HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Akisik, F. M. Articles by Lall, C. Search for Related Content PubMed PubMed Citation Articles by Akisik, F. M. Articles by Lall, C. Related Collections Gastrointestinal Radiology Magnetic Resonance Imaging Related Articles

Abdominal MR Imaging at 3.0 T¹

¹ From the Department of Radiology, Indiana University School of Medicine, 550 N University Blvd, Suite UH 0279, Indianapolis, IN 46202. Presented as an education exhibit at the 2006 RSNA Annual Meeting. Received February 12, 2007; revision requested April 12 and received May 14; accepted May 22. F.M.A. is a consultant with Repligen and the Bracco Group, K.S. received a grant from Koninklijke Philips Electronics, and A.M.A. is a consultant with Repligen and Algotec Systems; all remaining authors have no financial relationships to disclose. Address correspondence to K.S. (e-mail: ksandras{at}iupui.edu).

	Abstract

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

 
Magnetic resonance (MR) imaging at 3.0 T offers an improved signal-to-noise ratio compared with that at 1.5 T. However, the physics of high field strength also brings disadvantages, such as increases in the specific absorption rate, in magnetic field inhomogeneity effects, and in susceptibility artifacts. The use of 3.0-T MR imaging for abdominal evaluations, in particular, has lagged behind that for other applications because of the difficulty of imaging a large volume while compensating for respiratory motion. At a minimum, abdominal MR imaging at 3.0 T requires modifications in the pulse sequences used at 1.5 T. Such modifications may include a decrease in the flip angle used for refocusing pulses and an increase in the repetition time for T1-weighted acquisitions. In addition, parallel imaging and other techniques (hyper-echo sequences, transition between pseudo steady states) may be used to maintain a high signal-to-noise ratio while decreasing acquisition time and minimizing the occurrence of artifacts on abdominal MR images.

© RSNA, 2007

	Introduction

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

 
Since the late 1980s, the predominant magnetic field strength used for clinical magnetic resonance (MR) imaging has been 1.5 T. Images acquired with 1.5-T MR imaging systems are of excellent diagnostic quality. However, with the introduction of clinical 3.0-T systems by most major vendors in recent years, a shift to the use of higher field strengths has occurred. The trend toward higher field strengths has been driven by the desire for improved image quality (the signal-to-noise ratio [SNR] increases with higher field strength) as well as commercial considerations.

Technical parameters that are especially important when discussing high-field-strength abdominal MR imaging include the SNR, chemical shift effects, specific absorption rate (SAR), radiofrequency effects, and susceptibility effects. The authors discuss the relevance of these parameters at 3.0 T and, where the comparison is appropriate, at 1.5 T. The substantial differences between T1- and T2-weighted pulse sequences used for 3.0-T MR imaging and those used for 1.5-T MR imaging are highlighted. The advantages and disadvantages of various parallel imaging techniques for improving the quality of 3.0-T abdominal MR images are reviewed. The utility of 3.0 T for diffusion-weighted imaging and MR spectroscopy also is considered.

	Important Technical Parameters

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

 
Signal-to-Noise Ratio
The net spin vector is proportional to the magnetic field strength (B₀). At body temperature the difference between parallel and antiparallel spins at 1.5 T is five per million nuclei. The number at 3.0 T is double that, and this increase in net spin produces a higher signal intensity than that at 1.5 T after the application of a radiofrequency excitation pulse (1). Since noise is independent of field strength, the SNR at 3.0 T also should be double that at 1.5 T.

However, the SNR can be improved by increasing imaging time (eg, by increasing the number of signals averaged). Thus, when comparing field strengths, it is better to evaluate the SNR achieved in a given amount of imaging time than to consider the SNR in isolation from other acquisition parameters. The simplified equation that defines the SNR in spin-echo (SE) sequences (2) is as follows: SNR B₀ · (N_ex · N_p/BW), where B₀ is the field strength, N_ex is the number of excitations, N_p is the number of phase-encoding lines, and BW is the receiver bandwidth.

