DOI:10.2214/AJR.07.2903
AJR 2008; 190:W140-W150
© American Roentgen Ray Society
Simple Changes to 1.5-T MRI Abdomen and Pelvis Protocols to Optimize Results at 3 T
Daniel Cornfeld1 and
Jeffery Weinreb
1 Both authors: Department of Diagnostic Radiology, Yale University School of
Medicine, PO Box 208042, New Haven, CT 06520-8042.
Received July 20, 2007;
accepted after revision September 10, 2007.
Address correspondence to D. Cornfeld
(daniel.cornfeld{at}yale.edu).
J. Weinreb belongs to the speakers bureau of GE Healthcare and is a
consultant for Bayer HealthCare.
WEB
This is a Web exclusive article.
CME
This article is available for CME credit. See
www.arrs.org
for more information.
Abstract
OBJECTIVE. Simply transposing 1.5-T abdominal and pelvic MRI
protocols to 3-T scanners does not result in optimized images. Simple
modifications can be made to preexisting 1.5-T protocols to obtain image
quality at 3 T that is comparable to that at 1.5 T.
CONCLUSION. We illustrate several simple modifications to 1.5-T body
protocols that maintain image quality at 3 T.
Keywords: 3 T • abdominal protocol • MRI • pelvic protocol
Introduction
A common perception about 3-T MRI scanners is that the quality of abdominal
and pelvic imaging is worse than that at 1.5 T. The benefits of imaging at 3 T
in the brain, spine, and musculoskeletal systems are not immediately obvious
in the abdomen and pelvis. Artifacts that are easily manageable at 1.5 T may
be accentuated at 3 T and compromise diagnostic quality. Body imaging
protocols that are reliable at 1.5 T may be completely unsuitable at 3 T.
This article will show how knowledge of the differences between 3 and 1.5 T
can be used to direct modifications in standard 1.5-T protocols to produce
quality abdominal and pelvic images at 3 T. Specific challenges in abdominal
and pelvic imaging at 3 T are posed by dielectric and susceptibility
artifacts, in- and opposed-phase imaging, changes in optimal TR and TE, motion
artifacts, and increased specific absorption rates (SARs)
[1,
2].
Our body imaging protocols are based on our experiences with the
commercial, short-bore, 3-T scanner at our institution (Signa HDx, GE
Healthcare). However, for the purposes of this article we have attempted to
keep all terms vendor-neutral.
Dielectric Effect
The dielectric effect results in decreased signal in the middle of the
patient volume (Fig. 1A,
1B). This artifact has two
causes. First, at 3 T the wavelength of the radiofrequency pulse approximates
the diameter of the patient, resulting in standing wave effects that prevent
the radiofrequency pulse from exciting spins in the center of the volume
[3,
4]. Second, eddy currents
induced by the radiofrequency pulse generate a magnetic field that opposes the
main magnetic field, also decreasing excitation
[5]. Both effects are more
pronounced at 3 T than at 1.5 T. Artifact from the dielectric effect is more
pronounced in patients with a large amount of intraabdominal water or fat.
View larger version (150K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —63-year-old man with ascites. Axial T2-weighted single-shot
fast spin-echo image at 3 T through mid abdomen shows loss of signal in center
of patient volume secondary to dielectric effect.
|
|
The dielectric effect may be reduced, but not eliminated, by placing a
specially designed pad (called a radiofrequency cushion or dielectric pad)
between the patient and the receiver coil (Fig.
2A,
2B). In the future, this
artifact will probably be managed with techniques such as radiofrequency
shimming that are still under development. For the present, dielectric
artifacts can seriously compromise the diagnostic quality of the image,
especially for fast spin-echo and single-shot techniques. Gradient-echo
sequences are less affected. Patients with large girth (including pregnant
patients) or abundant ascites are best scanned at 1.5 T (Figs.
3 and
4).
