| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Medicine (Cardiovascular Division) and Radiology, Beth Israel
Deaconess Medical Center, Harvard Medical School, Boston, MA.
2 Present address: Department of Clinical Radiology, University of Münster,
Albert-Schweitzer-Strasse 33, Münster 48129, Germany.
3 Philips Medical Systems, Best, The Netherlands.
Received May 5, 2003;
accepted after revision August 15, 2003.
Address correspondence to D. Maintz
(maintz{at}uni-muenster.de).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Free-breathing coronary MR angiography was performed in 12 patients using four imaging sequences (turbo field-echo, fast spin-echo, balanced fast field-echo, and spiral turbo field-echo). Quantitative comparisons, including signal-to-noise ratio, contrast-to-noise ratio, vessel diameter, and vessel sharpness, were performed using a semiautomated analysis tool. Accuracy for detection of hemodynamically significant disease (> 50%) was assessed in comparison with radiographic coronary angiography.
RESULTS. Signal-to-noise and contrast-to-noise ratios were markedly increased using the spiral (25.7 ± 5.7 and 15.2 ± 3.9) and balanced fast field-echo (23.5 ± 11.7 and 14.4 ± 8.1) sequences compared with the turbo field-echo (12.5 ± 2.7 and 8.3 ± 2.6) sequence (p < 0.05). Vessel diameter was smaller with the spiral sequence (2.6 ± 0.5 mm) than with the other techniques (turbo field-echo, 3.0 ± 0.5 mm, p = 0.6; balanced fast field-echo, 3.1 ± 0.5 mm, p < 0.01; fast spin-echo, 3.1 ± 0.5 mm, p < 0.01). Vessel sharpness was highest with the balanced fast field-echo sequence (61.6% ± 8.5% compared with turbo field-echo, 44.0% ± 6.6%; spiral, 44.7% ± 6.5%; fast spin-echo, 18.4% ± 6.7%; p < 0.001). The overall accuracies of the sequences were similar (range, 74% for turbo field-echo, 79% for spiral). Scanning time for the fast spin-echo sequences was longest (10.5 ± 0.6 min), and for the spiral acquisitions was shortest (5.2 ± 0.3 min).
CONCLUSION. Advantages in signal-to-noise and contrast-to-noise ratios, vessel sharpness, and the qualitative results appear to favor spiral and balanced fast field-echo coronary MR angiography sequences, although subjective accuracy for the detection of coronary artery disease was similar to that of other sequences.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Favorable multicenter data using 3D free-breathing coronary MR angiography have recently been shown for detection of proximal and mid coronary artery disease [2]. This technique uses navigator gating and motion correction, a T2 prepulse and a frequency-selective fat-suppression prepulse for contrast enhancement, and a 3D turbo field-echo sequence for data acquisition [3, 4]. Alternative free-breathing coronary MR angiography sequences have been introduced. Black-blood, fast spin-echo coronary MR angiography facilitates exclusive visualization of the coronary lumen, reduced susceptibility to turbulent flow in regions of focal stenosis, a reduced sensitivity to arti-facts induced by metallic implants, and a high contrast-to-noise ratio [5, 6].
Steady-state free precession coronary MR angiography, also known as "true FISP" (Siemens, Erlangen, Germany), "balanced fast field-echo" (Philips Medical Systems, Best, The Netherlands), and "fast imaging employing steady-state acquisition" (General Electric Medical Systems, Milwaukee, WI), enables endogenous contrast enhancement without specific magnetization preparation prepulses or exogenous enhancement [7, 8]. Improvements in the signal-to-noise and contrast-to-noise ratios using steady-state free precession instead of turbo field-echo have been described [9].
Spiral coronary MR angiography offers more efficient k-space sampling, thereby facilitating an increase in the signal-to-noise ratio and reduced scanning times [10, 11].