The factors N_ex and N_p are measures of the image acquisition time. According to the preceding formula, when other parameters remain constant, the SNR at 3.0 T is double that at 1.5 T. However, there are factors that tend to counteract this benefit. At higher field strengths, the two most important effects are the increases in tissue T1 values and in the radiofrequency energy (the SAR) needed to magnetically excite the protons in tissue. The increased T1 values produce reductions both in signal intensity and in tissue contrast, effects that may require an increase in repetition time (TR) and, hence, an increase in imaging time. The increased SAR likewise limits the choice of pulse sequence and may necessitate increased imaging time. The two parameters together reduce the actual gain in SNR at 3.0 T, which may vary by a factor of 0.8 to 5.6 compared with the SNR at 1.5 T (3).

Chemical Shift Artifacts
Chemical shift artifacts, which may adversely affect the diagnostic quality of images, are increased at high field strengths. Misregistration artifact results from a slight difference between the precessional frequency of water protons and that of fat protons (methylene in triglyceride and fatty acids). This difference in frequency causes spatial misregistration of the signals from fat and from water by one or more pixels in the frequency encoding direction. A misregistration artifact commonly occurs where the signals from fat and water are superposed; it appears as a bright line on one side of an organ and a dark line on the opposite side. The shift is proportional to the magnetic field strength and is 223 Hz and 447 Hz at 1.5 T and 3.0 T, respectively (4). A misregistration artifact is easily recognized and usually is not a cause for concern. The artifact can be reduced by increasing the receiver bandwidth (5), which in turn leads to a reduction of the SNR. Doubling the bandwidth has the effect of reducing the SNR by about 30% (1,6). An alternative method for reducing the severity of misregistration artifact is to use a fat suppression technique; however, such techniques have the disadvantage of increasing the acquisition time, the SAR, or both.

The second type of chemical shift artifact, which is sometimes called in-phase–out-of-phase effect or cancellation artifact, also is due to the difference in the precessional frequencies of fat and water. Chemical shift artifact appears on gradient-echo images because the signals of water and fat are alternately summed or subtracted (canceled), depending on the echo time (TE). The effect is seen mostly in voxels in which the signal intensity of fat is approximately the same as the signal intensity of water (Fig 1). It does not appear on images obtained with standard SE sequences, because SE sequences contain one or more refocusing radiofrequency pulses that ensure that the water and fat signals are in phase regardless of the TE. The artifact is not related to the direction of frequency encoding or phase encoding, and its magnitude does not vary with the magnetic field strength; however, the time at which the artifact occurs during acquisition is field strength dependent (6). At MR imaging with 1.5 T, the fat and water protons are in phase when the TE is an even multiple of 2.2 msec and out of phase when the TE is an odd multiple of 2.2 msec. With 3.0 T, the water and fat protons are out of phase when the TE is an odd multiple of 1.1 msec.

View larger version (123K): [in this window] [in a new window] [Download PPT slide]   Figure 1a.  Chemical shift artifact of the second type (cancellation artifact). T1-weighted gradient-echo images obtained at 1.5 T (TR msec/TE msec, 130/2.2; flip angle, 70°) (a) and 3.0 T (111/6.15; flip angle, 70°) (b) show areas of reduced signal intensity (arrows) at the interface of organs and surrounding fat. Reduced contrast between the liver and the spleen at 3.0 T is probably due to the longer T1 and a resultant decrease in liver signal intensity. Because the TE was higher at 3.0 T than at 1.5 T, high iron content in the liver also may have contributed to the signal intensity reduction.

View larger version (126K): [in this window] [in a new window] [Download PPT slide]   Figure 1b.  Chemical shift artifact of the second type (cancellation artifact). T1-weighted gradient-echo images obtained at 1.5 T (TR msec/TE msec, 130/2.2; flip angle, 70°) (a) and 3.0 T (111/6.15; flip angle, 70°) (b) show areas of reduced signal intensity (arrows) at the interface of organs and surrounding fat. Reduced contrast between the liver and the spleen at 3.0 T is probably due to the longer T1 and a resultant decrease in liver signal intensity. Because the TE was higher at 3.0 T than at 1.5 T, high iron content in the liver also may have contributed to the signal intensity reduction.

The value of B₁ at 3.0 T is double that at 1.5 T. Therefore, the SAR is increased fourfold at 3.0 T compared with the SAR obtained with application of the same pulse sequence at 1.5 T. Several pulse sequence parameters affect the SAR, but the sequence may be altered to ameliorate these SAR effects (Table 1).