View larger version (51K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —46-year-old woman with normal pelvis. Axial T1-weighted fast
spin-echo images through pelvis at 3 T show large amount of dielectric
artifact centrally (A) and resolution of artifact with dielectric pad
(B).
|
|
View larger version (64K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —46-year-old woman with normal pelvis. Axial T1-weighted fast
spin-echo images through pelvis at 3 T show large amount of dielectric
artifact centrally (A) and resolution of artifact with dielectric pad
(B).
|
|
View larger version (144K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 — 26-year-old woman with abdominal pain. Axial T2-weighted
single-shot fast spin-echo image through mid abdomen at 3 T with dielectric
pad shows loss of signal in center of image despite use of pad. In some
patients, dielectric pad cannot remove artifact completely.
|
|
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —55-year-old man with cirrhosis, ascites, and elevated
bilirubin level. Coronal T2-weighted single-shot fast spin-echo image through
mid abdomen at 3 T with dielectric pad. Despite signal loss over center of
liver secondary to dielectric effect, common duct (arrow) is clearly
seen to ampulla. Duct is normal caliber. No obstructing mass or stone is
present. This patient did not have biliary obstruction as a cause of increased
bilirubin. In some patients, diagnosis is still made despite dielectric
artifact. This patient should have been scanned on 1.5-T scanner.
|
|
Susceptibility Artifacts
Susceptibility artifacts are due to local field distortions induced by
paramagnetic substances such as air and metal in the main magnetic field. The
effect worsens as the magnetic field strength increases (Fig.
5A,
5B). It can be mitigated by
imaging using a shorter TE (which can be accomplished by imaging at a higher
receiver bandwidth) and by imaging at higher spatial resolution
[6] (Fig.
6A,
6B). Imaging at higher
bandwidth results in a decreased signal-to-noise ratio (SNR), but this is
theoretically offset by the increased signal achieved by imaging at a shorter
TE. We found that higher-bandwidth imaging at 3 T mitigates susceptibility
artifacts, does not significantly compromise overall image quality, and
shortens breath-hold times (Fig.
7A,
7B,
7C). Another way to shorten
the TE is to decrease resolution in the frequency direction—that is,
decrease the frequency matrix. This will also result in a higher SNR
throughout the image.
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —55-year-old woman with history of metastatic colon cancer to
liver after wedge resection in right posterior lobe. T1-weighted gradient-echo
image through liver with TE of 2.15 milliseconds obtained on 1.5-T scanner. At
1.5 T, this is an out-of-phase image and india ink artifact can be seen at
interface of abdominal organs and peritoneal fat. Note amount of blooming
around surgical clips in right posterior liver (arrow).
|
|
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —55-year-old woman with history of metastatic colon cancer to
liver after wedge resection in right posterior lobe. T1-weighted gradient-echo
image through liver with TE of 2.3 milliseconds obtained on 3-T scanner. At 3
T, this is an in-phase image and no india ink artifact is seen. Note how much
more blooming is seen around surgical clips (arrow) despite the same
short TE.
|
|
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A —46-year-old woman with history of breast cancer. Sagittal
contrast-enhanced 3D T1-weighted fat-saturated fast spoiled gradient-recalled
echo image obtained at 1.5 T through breast (TR/TE, 6.5/3.2; flip angle,
10°; voxel size, 3 x 0.78 x 1.04 mm for a volume of 2.4
mm3). Note amount of blooming artifact from surgical clips
(arrow).
|
|
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B —46-year-old woman with history of breast cancer. Sagittal
contrast-enhanced 3D T1-weighted fat-saturated fast spoiled gradient-recalled
echo image obtained at 3 T through breast (6.7/2.6; flip angle, 10°; voxel
size, 2.2 x 0.625 x 1.04 mm for a volume of 1.4 mm3).
Amount of blooming from surgical clips (arrow) is the same as on
image obtained at 1.5 T. This is caused by shorter TE and higher spatial
resolution.
|
|
View larger version (92K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A —35-year-old woman with pelvic pain; images obtained at 3 T.
