Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Apr 1.
Published in final edited form as: Magn Reson Med. 2013 Jul 2;71(4):1511–1517. doi: 10.1002/mrm.24803

Homogenous Fat Suppression for Bilateral Breast Imaging Using Independent Shims

Misung Han 1,2,5,⋆, Charles H Cunningham 3,4, John M Pauly 2, Bruce L Daniel 1, Brian A Hargreaves 1
PMCID: PMC4085331  NIHMSID: NIHMS597950  PMID: 23821305

Abstract

Purpose

To demonstrate the capability of incorporating independent shims into a dual-band spectral-spatial excitation and to compare fat suppression between standard global shims and independent shims for in vivo bilateral breast imaging at 1.5T.

Methods

A dual-band spectral-spatial excitation pulse was designed by interleaving two flyback spectral-spatial pulses, playing one during positive gradient lobes and the other during negative gradient lobes. Each slab was enabled to have an independent spatial offset, spectral offset, and slab-phase modulation by modulating RF phase, and independent linear shims were incorporated by playing extra shim gradients. Phantom experiments were performed to demonstrate the functionality of the pulse, and in vivo experiments were performed for ten healthy volunteers to compare fat suppression between standard shims and independent shims.

Results

The phantom experiments confirmed that the dual-band pulse can provide independent spectral and spatial offsets and linear shims to the two slabs. Independent shims provided qualitatively more homogeneous fat suppression than standard shims in seven out of ten subjects, with equivalent fat suppression in two of the other three subjects.

Conclusion

Incorporating independent shims into the dual-band spectral-spatial excitation can provide homogeneous fat suppression in bilateral breast imaging.

Keywords: dual-band spectral-spatial excitation, independent shims, bilateral breast MRI, fat suppression

Introduction

Breast MRI is highly sensitive in detecting breast abnormalities and staging breast cancers. Recently, it has emerged as a screening tool supplementing conventional X-ray mammography [1]. Several studies have shown that MRI has much higher sensitivity in detecting cancer than X-ray mammography, especially for high-risk patients [2,3]. Central to breast MRI exams is T1-weighted dynamic contrast-enhanced MRI, which requires high temporal and spatial resolution and homogenous fat suppression. High temporal resolution provides accurate characterization of signal enhancement patterns, while high spatial resolution offers detailed visualization of lesion morphology. Fat suppression improves identification of breast tumors because unsuppressed fat can generate signal intensity similar to that of enhanced tumors [4].

Various techniques have been used to provide fat suppression in dynamic contrast-enhanced breast MRI. Subtraction of pre-contrast images from post-contrast images [5, 6] does not increase acquisition time, but can generate misregistration artifacts due to patient motion between frames. Chemically selective fat saturation [7, 8] is the most commonly used because it is simple to apply and the scan time increase is minimal with intermittent application of the saturation pulse. An alternative method is to use a water-only excitation that selectively excites only water by using a spatially and spectrally selective excitation pulse [9]; this technique is less sensitive to static (B0) field and radio frequency (B1) field inhomogeneities at a cost of an echo-time increase. Multi-point Dixon techniques are fat-water separation methods that exploit multiple echo images [10–12], and are more resilient to B0 field inhomogeneities. However, they may require longer acquisition times to image at multiple echo times, and they are often sensitive to motion and other displacement artifacts.

Homogeneous fat suppression is challenging for bilateral breast imaging due to significant field inhomogeneities over the two breasts. The hemispherical shape of the breast and presence of the heart and lungs close to the breast generate oddly shaped interfaces between air and tissue, resulting in susceptibility-induced B0 inhomogeneities. These field inhomogeneities within one breast are primarily linear in the anterior-to-posterior direction, and first-order shimming can reduce the field inhomogeneities substantially [13, 14]. However, optimal first-order shim fields can differ between the two breasts due to anatomical asymmetries, and thus standard global first-order shimming may not reduce the field inhomogeneities sufficiently for bilateral breast imaging. Previously, it was shown that using a separate center frequency for each breast can provide a more homogeneous B0 field over the two breasts [15].