Hyperechoes are produced by the application of varying flip angles and varying pulse phases that are symmetrically arranged around a central 180° pulse to completely refocus magnetization (7,8). The signal of the hyperecho is complex and influenced by many factors, including transverse magnetization, longitudinal magnetization, and stimulated echo components of individual radiofrequency pulses, as well as the relaxation and diffusion effects that accumulate between each radiofrequency pulse (5). The TE of hyperecho sequences must be increased to produce T2-weighted contrast similar to that provided by standard fast SE sequences. Hyperecho sequences also may be used in diffusion-weighted imaging (9) and echoplanar imaging.

TRAPS is based on the observation that once a static pseudo steady state is attained, refocusing flip angles can be varied freely, with negligible signal loss. By using higher flip angles only for the central part of k-space, reductions in SAR by factors of three to six may be achieved (10).

Radiofrequency Effects
Artifacts may result also from the interactions of radiofrequency waves used in MR imaging. The frequency of the waves increases proportionally with increasing magnetic field strength. Consequently, the wavelength of radiofrequency waves used for tissue excitation is 23 cm at 3.0 T, compared with 47 cm at 1.5 T (1). When the radiofrequency wavelength is similar to the diameter of the torso (as is the case at 3.0 T but not at 1.5 T), standing waves may occur (3). Standing waves may cause spatial variation in the amplitude of a radiofrequency excitation pulse and thereby produce an artifact that may completely degrade image quality.

In addition, a rapidly changing magnetic field may produce eddy currents, which may further reduce the intensity of the radiofrequency transmission (1,5). The likelihood of eddy currents occurring is proportional to the frequency of the radiofrequency waves, with the currents being more likely to occur at higher field strengths.

When they occur in combination, these effects produce ill-defined areas with little or no signal at abdominal imaging, particularly in patients with a large girth or in those with ascites. This problem can be ameliorated in most cases by placing a dielectric blanket on the anterior aspect of the torso (Fig 2); however, that remedy is not always effective (1,3). Future improvements in radiofrequency transmission technology may help reduce the inhomogeneity that leads to this artifact.

View larger version (124K): [in this window] [in a new window] [Download PPT slide]   Figure 2a.  Standing waves artifact. (a) T2-weighted fast SE image obtained at 3.0 T (1000/105; flip angle, 115°) in a 32-year-old woman believed to have chronic pancreatitis shows a reduction in signal intensity of most of the left lateral liver lobe (arrow). (b) T2-weighted fast SE image obtained with the same sequence as a shows mitigation of the artifact with the placement of a water-filled cushion (arrowhead) over the abdomen, a method that is effective in some but not all cases.

View larger version (127K): [in this window] [in a new window] [Download PPT slide]   Figure 2b.  Standing waves artifact. (a) T2-weighted fast SE image obtained at 3.0 T (1000/105; flip angle, 115°) in a 32-year-old woman believed to have chronic pancreatitis shows a reduction in signal intensity of most of the left lateral liver lobe (arrow). (b) T2-weighted fast SE image obtained with the same sequence as a shows mitigation of the artifact with the placement of a water-filled cushion (arrowhead) over the abdomen, a method that is effective in some but not all cases.

View larger version (132K): [in this window] [in a new window] [Download PPT slide]   Figure 3a.  Susceptibility artifact. Axial gadolinium-enhanced T1-weighted gradient-echo abdominal MR images at similar levels, obtained at 1.5 T (4.85/2.48; flip angle, 12°) (a) and 3 months later at 3.0 T (4.47/1.75; flip angle, 10°) (b), show siderotic nodules (arrowhead) in a 42-year-old woman with alcoholic cirrhosis. The nodules are more visible at 3.0 T than at 1.5 T because the magnetic susceptibility effect of iron content is increased at the higher field strength. A splenic varix also is depicted (arrow).

View larger version (134K): [in this window] [in a new window] [Download PPT slide]   Figure 3b.  Susceptibility artifact. Axial gadolinium-enhanced T1-weighted gradient-echo abdominal MR images at similar levels, obtained at 1.5 T (4.85/2.48; flip angle, 12°) (a) and 3 months later at 3.0 T (4.47/1.75; flip angle, 10°) (b), show siderotic nodules (arrowhead) in a 42-year-old woman with alcoholic cirrhosis. The nodules are more visible at 3.0 T than at 1.5 T because the magnetic susceptibility effect of iron content is increased at the higher field strength. A splenic varix also is depicted (arrow).