Unenhanced axial T1-weighted fat-saturated 3D fast spoiled gradient-recalled
echo image through pelvis (TR/TE, 3.5/1.98; flip angle, 12°; bandwidth, 63
kHz; imaging time, 27 seconds; field of view, 36 cm; slice thickness, 4 mm;
matrix, 320 x 256).
|
|
View larger version (87K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7B —35-year-old woman with pelvic pain; images obtained at 3 T.
Unenhanced axial T1-weighted fat-saturated 3D fast spoiled gradient-recalled
echo image through pelvis (3.8/1.69; flip angle, 12°; bandwidth, 83 kHz;
imaging time, 24 seconds; field of view, 36 cm; slice thickness, 4 mm; matrix,
320 x 256).
|
|
View larger version (92K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7C —35-year-old woman with pelvic pain; images obtained at 3 T.
Unenhanced axial T1-weighted fat-saturated 3D fast spoiled gradient-recalled
echo image through pelvis (3.5/1.55; flip angle, 12°; bandwidth, 100 kHz;
imaging time, 22 seconds; field of view, 36 cm; slice thickness, 4 mm; matrix,
320 x 256). There is no visible difference in signal between images
A, B, and C. Voxel size is the same. Increased bandwidth (which
reduces signal) results in decreased TE (which increases signal due to less
T2* decay), and these effects counteract each other. However, breath-hold time
is shorter with longer-bandwidth acquisition, which results in less potential
motion artifact from breathing.
|
|
In- and Opposed-Phase Imaging
The time between the in- and opposed-phase echoes is decreased at 3 T. The
first opposed-phase echo is at 1.1 milliseconds and the first in-phase echo is
at 2.2 milliseconds (compared with 2.2 and 4.4 milliseconds at 1.5 T). Scanner
receivers do not have sufficiently high bandwidths to receive both the first
opposed-phase and the first in-phase echoes in a single 2D acquisition
[1]. By decreasing the flip
angle, decreasing the TR, and increasing the bandwidth, 2D images can be
obtained at a TE of close to 1.1 milliseconds, the first opportunity for an
opposed-phase echo. An image with an in-phase TE close to 2.2 can be then be
obtained in a separate acquisition (Fig.
8A,
8B,
8C,
8D). However, decreasing the
TR and flip angle decreases the contrast between liver and spleen compared
with similar images at 1.5 T. This is probably secondary to increased T1 times
for liver and spleen at 3 T
[7]. Also, acquiring the two
echoes as separate acquisitions allows room for operator error (i.e., the TR,
flip angle, or matrix could be different for each acquisition) and
misregistration of images. Prototype sequences have remedied this problem
(Fig. 9A,
9B,
9C,
9D).
View larger version (118K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8A —25-year-old man with geographic fatty infiltration of liver.
T1-weighted out-of-phase 2D gradient-echo image obtained at 1.5 T (TR/TE,
250/2.2; flip angle, 80°; slice thickness, 5 mm; interval, 5 mm; matrix,
288 x 224).
|
|
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8B —25-year-old man with geographic fatty infiltration of liver.
T1-weighted in-phase 2D gradient-echo image obtained at 1.5 T (250/4.4; flip
angle, 80°; slice thickness, 5 mm; interval, 5 mm; matrix, 288 x
224). Images A and B were obtained in the same breath-hold,
which ensures that TR, flip angle, field of view, matrix, and receiver
bandwidth are identical for each acquisition and eliminates possibility of
image misregistration.
|
|
View larger version (107K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8C —25-year-old man with geographic fatty infiltration of liver.
T1-weighted out-of-phase 2D gradient-echo image obtained at 3 T (175/1.4; flip
angle, 60°; slice thickness, 5 mm; interval, 5 mm; matrix, 288 x
224).
|
|
View larger version (98K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8D —25-year-old man with geographic fatty infiltration of liver.