Recently, a dual-band spectral-spatial excitation pulse was proposed as an effective excitation method for bilateral breast imaging, having a capability to provide independent center frequency and linear shims to each breast [16–18]. This dual-band pulse interleaves two flyback spectral-spatial excitation pulses [19] on different slab-select gradient polarities, thus two slabs can be excited independently without perturbing magnetization from the other slab. Independent slab-phase modulation can also be incorporated, which allows for simultaneous imaging of both slabs without the need to image the non-excited space between the two slabs [20]. This reduces the scan time and can improve image quality when parallel imaging is applied [21].

In this work, we compare fat suppression for bilateral breast imaging with and without using independent shims for dual-band spectral-spatial excitation. Our hypothesis is that applying independent shims can simultaneously provide homogeneous fat suppression to both breasts, equivalent to the case where each breast were separately shimmed and imaged. We first demonstrate the functionality of the dual-band spectral-spatial pulse with phantom experiments. Then we apply the pulse to in vivo bilateral breast MRI to compare fat suppression with standard shims and independent shims at 1.5T.

Methods

Dual-Band Spectral-Spatial Excitation

RF Pulse Design and Spectral-Spatial Profile

A dual-band spectral-spatial excitation pulse was designed for slab excitation and fat suppression at 1.5T, using gradient systems with a maximum strength of 40 mT/m and a maximum slew rate of 150 mT/m/ms (Fig. 1a). The slab-select gradient was consisted of trapezoidal lobes of alternating polarity, with a period of 2 ms to locate the first sidelobes 500 Hz away from the main lobe. The rise and decay times of each trapezoidal lobe were 0.1 ms. The two flyback spectral-spatial excitation pulses were played on different gradient polarities, having equivalent spectral RF envelopes and spatial RF subpulses. Each RF subpulse was played over the entire trapezoid lobe including ramp times, and had a time-bandwidth product of 14.7. For each slab, the spatial offset was given by applying linear phase on the corresponding RF subpulses (quadratic phase during the gradient ramp times), which should be rewound before the next RF subpulse not to affect the spectral profile. The spectral offset was given by applying linear phase on the overall RF envelope.

Figure 1.

Figure 1

Open in a new tab

Dual-band spectral-spatial excitation with independent spatial and spectral offsets. (a) The dual-band spectral-spatial RF pulse is designed by interleaving two flyback spectral-spatial RF pulses to be played on different gradient polarities. The waveforms here are to provide a flip angle of 20°, a thickness of 10 cm for each slab, and spatial offsets of 15 cm and −5 cm. Θspatial(t) is an RF phase waveform to provide spatial offsets, where the average slope of the two different ramps during either positive or negative constant slab-select gradient is associated with the separation of the two slabs, while the difference slope between the two ramps is associated with the shift of the combination of the two slabs to off-center. (b) shows the theoretical spectral-spatial magnetization profile using the waveforms in (a). (c) Θspectral(t) is a phase waveform to provide independent spectral offsets of ±100 Hz, and (d) shows the theoretical magnetization profile by incorporating this phase waveform in addition to Θspatial(t). The spectral and spatial offset waveforms may be more easily understood by noting that they apply linear phase in excitation kz − kf space.

The waveforms shown in Fig. 1a are to provide a flip angle of 20°, a thickness of 10 cm for each slab, and spatial offsets of 15 cm and −5 cm. The theoretical spectral-spatial magnetization profile using these waveforms is shown in Fig. 1b. Figure 1d is the theoretical profile with having spectral offsets of ±100 Hz, with incorporating an additional RF phase Θspectral(t) in Fig. 1c. The spatial profile was sharp, with having an 18% transition width. The spectral profile had a 158 Hz full width at half maximum (FWHM).

Independent Linear Shims

Different linear shims can be applied to each slab if an alternating shim gradient amplitude can be played during excitation. If we use the built-in scanner shim channel to provide the average of the desired shim gradients, then we only need to apply the differential shim using the additional gradient waveform. Gshim(t) in Fig. 2 is a designed shim gradient waveform to provide a shim gradient of Gδ/2 to one slab and −Gδ/2 to the other slab. The blips were played during the gradient ramps to ensure that the k-space trajectory of the shim gradientent, kshim(t), is linearly increasing during each slab excitation as if one shim value were constantly applied. In particular, the shim gradient waveform was designed to make the two dashed lines (representing linear k-space trajectories experienced by each slab) intersect at the center of the RF pulse instead of at the start or end of the RF pulse to minimize the required maximum blip area.

Figure 2.