	Parallel Imaging Techniques

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

In parallel imaging, the range of sampled phase-encoding lines in k-space determines the spatial resolution: A larger range results in a higher resolution. In addition, the distance between phase-encoding lines in k-space determines the FOV: Wider spacing between lines yields a smaller FOV. In SENSE acquisitions, an initial low-resolution full-FOV calibration sequence is applied to collect the sensitivity profiles of the individual receiver coils. Subsequently, a rapid acquisition is performed with high spatial resolution and a small FOV. The true positions of the aliased signals in the high-resolution image data set then are mathematically calculated by using the calibration data (Fig 4). In theory, the acquisition time may be reduced by a factor (denoted as the acceleration factor R) equivalent to the number of independent coil elements used, which currently ranges from four to 32. However, in practice, only a time reduction by a factor of two to three is possible, because the SNR is inversely proportional to the square root of R and becomes unacceptable with higher R values.

View larger version (21K): [in this window] [in a new window] [Download PPT slide]   Figure 4a.  Comparison between conventional MR imaging and SENSE (image domain–based parallel imaging). (a) Diagram shows the traditional method of obtaining a high-resolution MR image with a full FOV. (b) Diagram of the SENSE method shows the acquisition of a reduced number (by a factor of two to four) of phase-encoding lines. The range of the phase-encoding lines is the same as that with nonparallel imaging techniques and provides the same high resolution; however, the spacing between the lines is greater, producing an aliased image with a smaller FOV. A calibration acquisition, which is performed before the image acquisition, supplies spatial information based on the different physical locations of the independent coil elements (typically, there are four to 32 such elements). This information is used to mathematically calculate the correct position of the aliased pixels. The combination of the calibration acquisition with SENSE imaging results in a considerable reduction of acquisition time while providing high-resolution, full-FOV images comparable to those obtainable with a nonparallel imaging technique. FT = Fourier transformation.

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 4b.  Comparison between conventional MR imaging and SENSE (image domain–based parallel imaging). (a) Diagram shows the traditional method of obtaining a high-resolution MR image with a full FOV. (b) Diagram of the SENSE method shows the acquisition of a reduced number (by a factor of two to four) of phase-encoding lines. The range of the phase-encoding lines is the same as that with nonparallel imaging techniques and provides the same high resolution; however, the spacing between the lines is greater, producing an aliased image with a smaller FOV. A calibration acquisition, which is performed before the image acquisition, supplies spatial information based on the different physical locations of the independent coil elements (typically, there are four to 32 such elements). This information is used to mathematically calculate the correct position of the aliased pixels. The combination of the calibration acquisition with SENSE imaging results in a considerable reduction of acquisition time while providing high-resolution, full-FOV images comparable to those obtainable with a nonparallel imaging technique. FT = Fourier transformation.

View larger version (16K): [in this window] [in a new window] [Download PPT slide]   Figure 5.  Diagram of GRAPPA (k-space–based parallel imaging). At MR imaging with GRAPPA, only a fraction of the standard number of phase-encoding lines are acquired (black indicates acquired lines). Additional lines (medium gray) in the center of k-space are filled to maintain a high SNR. "Missing" k-space lines (light gray)—that is, lines that are not acquired—are calculated by using weighting factors derived from individual coil sensitivities. In GRAPPA, unlike older k-space–based parallel imaging techniques (eg, auto-SMASH), all k-space lines from each coil element are used to calculate the missing k-space lines. FT = Fourier transformation.

	Routine 3.0-T Abdominal Imaging Techniques

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

 
T1-weighted Sequences
The standard T1-weighted sequences used in abdominal imaging are gradient-echo in-phase and out-of-phase sequences and gadolinium-enhanced three-dimensional fat-suppressed multiphasic sequences. The longitudinal relaxation times (T1) of solid abdominal organs generally increase with increasing field strength. However, the increase is not uniform; for instance, the mean T1 of liver increases by 38% while that of the renal cortex increases by only 18% with an increase in field strength from 1.5 T to 3.0 T (16). The signal intensity in T1-weighted gradient-echo sequences is governed by the following simplified equation (2): SI sin (1 – e^–TR/T1), where SI is the signal intensity, is the flip angle, and e is the base of natural logarithms (approximately 2.718).