T1-weighted in-phase 2D gradient-echo image obtained at 3 T (175/2.2; flip
angle, 60°; slice thickness, 5 mm; interval, 5 mm; matrix, 288 x
224). Images C and D are obtained in separate breath-holds. If
technologist inadvertently changes TR or matrix to optimize breath-hold time,
the two acquisitions will not match. Also, if patient breathes differently
between the two acquisitions, image misregistration will result. Compared with
A and B, contrast between liver and spleen in images C
and D is decreased. Shorter TR and smaller flip angle, combined with
longer T1 relaxation times at 3 T, result in less T1 weighting.
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A —45-year-old man with diffuse fatty infiltration of liver.
T1-weighted out-of-phase 2D gradient-echo image obtained at 3 T (TR/TE,
175/1.4; flip angle, 60°; slice thickness, 5 mm; interval, 5 mm; matrix,
288 x 224).
|
|
View larger version (118K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B —45-year-old man with diffuse fatty infiltration of liver.
T1-weighted in-phase 2D gradient-echo image obtained at 3 T (175/2.2; flip
angle, 60°; slice thickness, 5 mm; interval, 5 mm; matrix, 288 x
224). Images A and B are obtained in separate breath-holds.
|
|
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9C —45-year-old man with diffuse fatty infiltration of liver.
T1-weighted out-of-phase 3D dual-echo fast spoiled gradient-recalled echo
image obtained at 3 T (4/1.3; flip angle, 12°; slice thickness, 4 mm;
interval, 4 mm; matrix, 320 x 224).
|
|
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9D —45-year-old man with diffuse fatty infiltration of liver.
T1-weighted in-phase 3D dual-echo fast spoiled gradient-recalled echo image
obtained at 3 T (4/2.2; flip angle, 12°; slice thickness, 4 mm; interval,
4 mm; matrix, 320 x 224). Images C and D are obtained in
same breath-hold. Liver is darker in C than in A because TE is
closer to true out-of-phase TE of 1.1 milliseconds. At time of this writing,
this sequence is not a commercial product. However, this should solve reported
problems with chemical shift imaging at 3 T.
|
|
Changes in Optimal TR and TE
At 3 T, the SNR in the liver changes significantly over the range of
normally used TEs (Fig. 10A,
10B,
10C,
10D,
10E). We prefer the increased
signal in the liver achieved with the shorter TE because overall image quality
is increased and the appearance of the liver is comparable to that obtained at
1.5 T. Because we do not use the T2 single-shot images as our primary tool for
detecting hepatic neoplasms, we do not think that our ability to evaluate
hepatic neoplasms is decreased. The TE for single-shot images should therefore
be shortened compared with corresponding values at 1.5 T. In addition,
long-echo-train imaging produces image blurring secondary to changing TEs
across the image acquisition (each line of k-space is filled at a different
TE). Single-shot images at 3 T are especially susceptible because of their
long echo-trains and the rapidly decaying signal. Parallel imaging shortens
the echo-train length and should be used whenever possible (Fig.
11A,
11B). In contrast to the
liver, TEs in the pelvis should be lengthened to increase the conspicuity of
the uterine zonal anatomy (Fig.
12A,
12B,
12C,
12D,
12E).
View larger version (121K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10A —34-year-old healthy female volunteer with normal liver. Axial
T2-weighted single-shot fast spin-echo images through abdomen at 3 T with
varying effective TEs of 70 (A), 90 (B), 105 (C), 120
(D), and 140 (E) milliseconds. Each image was acquired as a
single slice in a single breath-hold (field of view, 36 cm; slice thickness, 5
mm; interval, 6 mm; matrix, 320 x 224). Note progressive signal loss
throughout liver and remainder of abdomen as TE increases from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted single-shot fast spin-echo
acquisition is 100 milliseconds. At 3 T, we use TE of 70 milliseconds.
