Rapid T₁ mapping using multislice echo planar imaging

Abstract

Determination of neurological pathology in white matter disease can be made in a semiquantitative way from T 1 -or T 2 -weighted images. A higher level of quantification based on measured T 1 or T 2 values has been either limited to specific regions of interest or to low-resolution maps. Higher-resolution T 1 maps have proved difficult to obtain due to the excessively long scan times required using conventional techniques. In this study, clinically acceptable images are obtained by using single-shot echo planar imaging (EPI) with an acquisition scheme that maximizes signal-to-noise while minimizing the scan time. The recent interest in methods for fast T 1 mapping (1-5), particularly in the human brain, highlights the increased desire for more quantitative in vivo measurements using MRI. For example, in multiple sclerosis the number of "black-holes" on T 1 -weighted scans has shown some correlation with disease severity (6). However, hypointensity on a conventional MR scan is not an objective measure, and results can be difficult to compare between research sites, due to variations in the sequence parameters used and the subjective criterion used by the person interpreting the scan. The use of T 1 mapping as a more quantitative approach has many attractions, as the value of T 1 may report more about the underlying pathology than does visual interpretation of T 1 -weighted scans (7).The main reason for T 1 maps not being used more routinely is the lengthy time it can take to obtain such maps using conventional methods. The standard inversion recovery pulse sequence involves considerable delays in the period between inversion and excitation pulses (8). This has limited T 1 maps to a few low-resolution slices, often with acquisition times of greater than 1 hr. To be clinically useful, T 1 maps need to cover a large volume of the brain with good resolution, and be acquired in scan times of minutes rather than hours.A considerable increase in the speed of T 1 mapping can be achieved by implementing the Look-Locker method (9). This method uses many low flip angle acquisitions to inspect a single recovery curve following an inversion pulse, thereby reducing the need for long delays between several inversions. The Look-Locker method (and other variants) greatly reduces the time taken to measure T 1 ; however, there are some penalties, notably in reduced signal-to-noise and in the sensitivity of the T 1 fit to the flip angle used for the excitation pulses. A combination of EPI and the Look-Locker approach (LL-EPI) probably represents the fastest T 1 mapping sequence that can be achieved, with a T 1 recovery curve from a single slice being acquired in 3 sec (10). Clearly, though, this method has the same inherent problems as the conventional LookLocker approach.Here we use an approach that uses EPI together with a more conventional [invert]-TI-[excite] strategy. This enables accurate whole brain, T 1 mapping to be carried out in only 3 min. METHODSAll experiments were carried out on a 3.0 Tesla whole body scanner with a V...

Modeling the effects of dispersion and pulsatility of blood flow in pulsed arterial spin labeling

“…The actual T 1 in the deep gray ROI 1 is likely to be shorter than this (17), and in the occipital ROI 2 the value may be longer (18). Changing the tissue T 1 value will affect all three models in a similar way-primarily affecting the decay rate of the falling portion of the kinetic curve, but also the height of the peak.…”

Radio frequency magnetic field mapping of a 3 Tesla birdcage coil: Experimental and theoretical dependence on sample properties

“…For example, it has been shown recently that by actively modulating the TX power as function of the axial slice position it is possible to partly compensate for axial B 1 inhomogeneity of 3 T birdcage coils (33). Moreover, the knowledge of the B 1 spatial distribution at 3 T is of importance for a posteriori MRI correction techniques such as relaxation time measurements (34) and image segmentation (35). As an additional application, accurate RF B 1 mapping may also find use in low frequency (100 -300 MHz) EPR imaging and Overhauser MRI (36).…”

Error reduction and parameter optimization of the TAPIR method for fast T1 mapping

“…As noted in a number of previous publications (8,11), it is extremely important for accurate and precise reconstruction of T 1 relaxation times to account for inversion pulse imperfections. Thus, it is common to assume that the measure of the inversion pulse (in)efficiency should be included in the calculation of the T 1 maps.…”

“…where x 1 …x n are parameters to be fitted from the experimental data such as M 0 , [8] where is the measurement error (standard deviation) of the experimental data points. The diagonal elements of the covariance matrix C ii are equal to the variance of the parameters x i , and the off-diagonal elements reflect the correlations between the parameters.…”

“…The interpretation of the predicted variance of the fitted parameters requires explanation: clearly, the absolute value of (x i ) depends on the model parameters and the "raw" noise level, given by in Eq. [8]. In principle, this noise level is an absolute value and is independent of sample properties (M 0 and T 1 ) and/or sequence parameters (TR, , etc).…”