The prolongation of T1 at higher field strengths causes the signal intensity to decrease, an effect that may be lessened by using higher TR values. However, a longer TR may result in an increase in the acquisition time to more than 25 seconds, which is the approximate upper limit of a comfortable breath hold. Consequently, the quality of two-dimensional T1-weighted abdominal MR images obtained at 3.0 T is no better than and often is inferior to that of similar images obtained at 1.5 T. It may be possible to use parallel imaging to keep the acquisition time short enough to allow completion of the sequence during a single breath hold.

Three-dimensional gradient-echo sequences are used for MR angiography and gadolinium-enhanced imaging. Each echo in these sequences supplies data about a three-dimensional volume of tissue instead of a single two-dimensional section. Two phase-encoding gradients instead of one are applied before the readout gradient, to allow reconstruction into tomographic sections. These sequences differ from two-dimensional sequences in that they result in a higher SNR and allow thinner sections (2-mm contiguous sections are possible), smaller flip angles (typically 10°–15°), and more-homogeneous fat suppression (5,17). The effects of paramagnetic agents such as gadolinium are more pronounced with higher field strength, and this enables better contrast-enhanced MR imaging and MR angiography at 3.0 T than at 1.5 T (Fig 6).

View larger version (89K): [in this window] [in a new window] [Download PPT slide]   Figure 6a.  Comparison of contrast-enhanced MR angiographic images obtained at different magnetic field strengths after the administration of 30 mL of gadobenate dimeglumine (MultiHance; Bracco, Princeton, NJ). (a) Coronal image obtained at 1.5 T in a 57-year-old man with uncontrolled hypertension shows two right renal arteries and a high-grade focal stenosis at the ostium of the left main renal artery (arrowhead). (b) Oblique coronal view obtained at 3.0 T in a 44-year-old woman with a renal transplant (arrowhead) and abnormally high flow velocity in the transplant artery at Doppler ultrasonography clearly shows a normal origin (arrow) and course of the artery. Note that the smaller branches of the aorta and iliac arteries are more visible at 3.0 T because of the enhanced paramagnetic effect of gadolinium at the higher field strength.

View larger version (130K): [in this window] [in a new window] [Download PPT slide]   Figure 6b.  Comparison of contrast-enhanced MR angiographic images obtained at different magnetic field strengths after the administration of 30 mL of gadobenate dimeglumine (MultiHance; Bracco, Princeton, NJ). (a) Coronal image obtained at 1.5 T in a 57-year-old man with uncontrolled hypertension shows two right renal arteries and a high-grade focal stenosis at the ostium of the left main renal artery (arrowhead). (b) Oblique coronal view obtained at 3.0 T in a 44-year-old woman with a renal transplant (arrowhead) and abnormally high flow velocity in the transplant artery at Doppler ultrasonography clearly shows a normal origin (arrow) and course of the artery. Note that the smaller branches of the aorta and iliac arteries are more visible at 3.0 T because of the enhanced paramagnetic effect of gadolinium at the higher field strength.

The transverse relaxation time (T2) is relatively independent of field strength (16). However, susceptibility effects are increased at higher field strength, and these may be mitigated by using a shorter TE (<80 msec). This, in turn, may require a higher readout bandwidth, with a consequent reduction in SNR. Shorter echo trains are possible with parallel imaging. Other advantages of parallel imaging include reduced SAR effects and reduced image blurring at T2-weighted fast SE imaging (5). Hyperechoes or TRAPS also may be used to reduce SAR effects (see "Specific Absorption Rate").

The improved SNR at 3.0 T is evident in the depiction of small intrahepatic biliary ducts at three-dimensional MR cholangiopancreatography (Fig 7) (18). However, the benefits of 3.0-T imaging are lost in many cases. As discussed above, standing wave artifacts may severely degrade the quality of T2-weighted images obtained at 3.0 T in patients who are obese or who have ascites (Fig 8). Standing wave effects are more pronounced in pregnant women because of their large girth and the presence of amniotic fluid. In addition, there is increased concern for fetal safety at high field strengths; we do not perform fetal MR imaging at 3.0 T.