|
|
View larger version (115K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10B —34-year-old healthy female volunteer with normal liver. Axial
T2-weighted single-shot fast spin-echo images through abdomen at 3 T with
varying effective TEs of 70 (A), 90 (B), 105 (C), 120
(D), and 140 (E) milliseconds. Each image was acquired as a
single slice in a single breath-hold (field of view, 36 cm; slice thickness, 5
mm; interval, 6 mm; matrix, 320 x 224). Note progressive signal loss
throughout liver and remainder of abdomen as TE increases from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted single-shot fast spin-echo
acquisition is 100 milliseconds. At 3 T, we use TE of 70 milliseconds.
|
|
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10C —34-year-old healthy female volunteer with normal liver. Axial
T2-weighted single-shot fast spin-echo images through abdomen at 3 T with
varying effective TEs of 70 (A), 90 (B), 105 (C), 120
(D), and 140 (E) milliseconds. Each image was acquired as a
single slice in a single breath-hold (field of view, 36 cm; slice thickness, 5
mm; interval, 6 mm; matrix, 320 x 224). Note progressive signal loss
throughout liver and remainder of abdomen as TE increases from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted single-shot fast spin-echo
acquisition is 100 milliseconds. At 3 T, we use TE of 70 milliseconds.
|
|
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10D —34-year-old healthy female volunteer with normal liver. Axial
T2-weighted single-shot fast spin-echo images through abdomen at 3 T with
varying effective TEs of 70 (A), 90 (B), 105 (C), 120
(D), and 140 (E) milliseconds. Each image was acquired as a
single slice in a single breath-hold (field of view, 36 cm; slice thickness, 5
mm; interval, 6 mm; matrix, 320 x 224). Note progressive signal loss
throughout liver and remainder of abdomen as TE increases from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted single-shot fast spin-echo
acquisition is 100 milliseconds. At 3 T, we use TE of 70 milliseconds.
|
|
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10E —34-year-old healthy female volunteer with normal liver. Axial
T2-weighted single-shot fast spin-echo images through abdomen at 3 T with
varying effective TEs of 70 (A), 90 (B), 105 (C), 120
(D), and 140 (E) milliseconds. Each image was acquired as a
single slice in a single breath-hold (field of view, 36 cm; slice thickness, 5
mm; interval, 6 mm; matrix, 320 x 224). Note progressive signal loss
throughout liver and remainder of abdomen as TE increases from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted single-shot fast spin-echo
acquisition is 100 milliseconds. At 3 T, we use TE of 70 milliseconds.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11A —34-year-old healthy female volunteer with normal pancreas.
Axial T2-weighted single-shot fast spin-echo images through abdomen at 3 T
with (A) and without (B) parallel imaging (TE, 70 milliseconds;
field of view, 36 cm; slice thickness, 5 mm; interval, 6 mm; matrix, 320
x 224). Parallel imaging acceleration factor in A is 2. Edges in
A are sharper than in B.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11B —34-year-old healthy female volunteer with normal pancreas.
Axial T2-weighted single-shot fast spin-echo images through abdomen at 3 T
with (A) and without (B) parallel imaging (TE, 70 milliseconds;
field of view, 36 cm; slice thickness, 5 mm; interval, 6 mm; matrix, 320
x 224). Parallel imaging acceleration factor in A is 2. Edges in
A are sharper than in B.
|
|
View larger version (159K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12A —34-year-old healthy female volunteer with normal uterus.
Sagittal T2-weighted fast spin-echo images through pelvis at 3 T with
different effective TEs as indicated (field of view, 24 cm; slice thickness, 4
mm, interval, 5 mm; echo-train length, 28; matrix, 320 x 288). Notice
how soft-tissue contrast in uterus changes as TE changes from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted fast spin-echo in pelvis is
100 milliseconds. At 3 T, we use TE of 120–140 milliseconds, which also
produces good contrast in prostate.
|
|
View larger version (177K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12B —34-year-old healthy female volunteer with normal uterus.