Figure 2

Open in a new tab

Independent shims. Gshim(t) is a shim gradient waveform to provide a shim gradient of Gδ/2 for one slab and −Gδ/2 for the other slab. This shim waveform is played on each of the x, y, and z axes, and the z shim waveform is added to the slab-select gradient. The pre-encoding and post-encoding blips during the gradient ramp times ensure that the k-space trajectory of the shim gradient, kshim(t), increases linearly for each slab excitation. The peak blip amplitude can be minimized when two dashed lines, representing kshim(t) variation on each slab excitation, intersect at the center of the RF pulse. Note that the first blip changes its amplitude from ±Gδ/2 to 51Gδ in 0.1 ms (ramp time of one slab-select gradient lobe). When Gδ/2 is 146 uT/m, the peak blip amplitude is 14.9 mT/m, and the slope of this blip reaches the slew-rate limit (150 mT/m/ms).

By setting the independent spectral offset and playing the shim gradient waveform along the x, y, and z axes, an independent first-order shim gradient field can be generated for each slab. To play the blips with meeting a slew-rate limit of 150 mT/m/ms, the shim gradient (Gδ/2) could be applied up to ±146 uT/m along the x and y axis. Along the z axis, this shim gradient was more restricted as the slab-select gradient needed to combine. For example, when the thickness for each slab was 10 cm, up to ±113 uT/m of the shim gradient could be applied along the z axis. The gradient amplitude limit (40 mT/m) did not restrict the shim gradient.

Independent Slab-Phase Modulation

For simultaneous imaging of the two slabs that are separated in the slab direction, 3D phase-encoding must be performed over the two excited slabs and the unexcited space between them to avoid aliasing artifacts. However, independent slab-phase modulation of the two slabs can shift the slabs close together as if they were contiguous, eliminating the need to encode the unexcited space [20]. To achieve this, the phase of each flyback pulse should independently increase with the slab-phase encoding number.

Implemenation

To play the desired RF phase on the scanner, we predefined phase waveforms to provide spectral offsets, spatial offsets, and phase modulation in the pulse sequence program. Their amplitudes were then calculated based on the prescription by the host computer, and the final RF phase waveform was determined by summing the necessary phase waveforms. When independent slab-phase modulation was used, the RF phase waveform was redefined for each sequence repetition. The shim gradient waveform in Fig. 2 was also predefined in the program, and its amplitude was determined after the user prescribing the desired shim values for each slab.

Phantom Validation

We validated the dual-band spectral-spatial RF pulse by performing phantom experiments using a GE Signa Excite 1.5T scanner (GE Healthcare, Waukesha, WI). The RF pulse was incorporated into a standard 2D gradient-echo sequence, and the slab profiles were directly imaged by applying the phase-encode gradient on the same axis as the slab-select gradient. The capability of providing an independent spatial offset and linear shims to each slab was tested by using a slab phantom, and the capability of providing an independent spectral offset was tested by using a bottle phantom containing both tap water and peanut oil. For sequence parameters, a 25 ms echo time, a 100 ms repetition time, a thickness of 6.3 cm for each slab, and a 30 × 30 cm2 field-of-view (FOV) for the slab phantom and 15 × 15 cm2 FOV for the bottle phantom were used.

In Vivo Experiments

We scanned ten healthy volunteers using an eight-channel phased-array breast coil (GE Healthcare, Waukesha, WI). Informed consent was obtained from all subjects, and all examinations were conducted with the approval of the Institutional Review Board. Three-dimensional sagittal bilateral imaging was performed using the dual-band water-selective excitation combined with RF spoiling and flyback echo-planar imaging [22]. Other sequence parameters included a 8.5 ms echo time, a 40 ms repetition time, a 20° flip angle, a 20 × 20 cm2 FOV with 256 × 192 matrix (in-plane), four echo train length, half-Fourier, 64 sagittal slices per breast (total 128 slices), and a 1.5–2 mm slice thickness. Independent slab-phase modulation [20] was incorporated into the dual-band excitation to avoid encoding the empty space between the breasts.

For the dual-slab excitation, the spatial offsets were obtained during the graphical prescription. The optimal center frequency and linear shim values were measured by the scanner prescan calibration after locating a shim volume with a size of 12 × 12× 12 cm3 on each slab. Based on these optimal values, four different scans were conducted by varying the shim values as follows: (1) applying the left breast shim values to both breasts, (2) applying the right breast shim values to both breasts, (3) applying the average shim values to both breasts (standard shims), and (4) applying the left breast shim values to the left breast and right breast shim values to the right breast (independent shims). The four sets of images were compared, assessing whether the independently-shimmed excitation provided better, worse or equivalent fat suppression than the other combinations.