View larger version (138K): [in this window] [in a new window] [Download PPT slide]   Figure 7a.  Comparison of MR cholangiopancreatographic images obtained at different magnetic field strengths in the same patient after the administration of 16 µg of secretin (SecreFlo; Repligen, Waltham, Mass) for improved visualization of pancreatic ducts and 300 mL of oral ferumoxsil suspension (Gastromark; Mallinckrodt, Raleigh, NC) for reduction of signal from the stomach and duodenum. Image obtained at 1.5 T (3100/755; flip angle, 90°) (a) and image obtained 6 months later at 3.0 T (4090/493; flip angle, 90°) (b) provide comparable depiction of the common bile duct (white solid arrow) and pancreatic duct (black arrow), but b provides better depiction of the intrahepatic bile ducts (white arrowhead). Note the pancreas divisum, with the dorsal duct (black arrowhead) coursing in front of the distal common bile duct. The high signal intensity in the duodenum (dashed arrow) is due to the normal exocrine effect of secretin.

View larger version (144K): [in this window] [in a new window] [Download PPT slide]   Figure 7b.  Comparison of MR cholangiopancreatographic images obtained at different magnetic field strengths in the same patient after the administration of 16 µg of secretin (SecreFlo; Repligen, Waltham, Mass) for improved visualization of pancreatic ducts and 300 mL of oral ferumoxsil suspension (Gastromark; Mallinckrodt, Raleigh, NC) for reduction of signal from the stomach and duodenum. Image obtained at 1.5 T (3100/755; flip angle, 90°) (a) and image obtained 6 months later at 3.0 T (4090/493; flip angle, 90°) (b) provide comparable depiction of the common bile duct (white solid arrow) and pancreatic duct (black arrow), but b provides better depiction of the intrahepatic bile ducts (white arrowhead). Note the pancreas divisum, with the dorsal duct (black arrowhead) coursing in front of the distal common bile duct. The high signal intensity in the duodenum (dashed arrow) is due to the normal exocrine effect of secretin.

View larger version (148K): [in this window] [in a new window] [Download PPT slide]   Figure 8.  Standing waves artifact. Coronal T2-weighted abdominal MR image obtained with a half-Fourier RARE sequence (6031/138; flip angle, 115°) in a 45-year-old woman shows an absence of normal signal in the central abdominal region. This effect, caused by standing waves, renders the image diagnostically useless.

	Diffusion-weighted Imaging

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

View larger version (123K): [in this window] [in a new window] [Download PPT slide]   Figure 9a.  Comparison of diffusion-weighted abdominal MR images obtained at 1.5 T (1000/68; flip angle, 90°) (a, b) and 3.0 T (850/69; flip angle, 90°) (c, d) and with variations in b value from 0 sec/mm2 (a, c) to 500 sec/mm2 (b, d) shows a higher SNR and better signal preservation despite use of a high b value at 3.0 T than at 1.5 T. Images obtained with a high b value are weighted more heavily toward diffusion and less heavily toward T2 and pseudodiffusion (movement based on intravoxel capillary perfusion) than are images obtained with a lower b value. In addition, with a high b value, the signal hyperintensity of the spleen, kidneys, and gallbladder (a result of T2 shine-through) is less evident and the SNR overall is decreased.

View larger version (94K): [in this window] [in a new window] [Download PPT slide]   Figure 9b.  Comparison of diffusion-weighted abdominal MR images obtained at 1.5 T (1000/68; flip angle, 90°) (a, b) and 3.0 T (850/69; flip angle, 90°) (c, d) and with variations in b value from 0 sec/mm2 (a, c) to 500 sec/mm2 (b, d) shows a higher SNR and better signal preservation despite use of a high b value at 3.0 T than at 1.5 T. Images obtained with a high b value are weighted more heavily toward diffusion and less heavily toward T2 and pseudodiffusion (movement based on intravoxel capillary perfusion) than are images obtained with a lower b value. In addition, with a high b value, the signal hyperintensity of the spleen, kidneys, and gallbladder (a result of T2 shine-through) is less evident and the SNR overall is decreased.

View larger version (100K): [in this window] [in a new window] [Download PPT slide]   Figure 9c.  Comparison of diffusion-weighted abdominal MR images obtained at 1.5 T (1000/68; flip angle, 90°) (a, b) and 3.0 T (850/69; flip angle, 90°) (c, d) and with variations in b value from 0 sec/mm2 (a, c) to 500 sec/mm2 (b, d) shows a higher SNR and better signal preservation despite use of a high b value at 3.0 T than at 1.5 T. Images obtained with a high b value are weighted more heavily toward diffusion and less heavily toward T2 and pseudodiffusion (movement based on intravoxel capillary perfusion) than are images obtained with a lower b value. In addition, with a high b value, the signal hyperintensity of the spleen, kidneys, and gallbladder (a result of T2 shine-through) is less evident and the SNR overall is decreased.