Sagittal T2-weighted fast spin-echo images through pelvis at 3 T with
different effective TEs as indicated (field of view, 24 cm; slice thickness, 4
mm, interval, 5 mm; echo-train length, 28; matrix, 320 x 288). Notice
how soft-tissue contrast in uterus changes as TE changes from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted fast spin-echo in pelvis is
100 milliseconds. At 3 T, we use TE of 120–140 milliseconds, which also
produces good contrast in prostate.
|
|
View larger version (164K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12C —34-year-old healthy female volunteer with normal uterus.
Sagittal T2-weighted fast spin-echo images through pelvis at 3 T with
different effective TEs as indicated (field of view, 24 cm; slice thickness, 4
mm, interval, 5 mm; echo-train length, 28; matrix, 320 x 288). Notice
how soft-tissue contrast in uterus changes as TE changes from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted fast spin-echo in pelvis is
100 milliseconds. At 3 T, we use TE of 120–140 milliseconds, which also
produces good contrast in prostate.
|
|
View larger version (182K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12D —34-year-old healthy female volunteer with normal uterus.
Sagittal T2-weighted fast spin-echo images through pelvis at 3 T with
different effective TEs as indicated (field of view, 24 cm; slice thickness, 4
mm, interval, 5 mm; echo-train length, 28; matrix, 320 x 288). Notice
how soft-tissue contrast in uterus changes as TE changes from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted fast spin-echo in pelvis is
100 milliseconds. At 3 T, we use TE of 120–140 milliseconds, which also
produces good contrast in prostate.
|
|
View larger version (170K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12E —34-year-old healthy female volunteer with normal uterus.
Sagittal T2-weighted fast spin-echo images through pelvis at 3 T with
different effective TEs as indicated (field of view, 24 cm; slice thickness, 4
mm, interval, 5 mm; echo-train length, 28; matrix, 320 x 288). Notice
how soft-tissue contrast in uterus changes as TE changes from 70 to 140
milliseconds. At 1.5 T, typical TE for T2-weighted fast spin-echo in pelvis is
100 milliseconds. At 3 T, we use TE of 120–140 milliseconds, which also
produces good contrast in prostate.
|
|
Motion
Motion artifacts are increased at 3 T. We initially thought that the motion
artifacts seen with fast spin-echo techniques could be decreased by shortening
the echo acquisition time [8].
This is achieved by shortening the echo-train length, increasing the
bandwidth, and using parallel imaging. However, we obtained better image
quality by increasing the echo-train, decreasing the bandwidth, and turning
off parallel imaging (Fig.
13A,
13B,
13C,
13D). This also increases the
SNR and allows higher-resolution imaging. In addition, placing the frequency
direction anterior-to-posterior reduces motion from the anterior pelvic wall.
Additional artifacts from bowel motion may be decreased by using an
antiperistaltic agent such as glucagon (GlucaGen, Novo Nordisk) or
hyoscine-N-butylbromide (Buscopan, Boehringer Ingelheim). The latter
is not approved for use in the United States (Fig.
14A,
14B). However, we do not
routinely administer antiperistaltic agents for pelvic imaging.
View larger version (186K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13A —72-year-old woman with pelvic pain and intramural fibroid.
Sagittal (A) and axial (B) T2-weighted fast spin-echo images
through pelvis at 3 T (TR/TE, 5,700/120; field of view, 24 cm; slice
thickness, 4 mm; interval, 5 mm; matrix, 320 x 224). Echo-train length
is 15; bandwidth is 60 kHz. Parallel imaging was used to reduce effective
echo-train length. Frequency direction is superior-to-inferior on sagittal
image and left-to-right on axial image. Saturation pulses were placed on
anterior abdominal wall to reduce artifact from breathing. Image quality is
poor.
|
|
View larger version (189K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13B —72-year-old woman with pelvic pain and intramural fibroid.