Results

Phantom Validation

The images from the phantom experiments are presented in Fig. 3, where (a–c) used a slab phantom and (d–f) used a water/fat bottle phantom. Figure 3a–b demonstrates the ability to offer an arbitrary spatial offset to each slab. The two slabs are located equally separated from z = 0 in Fig. 3a while one slab is located farther away from the center in Fig. 3b. The capability of applying independent shims is demonstrated in Fig. 3c. Here, the shim gradient waveform was played along the x axis with prescribing an average x shim of 20.3 Hz/cm, providing a 33.2 Hz/cm shim to slab 1 and a 7.3 Hz/cm shim to slab 2. The linear variation of the spin resonant frequencies along the x axis (996 Hz/FOV for slab 1 and 219 Hz/FOV for slab 2) results in light to dark transition of signal, which reflects the spectral profile of the spectral-spatial pulse. Figure 3d–f demonstrates the property of providing independent spectral offsets. In Fig. 3d the center frequencies of the two slabs were set only to excite water. In Fig. 3e–f, the independent center frequencies were set to excite water for one slab and fat for the other slab.

Figure 3.

Figure 3

Open in a new tab

Slab profiles using the dual-band spectral-spatial pulse. (a–c) are acquired with an uniform slab phantom, while (d–f) are acquired with a water/fat bottle phantom. (a–b) demonstrate the ability to excite two slabs in arbitrary locations. In (c), a shim gradient is applied along the x axis, generating different signal profiles between the two slabs. (d–f) demonstrate the ability to provide an independent center frequency to each slab. In (d), only water is excited in both slab, while in (e–f), water is excited for one slab and fat is excited for the other slab. Note the signal reduction at the edge of slab 2 (denoted by an arrow in (f)) due to magnetic field inhomogeneities caused by susceptibility differences at the air/phantom boundary.

In Vivo Experiments

The optimal center frequencies and optimal shim gradients measured for both slabs in all ten subjects by the scanner prescan calibration routine are listed in Tab. 1. All values varied between the two breasts and over different subjects. The optimal shim gradient for the anterior to posterior axis was higher than those for the other axes, indicating that the most significant linear field variation is in the anterior to posterior direction as expected.

Table 1.

Optimal shim gradients from the left and right breast slabs for the ten subjects

Subject Center
Frequency
Differencea 		Linear Shim gradienta Fat Suppressionc
Left Slab Right Slab Difference
xb yb zb x y z x y z
1 31 1.53 38.15 9.16 −10.68 42.73 0 12.21 −4.58 9.16 Unchanged
2 15 −3.05 21.36 10.68 −9.16 32.05 6.10 6.10 −10.68 4.58 Improved
3 39 3.05 32.05 10.68 −10.68 48.83 10.68 13.73 −16.79 0 Improved
4 27 −12.21 32.05 1.53 −4.58 38.15 −13.73 −7.63 −6.10 15.26 Improved
5 52 −16.79 28.99 −10.68 1.53 28.99 −10.68 −18.31 0 0 Improved
6 35 3.05 38.15 9.16 −12.21 50.36 7.63 15.26 −12.21 1.53 Unchanged
7 −20 6.10 42.73 6.10 −9.16 47.30 0 15.26 −4.578 6.10 Improved
8 9 6.10 21.36 1.53 −12.21 25.94 6.10 18.31 −4.58 −4.58 Improved
9 32 −4.58 35.10 −9.16 −7.63 36.62 −4.58 3.05 −1.53 −4.58 Improved
10 46 −7.63 42.73 −21.36 −10.68 38.15 −12.21 3.05 4.58 −9.16 Worse
Open in a new tab
a

The center frequency difference between the left slab and the right slab is shown in Hz, and the linear shim gradients along the three axes for the two slabs and the differences between them are shown in uT/m.

b

The x, y, and z axes are in the right/left direction, anterior/posterior direction, and superior/inferior direction, respectively.

c

Here we denote whether fat suppression is visually improved, unchanged, or worse by using independent shims than using standard shims.

