k‐Space filtering in 2D gradient‐echo breath‐hold hyperpolarized ³He MRI: Spatial resolution and signal‐to‐noise ratio considerations

Abstract

In this work some of the factors that can influence the signal-tonoise ratio (SNR) and spatial resolution in MR images of inhaled hyperpolarized gases are systematically addressed. In particular, the effects of RF depletion of longitudinal polarization and image gradient diffusion dephasing were assessed in terms of their contribution to a k-space filter. By means of theoretical simulations and a novel method of experimental validation using a variable transverse magnetization of the 1 H signal, systematic quantitative and qualitative investigations of the effects of k-space filtering intrinsic to imaging of hyperpolarized gas were made. A 2D gradient-echo image is considered for a range of flip angles with centric, sequential, and half-Fourier Cartesian phase-encoding strategies, and the results are assessed in terms of SNR and spatial resolution in the reconstructed images. Centric phase encoding was found to give the best SNR at higher flip angles, with a trade-off in spatial resolution compared to sequential phase encoding. A half-Fourier approach potentially offers increased SNR through the use of higher flip angles without compromising the spatial resolution, which is comparable to that achieved with sequential encoding. Magn Reson Med 47:687-695, 2002.

A flexible 32‐channel receive array combined with a homogeneous transmit coil for human lung imaging with hyperpolarized 3He at 1.5 T

“…A 2D SPGR sequence without phase encoding gradients was used for this purpose. A single coronal slice of 40 cm thickness was used which extended beyond the lungs, thus avoiding complications in the flip angle calculation resulting from the slice profile flip angle distribution (12). Depletion of hyperpolarization due to RF irradiation then allows estimation of the flip angle a following the equation:…”

“…This is expected, as the RF calibration was done using a nonselective RF pulse, while the B 1 þ maps were acquired using a 2D slice-selective sequence. Slice profile effects are known to reduce the effective measured flip angle in 3 He imaging (12).…”

“…Every RF pulse destroys a fraction of the finite available longitudinal magnetization, and image SNR strongly depends on the delivered flip angle and hence homogeneity of the transmit field. For example, for a Spoiled Gradient Recalled Echo (SPGR) sequence with constant flip angle and sequential phase encoding, the optimal flip angle is given by a opt ¼ arctanð1= ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi N k À 1 p Þ, with N k being the number of RF pulses to the center of k-space (12). A homogeneous transmit field ensures that the actual flip angle is close to a opt across the whole volume.…”

Achieving high spatial resolution and high SNR in low-field MRI of hyperpolarised gases with Slow Low Angle SHot

“…This will be addressed in Section 3 and in B. Still, evaluating the impact of gas diffusion on image resolution and intensity using a free diffusion assumption is a standard approach [5,23,24]. It provides a convenient way to (over) estimate signal attenuation and to define ranges of acquisition parameters that avoid strong diffusion-induced losses.…”

“…They are conveniently discussed using, for instance, 3/ π 2 D =1.6 ms/mm 2 (the value for pure 3 He gas). Equation 9 shows that, as derived in [23], millimetre resolutions in the phase-encode direction can be obtained using conventional hardware and 0.5-1 ms gradient durations. Similar resolutions can be obtained in the read directions as well, provided that ultra-fast (sub-ms) acquisition is performed (Eq.…”