic T1 ρ′ maps derived from these data. This behavior is quite different than that of 1H2O signals from an aqueous protein solution of roughly the same macromolecular volume fraction as tissue. The ROI-averaged decays from the two acquisitions can be simultaneously accommodated by a two-site-exchange model for an equilibrium isochronous process whose exchange condition is modulated during the pulse. The model employs a two-spin description of dipolar interaction fluctuations in each site. The intrinsic site R1ρ(≡ T1 ρ- 1) value is sensitive to fluctuations at the effective Larmor frequency (ωeff) in the rotating frame, and this is modulated differently during the two types of AFP pulses. Agreement with the data is quite good for site orientation correlation time constants characteristic of macromolecule-interacting water (site A) and bulk-like water (site B). Since R1ρA is significantly modulated while R1ρB is not, the intrinsic relaxographic shutter-speed for the process (≡|R1ρA - R1ρB|), and thus the exchange condition, is modulated. However, the mean residence time (67 ms) and intrinsic population fraction (0.2) values found for site A are each rather larger than might be expected, suggesting a disproportionate role for the water molecules known to be "buried" within the large and concentrated macromolecules of in vivo tissue.
AB - Longitudinal relaxation in the rotating frame (T1ρ) is the dominant mechanism during a train of adiabatic full passage (AFP) RF pulses with no interpulse intervals, placed prior to an excitation pulse. Asymptotic apparent time constants (T1 ρ′) were measured for human occipital lobe 1H2O at 4 T using brief imaging readouts following such pulse trains. Two members of the hyperbolic secant (HSn) AFP pulse family (n = 1 or 4; i.e., arising from different amplitude- and frequency-modulation functions) were used. These produced two different non-monoexponential signal decays during the pulse trains. Thus, there are differing contrasts in asymptotic T1 ρ′ maps derived from these data. This behavior is quite different than that of 1H2O signals from an aqueous protein solution of roughly the same macromolecular volume fraction as tissue. The ROI-averaged decays from the two acquisitions can be simultaneously accommodated by a two-site-exchange model for an equilibrium isochronous process whose exchange condition is modulated during the pulse. The model employs a two-spin description of dipolar interaction fluctuations in each site. The intrinsic site R1ρ(≡ T1 ρ- 1) value is sensitive to fluctuations at the effective Larmor frequency (ωeff) in the rotating frame, and this is modulated differently during the two types of AFP pulses. Agreement with the data is quite good for site orientation correlation time constants characteristic of macromolecule-interacting water (site A) and bulk-like water (site B). Since R1ρA is significantly modulated while R1ρB is not, the intrinsic relaxographic shutter-speed for the process (≡|R1ρA - R1ρB|), and thus the exchange condition, is modulated. However, the mean residence time (67 ms) and intrinsic population fraction (0.2) values found for site A are each rather larger than might be expected, suggesting a disproportionate role for the water molecules known to be "buried" within the large and concentrated macromolecules of in vivo tissue.
KW - Adiabatic pulses
KW - Dipolar relaxation
KW - Shutter-speed
KW - T contrast
UR - http://www.scopus.com/inward/record.url?scp=33744969765&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33744969765&partnerID=8YFLogxK
U2 - 10.1016/j.jmr.2006.04.002
DO - 10.1016/j.jmr.2006.04.002
M3 - Article
C2 - 16675277
AN - SCOPUS:33744969765
SN - 1090-7807
VL - 181
SP - 135
EP - 147
JO - Journal of Magnetic Resonance
JF - Journal of Magnetic Resonance
IS - 1
ER -