Note
Go to the end to download the full example code.
Shuffled echo-resolved 3D FSE#
Shuffled 3D FSE uses the same refocusing-train design as conventional FSE and differs in two separate choices. The support is a variable-density Poisson-disc draw, which determines which views are acquired. The ordering determines when each view is acquired: as in T2 Shuffling [TAM17], every echo time samples views over the whole sampled extent, and the echo of each view is drawn as a complementary Poisson-disc split of the support [LEV17], so the views of each echo are themselves a variable-density Poisson-disc pattern and neighbouring echoes do not repeat each other’s neighbourhoods. This is the sampling condition of echo-resolved subspace reconstruction [TAM17], with the aliasing of each echo kept away from the point-spread function’s main lobe; no reconstruction is performed here.
from pypulseqpp import sequences
P = {
"n_x": 128,
"n_y": 96,
"n_z": 48,
"fov_x": 0.20,
"fov_y": 0.20,
"fov_z": 0.12,
"etl": 64,
"te": 100e-3,
"tr": 1.4,
"ry": 2,
"rz": 2,
"n_acs_y": 8,
"n_acs_z": 6,
"n_dummy": 0,
"ordering": "shuffling",
"flip_modulation": "optimized",
"wave_amplitude": 0.0,
}
seq = sequences.fse3D_sequence(**P)
Variable-density sampling#
A fully sampled calibration region is embedded in a variable-density
Poisson-disc support, selected with
make_cartesian_plane_sampling(..., sampling="poisson"). The echo index of
each view is then assigned by make_complementary_order()
rather than by its distance from the k-space centre: the views of each echo
form a Poisson-disc pattern of their own, and views of neighbouring echoes
keep a minimum distance from one another.

Echo-time distribution#
The refocusing train is designed as in Conventional 3D fast spin echo, whose figures show the schedule; shuffling changes the assignment of views to echoes, not the refocusing schedule. With a 64-echo train, each echo index occurs throughout the sampled extent, so the contrast evolution along the train is not confined to a radial k-space band. The figure gives the distance of every acquired view from the k-space centre against its echo time: every echo time samples views from the centre to the edge of the support.

References#
Tamir JI, Uecker M, Chen W, Lai P, Alley MT, Vasanawala SS, Lustig M. T2 shuffling: sharp, multicontrast, volumetric fast spin-echo imaging. Magnetic Resonance in Medicine. 2017;77(1):180-195. https://doi.org/10.1002/mrm.26102
Levine E, Daniel B, Vasanawala S, Hargreaves B, Saranathan M. 3D Cartesian MRI with compressed sensing and variable view sharing using complementary Poisson-disc sampling. Magnetic Resonance in Medicine. 2017;77(5):1774-1785. https://doi.org/10.1002/mrm.26254
Total running time of the script: (0 minutes 0.617 seconds)