Figure 4 shows breast images from one subject (Subject 8). Figure 4a–d shows axially reformatted images close to the inferior edges of the breasts using the four different shim methods. When the optimal left shim values or the optimal right shim values were used for both breasts (a,b), the ipsilateral breast achieved excellent fat suppression while the contralateral breast had regions with unsuppressed fat shown by arrows. When average shim values were used for both breasts (c), failure of fat suppression was still seen in both breasts. With independent shims (d), the fat suppression in each breast was as good as when the optimal shim values for that breast were used. Figure 4e–h shows sagittal slices from the right and left breasts from the same subject. When standard shims were used, fat suppression failed in the inferior edges of the breasts (arrows), which often possess significant field inhomogeneities; however, when the independent shims were used, improved fat suppression was observed in those regions.

Figure 4.

Figure 4

Open in a new tab

Fat-suppressed breast images from one subject. Axially reformatted images from the left breast shims (a), right breast shims (b), standard shims (c), and independent shims (d) are shown. The regions of fat suppression failure are denoted by arrows. (e–h) Sagittal images from the right and left breasts (corresponding to the dashed lines in (a)) demonstrate that independent shims achieve better fat suppression in the inferior edges of the breasts (arrows) than standard shims. The dashed line in (e) corresponds to the position of the axial images (a–d).

Independent shims provided fat suppression over the two breasts as if each breast were separately scanned with its optimal shim gradient field. In seven out of the ten subjects, fat suppression was qualitatively better with independent shims than with standard shims. For two other subjects, fat suppression was similar between both shim methods because there were no visible artifacts in the images acquired with standard shims. For the other subject, independent shims generated more artifacts, possibly due to suboptimal shim numbers from the improper prescription of the shim volumes, or suboptimal shim settings that may have resulted from the automated shimming algorithm itself.

Discussion

For bilateral breast MRI, uniform fat suppression over the two breasts is difficult due to large field inhomogeneities. Most scanners provide automatic first-order shimming to reduce field inhomogeneities, or scanner operators can manually adjust linear shim fields by changing shim values through the user interface. However, optimal linear shim values for the right and left breasts can be different. The dual-band spectral-spatial excitation allows the slabs to have independent first-order shims. With independent shims, better fat suppression can be achieved with a reduction of B0-induced artifacts such as unsuppressed fat or suppressed water. This can provide better delineation of breast tissue and enhanced tumors for contrast-enhanced MRI.

The independent shims could be applied in various schemes with our dual-band pulse, but setting the average shim gradients in the hardware and only playing the differential using a predesigned shim gradient waveform can be the most efficient approach in terms of sequence programming. The shim gradient waveform should be carefully designed; the amplitude should alternate between the two RF subpulses but this alternation should not provide any adverse effects on offering each slab a desirable shim field. The blips during the slab-select gradient ramps can allow for each slab to experience an unvarying shim field as if a constant gradient were applied. Ideally, no RF should be played when the blips are applied not to degrade a spectral-spatial magnetization profile. With our RF subpulses, which were played over the gradient ramps, the distortion of the profile was minimal based on the simulation, though. Eddy currents will be induced with this rapidly changing shim gradient waveform, but alternating the polarity of the gradient may cause some cancellation of eddy current effects, and actually reduce their overall effect. Because the slopes of the blips increase with the shim gradient, the shim gradient is restricted by system slew-rate limits. The proposed shim gradient waveform is designed to minimize the peak blip amplitude for a given shim gradient, which will maximize the shim gradient applicable. With our design, a shim gradient over ±100 uT/m could be applied, which will be probably sufficient for 1.5T breast MRI based on our measurements.

The dual-band spectral-spatial pulse can be also designed for 3T breast imaging. The total RF pulse duration can be shorter at 3T due to the doubling of the frequency difference between water and fat and restricting the subpulse length.

The spectral-spatial excitation could also be used as a 90° fat saturation pulse by exciting fat instead of water and then dephasing the fat [23]. In this case, fat saturation can be performed independently over the two breasts, providing improved robustness to both B0 and B1 field inhomogeneities. Independent shims could be applied (as discussed in this work) to reduce B0 field inhomogeneities, while B1 variations could be mitigated by applying a different RF amplitude for each breast to provide effectively similar flip angles. This is especially useful at 3T where B1 fields can vary by up to a factor of two between the left and the right breasts [24, 25].