View larger version (87K): [in this window] [in a new window] [Download PPT slide]   Figure 9d.  Comparison of diffusion-weighted abdominal MR images obtained at 1.5 T (1000/68; flip angle, 90°) (a, b) and 3.0 T (850/69; flip angle, 90°) (c, d) and with variations in b value from 0 sec/mm2 (a, c) to 500 sec/mm2 (b, d) shows a higher SNR and better signal preservation despite use of a high b value at 3.0 T than at 1.5 T. Images obtained with a high b value are weighted more heavily toward diffusion and less heavily toward T2 and pseudodiffusion (movement based on intravoxel capillary perfusion) than are images obtained with a lower b value. In addition, with a high b value, the signal hyperintensity of the spleen, kidneys, and gallbladder (a result of T2 shine-through) is less evident and the SNR overall is decreased.

	MR Spectroscopy

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

MR spectroscopy performed at 3.0 T is associated with a higher SNR compared with that at 1.5 T. In addition, the increased chemical shift at 3.0 T allows greater separation between adjacent spectral resonance peaks. However, two other factors also are important at in vivo MR spectroscopy. The intrinsic line width of the spectrum is dependent on transverse relaxation (T2), which is mostly independent of field strength. In contrast, intravoxel magnetic field inhomogeneities due to susceptibility effects are worse at high field strength and cause reduced signal and increased line width. As a result, at MR spectroscopy of immobile organs such as the brain, the advantage of 3.0 T over 1.5 T is marginal (1). At abdominal MR spectroscopy, in contrast, the faster acquisition offered by a field strength of 3.0 T used with a parallel imaging technique results in substantially improved spectral resolution (Fig 10) (20).

View larger version (37K): [in this window] [in a new window] [Download PPT slide]   Figure 10a.  Comparison of proton (1H) MR spectroscopic images in normal volunteers at 1.5 T (a) and 3.0 T (b) shows a strong water resonance peak in both a and b. In a, there is a single broad lipid resonance peak (arrow); in b, the taller resonance peak of methylene [(CH2)n] (arrow) is separated from that of terminal methyl [R-CH3] (arrow-head). The increased thickness of the spectral line in b is due to increased inhomogeneity of the magnetic field (B0) at 3.0 T. Note the absence of the N-acetylaspartate and choline peaks observed in brain MR spectra.

View larger version (40K): [in this window] [in a new window] [Download PPT slide]   Figure 10b.  Comparison of proton (1H) MR spectroscopic images in normal volunteers at 1.5 T (a) and 3.0 T (b) shows a strong water resonance peak in both a and b. In a, there is a single broad lipid resonance peak (arrow); in b, the taller resonance peak of methylene [(CH2)n] (arrow) is separated from that of terminal methyl [R-CH3] (arrow-head). The increased thickness of the spectral line in b is due to increased inhomogeneity of the magnetic field (B0) at 3.0 T. Note the absence of the N-acetylaspartate and choline peaks observed in brain MR spectra.

	Conclusions

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

	Footnotes

 

Abbreviations: FOV = field of view, GRAPPA = generalized auto-calibrating partially parallel acquisition, RARE = rapid acquisition with relaxation enhancement, SAR = specific absorption rate, SE = spin echo, SENSE = sensitivity encoding, SMASH = simultaneous acquisition of spatial harmonics, SNR = signal-to-noise ratio, TE = echo time, TR = repetition time, TRAPS = transition between pseudo steady states

See also the article by Barth et al (pp 1445–1462) and the commentary by Sher (pp 1462–1464) in this issue.