Sagittal (A) and axial (B) T2-weighted fast spin-echo images
through pelvis at 3 T (TR/TE, 5,700/120; field of view, 24 cm; slice
thickness, 4 mm; interval, 5 mm; matrix, 320 x 224). Echo-train length
is 15; bandwidth is 60 kHz. Parallel imaging was used to reduce effective
echo-train length. Frequency direction is superior-to-inferior on sagittal
image and left-to-right on axial image. Saturation pulses were placed on
anterior abdominal wall to reduce artifact from breathing. Image quality is
poor.
|
|
View larger version (194K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13C —72-year-old woman with pelvic pain and intramural fibroid.
Sagittal (C) and axial (D) T2-weighted fast spin-echo images
through pelvis (5,700/120; field of view, 24 cm; slice thickness, 5 mm;
interval, 5 mm; matrix, 320 x 224). Echo-train length is 28. Parallel
imaging was not used. Frequency direction is anterior-to-posterior on both
sagittal and axial images. Image quality is significantly improved.
|
|
View larger version (181K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13D —72-year-old woman with pelvic pain and intramural fibroid.
Sagittal (C) and axial (D) T2-weighted fast spin-echo images
through pelvis (5,700/120; field of view, 24 cm; slice thickness, 5 mm;
interval, 5 mm; matrix, 320 x 224). Echo-train length is 28. Parallel
imaging was not used. Frequency direction is anterior-to-posterior on both
sagittal and axial images. Image quality is significantly improved.
|
|
View larger version (135K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14A — 27-year-old woman with torsed ovary and fallopian tube.
Axial T2-weighted fast spin-echo image through pelvis at 3 T (TR/TE,
5,700/120; field of view, 24 cm; slice thickness, 4 mm; interval, 5 mm;
matrix, 320 x 224). Echo-train length is 28. Parallel imaging was not
used. Frequency direction is anterior-to-posterior. Image is blurry because of
bowel motion.
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14B — 27-year-old woman with torsed ovary and fallopian tube.
Axial T2-weighted fast spin-echo image through pelvis after IV administration
of 1 mg of glucagon (5,700/120; field of view, 24 cm; slice thickness, 4 mm;
interval, 5 mm; matrix, 320 x 224). Echo-train length is 28. Parallel
imaging was not used. Frequency direction is anterior-to-posterior. Tortuous,
thickened fallopian tube is much more clearly seen (arrow) because
artifact from bowel motion is reduced.
|
|
Specific Absorption Rate
SAR, specific absorption rate, is a measure of the rate that energy is
absorbed by the body when exposed to radiofrequency pulses. Because SAR is
proportional to the square of the field strength, radiofrequency pulses at 3 T
deposit four times as much energy as radiofrequency pulses at 1.5 T. Many
protocol modifications are performed automatically by the scanner to reduce
the SAR. These techniques include modified radiofrequency pulse profiles,
increased dead time, and flip angle modulation techniques. An important area
in which limitations from SAR come into play is single-shot imaging because
long echo-trains deposit a lot of radiofrequency energy. To compensate, the
scanner increases the time between slice excitations, resulting in longer
scanning times. For example, a single-shot fast spin-echo sequence on our
1.5-T scanner obtains 30 slices in 25 seconds with 860 milliseconds between
each slice excitation. Obtaining 30 slices at 3 T takes 35 seconds with 1,200
milliseconds between each slice excitation. This is the difference between a
single- and a double-breath-hold acquisition and can result in image
misregistration on the two different breath-holds even if the patient is a
good breath-holder. Using respiratory triggering increases the time between
slice excitations to a level at which the SAR is not an issue and also
produces a stack of correctly registered images (Figs.
15A,
15B and S15, which is
available in the supplemental data for this article at
www.arrs.org).
View larger version (174K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 15A —36-year-old healthy male volunteer. See Figures S15A and S15B
at
www.arrs.org
for these cine imaging files. Single still image from stack of coronal
single-shot fast spin-echo images through liver and biliary tree (field of
view, 36 cm; slice thickness, 5 mm; interval, 6 mm; matrix, 320 x 224).