Although all four approaches provide potential advantages, no systematic comparison of these sequences in a common patient population has been described. Therefore, we undertook this prospective study of patients recently referred for coronary radiographic angiography to compare these sequences. Signal-to-noise ratio, contrast-to-noise ratio, vessel diameter, and vessel sharpness were compared using a comprehensive coronary vessel analysis tool [12]. Additionally, images were compared with radiographic coronary angiography for the assessment of their value for detection of clinically significant coronary artery disease.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Protocol
All scanning was performed on a commercial 1.5-T Gyroscan ACS-NT whole-body
system (Philips Medical Systems) equipped with cardiac software (CPR6,
Philips) and a commercial PowerTrak 6000 gradient system (Philips) (23 mT/m,
220-msec rise time).
For signal reception, a commercial five-element (two anterior, three posterior) cardiac synergy coil was used, with all five elements being applied for data acquisition. Subjects were examined with ECG leads placed on the anterior left hemithorax (vector ECG). All scanning was performed during free breathing. For coronary artery localization and for navigator positioning at the right hemidiaphragm, two scout scans were obtained. The first scout image was an ECG-triggered, free-breathing, multislice 2D segmented gradient-echo (TR/TE, 11/2.4; matrix, 256 x 128; field of view, 450 mm; slice thickness, 10 mm; interslice gap, 5 mm) scan with nine transverse, nine coronal, and nine sagittal interleaved acquisitions of the thorax. The total duration of this scout acquisition was less than 1 min. On these images, the navigator position at the dome of the right hemidiaphragm and the localized transverse 3D volume for the subsequent scout scan were planned. The transverse 3D volume of the second scout image was positioned to include the origins of both the left and the right coronary arteries as visualized on the first scout image. For the second scout image, the MR data were acquired in mid-diastole and at end-expiration using right hemidiaphragm navigator gating and real-time motion correction. This second scout scan included a T2 prepulse and a fat saturation prepulse for contrast enhancement and a segmented k-space echoplanar imaging sequence for signal readout. One shot included five radiofrequency excitations, followed by 11 readouts each. One shot was applied per cardiac cycle, resulting in an acquisition window of 97 msec (8.8/5.3). This 3D slab covered a 90-mm-thick volume with an in-plane resolution of 1.3 x 1.8 mm. Thirty overlapping slices with a reconstructed slice thickness of 3 mm were acquired.
Imaging Sequences
From the second scout data set, double oblique imaging planes were defined
for high-resolution coronary imaging with four data acquisition sequences
using a previously described three-point planning tool
[4]. For the left coronary
system, a transverse volume including the left main, the left anterior
descending, and the left circumflex coronary arteries was located on the
second scout scan. For the suppression of bulk cardiac motion related to
diaphragm displacement, real-time navigator gating and real-time motion
correction were used with a 5-mm gating window and a constant z-axis
correction factor of 0.6 [3,
5,
10].
Table 1 summarizes the most
important parameters of the four sequences. To avoid bias caused by fatigue
effects, we acquired sequences in random order.
|
Three-dimensional turbo field-echo sequences.— A 3D segmented k-space turbo field-echo sequence (2.4/7.2) used a flow insensitive T2 prepulse (50 msec) for contrast enhancement [3, 13], followed by a coronal anterior chest saturation prepulse, the navigator, a spectrally selective fat saturation pulse, and, finally, the turbo field-echo imaging sequence. The receiver bandwidth was 135 Hz/pixel. To compensate for suboptimal slice profiles associated with 3D slab excitations, imaging of a 39-mm-thick volume was performed with subsequent oversampling (in the slice selection direction) by a factor of 1.3. The 20 center slices with a slice thickness of 1.5 mm each were subsequently reconstructed, and the six most peripheral slices were rejected during reconstruction. A field of view of 360 mm and a 512 x 360 matrix yielded an in-plane resolution of 0.7 x 1.0 mm. Sixteen phase-encoding steps were sampled during each cardiac cycle (acquisition window, 115 msec). The k-space was sampled using centric ordering with priority for the low ky profiles (cartesian acquisition). One signal average was performed and no flow-compensating gradients were used. Data acquisition was performed in mid-diastole.