	References

Top Abstract Introduction Important Technical Parameters Parallel Imaging Techniques Routine 3.0-T Abdominal Imaging... Diffusion-weighted Imaging MR Spectroscopy Conclusions References

 

1. Schick F. Whole-body MRI at high field: technical limits and clinical potential. Eur Radiol 2005;15: 946–959.[CrossRef][Medline]
2. Edelstein WA, Glover GH, Hardy CJ, Redington RW. The intrinsic signal-to-noise ratio in NMR imaging. Magn Reson Med 1986;3:604–618.[Medline]
3. Merkle EM, Dale BM, Paulson EK. Abdominal MR imaging at 3T. Magn Reson Imaging Clin N Am 2006;14:17–26.[CrossRef][Medline]
4. Frayne R, Goodyear BG, Dickhoff P, Lauzon ML, Sevick RJ. Magnetic resonance imaging at 3.0 Tesla: challenges and advantages in clinical neurological imaging. Invest Radiol 2003;38:385–402.[CrossRef][Medline]
5. Hussain SM, Wielopolski PA, Martin DR. Abdominal magnetic resonance imaging at 3.0 T: problem or a promise for the future? Top Magn Reson Imaging 2005;16:325–335.[CrossRef][Medline]
6. Merkle EM, Dale BM. Abdominal MRI at 3.0 T: the basics revisited. AJR Am J Roentgenol 2006; 186:1524–1532.[Abstract/Free Full Text]
7. Hennig J, Scheffler K. Hyperechoes. Magn Reson Med 2001;46:6–12.[CrossRef][Medline]
8. Weigel M, Hennig J. Contrast behavior and relaxation effects of conventional and hyperecho-turbo spin echo sequences at 1.5 and 3 T. Magn Reson Med 2006;55:826–835.[CrossRef][Medline]
9. Frank LR, Wong EC, Liu TT, Buxton RB. Increased diffusion sensitivity with hyperechos. Magn Reson Med 2003;49:1098–1105.[CrossRef][Medline]
10. Hennig J, Weigel M, Scheffler K. Multiecho sequences with variable refocusing flip angles: optimization of signal behavior using smooth transitions between pseudo steady states (TRAPS). Magn Reson Med 2003;49:527–535.[CrossRef][Medline]
11. Pruessmann KP. Encoding and reconstruction in parallel MRI. NMR Biomed 2006;19:288–299.[CrossRef][Medline]
12. Glockner JF, Hu HH, Stanley DW, Angelos L, King K. Parallel MR imaging: a user’s guide. RadioGraphics 2005;25:1279–1297.[Abstract/Free Full Text]
13. Bammer R, Schoenberg SO. Current concepts and advances in clinical parallel magnetic resonance imaging. Top Magn Reson Imaging 2004; 15:129–158.[CrossRef][Medline]
14. Pruessmann KP. Parallel imaging at high field strength: synergies and joint potential. Top Magn Reson Imaging 2004;15:237–244.[CrossRef][Medline]
15. Blaimer M, Breuer F, Mueller M, Heidemann RM, Griswold MA, Jakob PM. SMASH, SENSE, PILS, GRAPPA: how to choose the optimal method. Top Magn Reson Imaging 2004;15:223–236.[CrossRef][Medline]
16. 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]
17. Martin DR, Friel HT, Danrad R, De Becker J, Hussain SM. Approach to abdominal imaging at 1.5 Tesla and optimization at 3 Tesla. Magn Reson Imaging Clin N Am 2005;13:241–254, v–vi.[CrossRef][Medline]
18. Merkle EM, Haugan PA, Thomas J, Jaffe TA, Gullotto C. 3.0- versus 1.5-T MR cholangiography: a pilot study. AJR Am J Roentgenol 2006; 186:516–521.[Abstract/Free Full Text]
19. Naganawa S, Kawai H, Fukatsu H, et al. Diffusion-weighted imaging of the liver: technical challenges and prospects for the future. Magn Reson Med Sci 2005;4:175–186.[CrossRef][Medline]
20. Katz-Brull R, Rofsky NM, Lenkinski RE. Breath-hold abdominal and thoracic proton MR spectroscopy at 3T. Magn Reson Med 2003;50:461–467.[CrossRef][Medline]

This article has been cited by other articles:

				J.-Y. Choi, M.-J. Kim, Y. E. Chung, J. Y. Kim, A. C Jones, J. de Becker, and M. van Cauteren Abdominal Applications of 3.0-T MR Imaging: Comparative Review versus a 1.5-T System RadioGraphics, July 1, 2008; 28(4): e30 - e30. [Abstract] [Full Text] [PDF]

				S. I. Sher Invited Commentary RadioGraphics, September 1, 2007; 27(5): 1462 - 1464. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Akisik, F. M. Articles by Lall, C.