Conclusions

Homogeneous fat suppression in bilateral breast imaging is challenging due to large B0 field inhomogeneities over the two breasts. A dual-band spectral-spatial pulse can excite two slabs in arbitrary locations and provide independent first-order shims to each slab. By applying independent shims, B0 field inhomogeneities over the two breasts can be further reduced than standard shims, and more homogeneous fat suppression can be achieved for bilateral imaging.

Acknowledgement

The authors thank Sandra Rodriguez and Ann Sawyer for their assistance with in vivo experiments and Dr. Pauline Worters for her assistance with manuscript preparation. This work was supported by NIH 2P41RR009784-11, NIH 1R01EB009055, the Richard M. Lucas Foundation, and GE Healthcare.

References

  • 1.Saslow D, Boetes C, Burke W, Harms S, Leach MO, Lehman CD, Morris E, Pisano E, Schnall M, Sener S, Smith RA, Warner E, Yaffe M, Andrews KS, Russell CA. American cancer society guidelines for breast screening with MRI as an adjunct to mammography. CA Cancer J Clin. 2007;57:75–89. doi: 10.3322/canjclin.57.2.75. [DOI] [PubMed] [Google Scholar]
  • 2.Kriege M, Brekelmans CT, Boetes C, Besnard PE, Zonderland HM, Obdeijn IM, Manoliu RA, Kok T, Peterse H, Tilanus-Linthorst MM, Muller SH, Meijer S, Oosterwijk JC, Beex LV, Tollenaar RA, de Koning HJ, Rutgers EJ, Klijn JG. Efficacy of MRI and mammography for breast-cancer screening in women with a familial or genetic predisposition. N Engl J Med. 2004;351:427–437. doi: 10.1056/NEJMoa031759. [DOI] [PubMed] [Google Scholar]
  • 3.Leach MO, Boggis CR, Dixon AK, et al. Screening with magnetic resonance imaging and mammography of a UK population at high familial risk of breast cancer: a prospective multicentre cohort study (MARIBS) Lancet. 2005;365:1769–1778. doi: 10.1016/S0140-6736(05)66481-1. [DOI] [PubMed] [Google Scholar]
  • 4.Kerslake RW, Carleton PJ, Fox JN, Imrie MJ, Cook AM, Read JR, Bowsley SJ, Buckley DL, Horsman A. Dynamic gradient-echo and fat-suppressed spin-echo contrast-enhanced MRI of the breast. Clin Radiol. 1995;50:440–454. doi: 10.1016/s0009-9260(05)83159-9. [DOI] [PubMed] [Google Scholar]
  • 5.Boetes C, Barentsz JO, Mus RD, Van Der Sluis RF, van Erning LJ, Hendriks JH, Holland R, Ruys SH. MR characterization of suspicious breast lesions with a gadolinium-enhanced Turbo-FLASH subtraction technique. Radiology. 1994;193:777–781. doi: 10.1148/radiology.193.3.7972823. [DOI] [PubMed] [Google Scholar]
  • 6.Flanagan FL, Murray JG, Gilligan P, Stack JP, Ennis JT. Digital subtraction in Gd-DTPA enhanced imaging of the breast. Clin Radiol. 1995;50:848–854. doi: 10.1016/s0009-9260(05)83106-x. [DOI] [PubMed] [Google Scholar]
  • 7.Haase A, Frahm J, Hänicke W, Matthaei D. 1H NMR chemical shift selective (CHESS) imaging. Phys Med Biol. 1985;30:341–344. doi: 10.1088/0031-9155/30/4/008. [DOI] [PubMed] [Google Scholar]
  • 8.Frahm J, Haase A, Hänicke W, Matthaei D, Bomsdorf H, Helzel T. Chemical shift selective MR imaging using a whole-body magnet. Radiology. 1985;156:441–444. doi: 10.1148/radiology.156.2.4011907. [DOI] [PubMed] [Google Scholar]
  • 9.Meyer CH, Pauly JM, Macovskiand A, Nishimura DG. Simultaneous spatial and spectral selective excitation. Magn Reson Med. 1990;15:287–304. doi: 10.1002/mrm.1910150211. [DOI] [PubMed] [Google Scholar]
  • 10.Dixon WT. Simple proton spectroscopic imaging. Radiology. 1984;153:189–194. doi: 10.1148/radiology.153.1.6089263. [DOI] [PubMed] [Google Scholar]