Two-dimensional fast spin-echo sequences.— Black-blood coronary MR angiography was performed (3-mm slice thickness, 1.5-mm overlap) using a 2D fast spin-echo sequence with a linear k-space acquisition scheme, 5.2-msec interecho spacing, 25-msec TE, echo-train length of 23, and TR of two cardiac cycles [6]. An initial 90° radiofrequency excitation was followed by repetitive 160° refocusing pulses. Half-Fourier sampling was used, with 60% of the profiles sampled in the phase-encoding direction. The acquisition window was 120 msec. The field of view was 360 mm with a 512 x 384 scan matrix (spatial resolution, 0.7 x 0.9 mm). No flow-compensating gradients or fat suppression prepulses were used. For enhanced black-blood characteristics, a dual-inversion prepulse [14] was applied immediately after the detection of the R-wave of the ECG. The first nonselective 180° inversion prepulse was followed by a 7-mm slice-selective 180° inversion prepulse at the anatomic level of interest. After the dual-inversion prepulse, a NAV-RESTORE prepulse [6] was used to allow navigator lung–liver interface detection in the presence of a nonselective inversion preceding the navigator excitation. The fast spin-echo imaging sequence followed the first inversion pulse with an inversion delay that was adjusted to null the signal of the inflowing blood at the time of the initial 90° pulse of the fast spin-echo sequence.
Three-dimensional balanced fast field-echo sequences.—These
sequences were a segmented 3D balanced fast field-echo sequence with a centric
ordered k-space acquisition scheme and a TR/TE of 4/2
[8]. Fourteen radiofrequency
excitations with a 75° constant excitation angle were applied during each
R-R interval, resulting in an acquisition window of 56 msec. The field of view
was 300 mm with a 256 scan matrix (in-plane resolution of 1.2 x 1.2 mm).
Twelve slices were acquired covering a volume 36 mm thick (effective slice
thickness, 3 mm). To account for suboptimal slice profiles associated with 3D
volume acquisitions, we performed oversampling by a factor of 1.3 in the slice
selection direction, with the most peripheral slices being rejected during
reconstruction. Twenty-four slices with a slice thickness of 1.5 mm each were
reconstructed using zero filling in the kz direction. The imaging
part of the sequence was performed in mid-diastole with a
heart-rate–dependent trigger delay. The use of 20 preparatory pulses
with a TR of 4 msec and a radiofrequency excitation angle of 75°
immediately preceding the imaging sequence yielded a good contrast between the
blood pool and the surrounding tissue, while minimizing flow-related artifacts
and the delay between navigator and imaging. During the dummy cycles, no
signal sampling was performed, but the frequency of the radiofrequency pulses
was adapted according to the navigator-detected position of the
lung–liver interface. For the suppression of chest wall motion
artifacts, a local saturation band with an excitation angle of 130° was
applied before the preparatory pulses and after the navigator pulse. No
additional prepulses or signal averaging was applied. However, a previously
described 
/2 prepulse was used before the 20 preparatory pulses to
obtain steady-state conditions and to suppress signal from fat
[15].