TE is 70 milliseconds. Parallel imaging acceleration factor is 2. Stack was
obtained during two separate breath-holds. Notice how diaphragm and liver
bounce up and down on every other image. Although patient held his breath
satisfactorily, breath-hold position was different for the two acquisitions.
Images are difficult to scroll through, and a small gallstone could
potentially be missed if it moved in and out of the imaging planes.
|
|
View larger version (168K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 15B —36-year-old healthy male volunteer. See Figures S15A and S15B
at
www.arrs.org
for these cine imaging files. Single still image from stack of coronal
single-shot fast spin-echo images through liver and biliary tree (field of
view, 36 cm; slice thickness, 5 mm; interval, 6 mm; matrix, 320 x 224).
TE is 70 milliseconds. Parallel imaging acceleration factor is 2. Stack was
obtained using respiratory triggering. Notice how diaphragm and liver are in
same position on each image. No misregistration is seen.
|
|
Conclusion
Simple protocol changes produce abdominal and pelvic images at 3 T that are
as good as those obtained at 1.5 T. A dielectric pad is always used to
compensate for dielectric artifact. Susceptibility artifacts are decreased by
using a higher receiver bandwidth and imaging at a shorter TE. Image quality
is preserved with the added benefit of shorter breath-hold times. Decreasing
the flip angle and shortening the TR allows 2D in- and out-of-phase imaging at
the first opposed-phased echo. Prototype 3D sequences can also be used to
acquire images at the first opposed- and in-phase echoes. Decreasing the TE to
70 milliseconds for single-shot imaging in the abdomen (vs 100 milliseconds at
1.5 T) produces images with increased signal. Parallel imaging shortens the
echo-train and results in sharper images. Increasing the TE to 120–140
milliseconds for T2-weighted fast spin-echo imaging in the pelvis (vs 100
milliseconds at 1.5 T) increases soft-tissue contrast in the uterus and
prostate. Image quality in the pelvis can be also increased by avoiding
parallel imaging, increasing the echo-train length, decreasing the receiver
band-width, and placing the frequency direction in the anteroposterior
direction. Respiratory triggering of single-shot sequences reduces
misregistration and reduces the SAR.
References
- Merkle E, Dale BM, Paulson EK. Abdominal MR imaging at 3T.
Magn Reson Imaging Clin N Am 2006;14
: 17–26[CrossRef][Medline]
- Martin D, Friel H, Danrad R, De Becker J, Hussain S. Approach to
abdominal imaging at 1.5 Tesla and optimization at 3 Tesla. Magn
Reson Imaging Clin N Am 2005;13
: 241–254[CrossRef][Medline]
- Brix B, Seebass M, Helwig G, Greibel J. Estimation of heat transfer
and temperature rise in partial-body regions during MR procedures: an
analytical approach with respect to safety considerations. Magn
Reson Imaging 2002; 20:65
–76[CrossRef][Medline]
- Tropp J. Image brightening in samples of high dielectric constant.
J Magn Reson 2004;167
: 12–24[CrossRef][Medline]
- Collins CM, Liu W, Schreiber W, Yang QX, Smith MB. Central
brightening due to constructive interference with, without, and despite
dielectric resonance. J Magn Reson Imaging2005; 21:192
–196[CrossRef][Medline]
- Li T, Mirowitz SA. Fast multi-planar gradient echo MR imaging:
impact of variation in pulse sequence parameters on image quality and
artifacts. Magn Reson Imaging 2004;22
: 807–814[CrossRef][Medline]
- de Bazelaire CM, Duhamel GD, Rofsky NM, Alsop DC. MR imaging
relaxation times of abdominal and pelvic tissues measured in vivo at 3.0 T:
preliminary results. Radiology 2004;230
: 652–659[Abstract/Free Full Text]
- Hussain S, van den Bos I, Oliveto J, Martin D. MR imaging of the
female pelvis at 3T. Magn Reson Imaging Clin N Am2007; 14:537
–544[CrossRef]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