  • 11.Glover GH. Multipoint Dixon technique for water and fat proton and susceptibility imaging. J Magn Reson Imaging. 1991;1:521–530. doi: 10.1002/jmri.1880010504. [DOI] [PubMed] [Google Scholar]
  • 12.Ma J. Breath-hold water and fat imaging using a dual-echo two-point dixon technique with an efficient and robust phase-correction algorithm. Magn Reson Med. 2004;52:415–419. doi: 10.1002/mrm.20146. [DOI] [PubMed] [Google Scholar]
  • 13.Maril N, Collins CM, Greenman RL, Lenkinski RE. Strategies for shimming the breast. Magn Reson Med. 2005;54:1139–1145. doi: 10.1002/mrm.20679. [DOI] [PubMed] [Google Scholar]
  • 14.Lee SK, Hancu I. Population Variabity of Susceptiblity-Induced B0 Field in Bilateral Breast MRI. Proceedings of the 20th Annual Meeting of ISMRM; Melbourne, Australia. 2012. p. 1475. [Google Scholar]
  • 15.Kim DH, Spielman DM, Daniel B. Automated Bilateral Shimming for Breast MRI. Proceedings of the 10th Annual Meeting of ISMRM; Honolulu, Hawaii, USA. 2002. p. 636. [Google Scholar]
  • 16.Pauly JM, Cunningham CH, Daniel BL. Independent dual-band spectral-spatial pulses. Proceedings of the 11th Annual Meeting of ISMRM; Toronto, Ontario, Canada. 2003. p. 966. [Google Scholar]
  • 17.Cunningham CH, Daniel BL, Rakow-Penner R, Pauly JM. Independent dual-band spectral-spatial pulses: implementation for bilateral breast MRI. Proceedings of the 13th Annual Meeting of ISMRM; Miami Beach, Florida, USA. 2005. p. 1849. [Google Scholar]
  • 18.Han M, Rodriguez S, Sawyer AM, Daniel BL, Cunningham C, Pauly JM, Hargreaves BA. Fat suppression with independent shims for bilateral breast MRI. Proceedings of the 17th Annual Meeting of ISMRM; Honolulu, Hawaii, USA. 2009. p. 580. [Google Scholar]
  • 19.Fredrickson JO, Meyer CH, Pelc NJ. Flow effects of spectral spatial excitation. Proceedings of the 5th Annual Meeting of ISMRM; Vancouver, British Vancouver, British Columbia, Canada. 1997. p. 113. [Google Scholar]
  • 20.Hargreaves BA, Cunningham CH, Pauly JM, Daniel BL. Independent phase modulation for efficient dual-band 3D imaging. Magn Reson Med. 2007;57:798–802. doi: 10.1002/mrm.21180. [DOI] [PubMed] [Google Scholar]
  • 21.Han M, Beatty PJ, Daniel BL, Hargreaves BA. Independent slab-phase modulation combined with parallel imaging in bilateral breast MRI. Magn Reson Med. 2009;62:1221–1231. doi: 10.1002/mrm.22115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Feinberg DA, Turner R, Jakab PD, Von Kienlin M. Echo-planar imaging with asymmetric gradient modulation and inner-volume excitation. Magn Reson Med. 1990;13:162–169. doi: 10.1002/mrm.1910130116. [DOI] [PubMed] [Google Scholar]
  • 23.Zur Y. Design of improved spectral-spatial pulses for routine clinical use. Magn Reson Med. 2000;43:410–420. doi: 10.1002/(sici)1522-2594(200003)43:3<410::aid-mrm13>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
  • 24.Kuhl CK, Kooijman H, Gieseke J, Schild HH. Effect of B1 inhomogeneity on breast MR imaging at 3.0T. Radiology. 2007;244:929–930. doi: 10.1148/radiol.2443070266. [DOI] [PubMed] [Google Scholar]
  • 25.Azlan CA, Di Giovanni P, Ahearn TS, Semple SI, Gilbert FJ, Redpath TW. B1 transmission-field inhomogeneity and enhancement ratio errors in dynamic contrast-enhanced MRI (DCE-MRI) of the breast at 3T. J Magn Reson Imaging. 2010;31:234–239. doi: 10.1002/jmri.22018. [DOI] [PubMed] [Google Scholar]

RESOURCES