With Institutional Review Board approval and informed
consent/assent, in vivo studies were performed on 22 consecutive pediatric
patients, ranging from 2.2 to 10.7 years of age. These patients consisted of
13 females and 9 males. They were referred for abdominal MRI with different
clinical indications that are specified inTable 1 . The protocol included dynamic contrast-enhanced imaging
performed with the patient freely breathing. Studies were performed on a GE
MR750 3T scanner (Waukesha, WI, USA) using a 32-channel cardiac coil and a
3D spoiled gradient echo acquisition sequence. A flip angle of 15°,
a readout bandwidth of ±100 khz, partial readout (0.6 of the full
readout) to achieve minimum echo time (1.2–1.3 ms), and minimum TR
(3.0–3.4 ms) were used for these studies. Motion was estimated using
Butterfly navigators (23 (link),32 ). For each TR, the navigator
acquisition was 0.10–0.12 ms long and reached a maximum k-space
radius of 0.4–0.6 cm−1. The acquisition of
3–4 temporal phases was designed using VDRad where each temporal
phase had an acceleration of ~6. In our reconstructions, data from
the different temporal phases are combined into one phase. Fat-suppression
was incorporated with a periodic spectral fat-inversion pulse (inversion
time of 9.0 ms) with 24–27 views following each fat-inversion pulse.
Data were acquired ~1.5 min after intravenous gadolinium-based
contrast administration during the venous phases.
All pediatric patient studies were acquired completely
free-breathing under light anesthesia. For comparison, a prospective
respiratory-triggered/gated scan was performed immediately after the
proposed scan using conventional respiratory bellows. These scans were
acquired using a trigger point of 30% of maximum and an acceptance
window of 30%.Table 1 describes ages, genders, specific contrast agent used, and other details
about each study. After the data acquisition, motion was estimated from the
navigators. Images were reconstructed using a combination of C/C++ and
Matlab (Mathworks, Natick, MA, USA).
consent/assent, in vivo studies were performed on 22 consecutive pediatric
patients, ranging from 2.2 to 10.7 years of age. These patients consisted of
13 females and 9 males. They were referred for abdominal MRI with different
clinical indications that are specified in
performed with the patient freely breathing. Studies were performed on a GE
MR750 3T scanner (Waukesha, WI, USA) using a 32-channel cardiac coil and a
3D spoiled gradient echo acquisition sequence. A flip angle of 15°,
a readout bandwidth of ±100 khz, partial readout (0.6 of the full
readout) to achieve minimum echo time (1.2–1.3 ms), and minimum TR
(3.0–3.4 ms) were used for these studies. Motion was estimated using
Butterfly navigators (23 (link),32 ). For each TR, the navigator
acquisition was 0.10–0.12 ms long and reached a maximum k-space
radius of 0.4–0.6 cm−1. The acquisition of
3–4 temporal phases was designed using VDRad where each temporal
phase had an acceleration of ~6. In our reconstructions, data from
the different temporal phases are combined into one phase. Fat-suppression
was incorporated with a periodic spectral fat-inversion pulse (inversion
time of 9.0 ms) with 24–27 views following each fat-inversion pulse.
Data were acquired ~1.5 min after intravenous gadolinium-based
contrast administration during the venous phases.
All pediatric patient studies were acquired completely
free-breathing under light anesthesia. For comparison, a prospective
respiratory-triggered/gated scan was performed immediately after the
proposed scan using conventional respiratory bellows. These scans were
acquired using a trigger point of 30% of maximum and an acceptance
window of 30%.
about each study. After the data acquisition, motion was estimated from the
navigators. Images were reconstructed using a combination of C/C++ and
Matlab (Mathworks, Natick, MA, USA).