We present a technique to measure the longitudinal relaxation time constant of venous blood (T1b) in vivo in a few seconds. The MRI sequence consists of a thick-slab adiabatic inversion, followed by a series of slice-selective excitations and single-shot echo planar imaging readouts. The time intervals between excitations were chosen so that blood in macroscopic vessels is fully refreshed between excitations, making the blood signal follow an unperturbed inversion recovery curve. Static tissue, which experiences the inversion and all excitation pulses, quickly reaches a steady state at a low signal as a result of partial saturation. This allows blood-filled voxels to be discriminated from those containing static tissue, and to be fitted voxel-by-voxel to a simple inversion recovery model. The sequence was tested on a flow phantom with the proposed method, yielding T1 values consistent to within 3% of those obtained using a conventional inversion recovery sequence with a spin-echo readout. The method was applied to seven adult volunteers and 18 neonates. The blood T1 of the neonates (1799±206ms; range, 1393-2035ms) was found to be more variable than that of adults (1717±39ms; range, 1662-1779ms). A linear correlation between the inverse of T1b and the haematocrit was established in 12 neonates (R2=0.90). We present a sequence capable of measuring T1blood in vivo in less than 30 s. This method was found to yield T1 values in agreement with a standard method in a flow phantom. T1blood was measured in small cohorts of adults and neonates. Adult T1blood values were found to be reproducible and in agreement with literature values. In neonates, blood T1 was found to be more variable than in adults and was correlated with the haematocrit.
- Arterial spin labelling
- Inversion recovery
- Longitudinal relaxation time constant
- Neonatal imaging
ASJC Scopus subject areas
- Molecular Medicine
- Radiology Nuclear Medicine and imaging