Three-dimensional spiral sequences.—The modular structure of this sequence was according to the turbo field-echo sequence [11]. A square field of view of 350 mm and a 512 x 512 matrix yielded an in-plane spatial resolution of 0.7 x 0.7 mm2 (slice thickness of 3.0 mm, 1.5 mm after zero filling). Spiral 3D data sets were acquired using the stack-of-spirals approach [16], in which conventional phase-encoding is used for spatial encoding in the slice-selection direction. Analogous to the cartesian acquisition, 10 phase-encoding steps were applied in the outer loop of the experiment to cover a slab of 30 mm. Each 2D sub–k-space was traversed using a variable angular speed spiral with a TR/effective TE of 26/1.5 msec and a sampling window of 20 msec. The readout bandwidth was set to 140 kHz. The prepared longitudinal magnetization was read out using a 90° excitation pulse as suggested by Thedens et al. [17]. In our study, two consecutive spiral interleaves (with no intervening prepulses or navigator) were acquired during each R-R interval to increase sampling efficiency [11]. A total of 42 interleaves were acquired using ramped radiofrequency excitation angles of 45° and 90° per R-R interval. Thus, for both interleaves, the same transverse magnetization was obtained. The starting angle between the subsequent interleaves (spiral trajectories) was 180°. The total cardiac acquisition window was 52 msec.
Quantitative Analysis of MR Images
For quantitative analysis and comparison of the different MR techniques, an
objective analysis tool was used
[12]. This tool allows
reformatting of the original MR data set and visualization of the coronary
anatomy in one image as well as semiautomatic determination of signal-to-noise
ratio, contrast-to-noise ratio, vessel length, vessel diameter, and vessel
sharpness. For visualization of the coronary tree, the user interactively
specifies a curved subvolume (enclosed in the 3D coronary data set) that
closely encompasses the coronary artery segments. Three-dimensional Delaunay
triangulation [18] and
parallel projection enable the simultaneous display of multiple coronary
segments in one 2D representation. To calculate vessel diameters and
sharpness, the user must define the segment of interest by marking multiple
points on the course of the vessel. The software then calculates the sharpness
of the vessel using a Deriche algorithm
[19]. This algorithm
calculates an edge image using a first-order derivative of the input image.
The local value in a Deriche image represents the magnitude of local change in
signal intensity (i.e., vessel sharpness). A vessel sharpness of 100% refers
to a maximum signal-intensity change at the vessel border. A low value
indicates inferior vessel sharpness, and vice versa.
Analysis of signal-to-noise and contrast-to-noise ratios was selectively
performed on a user-specified slice of the original 3D data. After we defined
regions of interest in the ascending aorta (SI[signal
intensity]blood), anterior to the chest wall
(SIair), and in the muscle of the left ventricular free
wall (SImuscle), the program calculated signal-to-noise
ratio and contrast-to-noise ratio according to the following
equations:
The scanning time and efficiency were recorded in all individuals. Scanning time included the time from initiation of the coronary MR angiography sequence until completion of scanning. Scan efficiency was defined as the number of R-R intervals accepted for reconstruction (navigator position in the gating window) divided by the total number of heartbeats required to complete a scan. For example, a scan efficiency of 50% would represent a scan in which data acquired from 50% of the heartbeats are accepted for reconstruction.
Qualitative Analysis of MR Images
Source coronary MR angiograms were evaluated by consensus of two
experienced observers who were unaware of the patients' clinical findings. The
original source images were analyzed by scrolling through individual slices
from the 3D data set with the use of a commercial software package (EasyVision
4.0, Philips Medical Systems). Sequences were interpreted randomly. Seven
coronary artery segments were evaluated: the left main coronary artery and the
proximal and mid segments of the left anterior descending coronary artery
(0–2 cm, 2–4 cm), the left circumflex coronary artery (0–1.5
cm, 1.5–3 cm), and the right coronary artery (0–2 cm, 2–5
cm). For each segment, image quality was visually graded
[20] as 1, indicating poor or
uninterpretable (coronary artery visible with markedly blurred borders or
edges); 2, good (coronary artery visible with moderately blurred borders or
edges); 3, very good (coronary artery visible with mildly blurred borders or
edges); or 4, excellent (coronary artery visible with sharply defined borders
or edges). If the segments were not imaged or if the image quality was graded
poor or uninterpretable (grade 1), no further evaluation was performed. Images
of good, very good, and excellent quality (grades 2–4) were further
classified according to the visual assessment of the coronary artery lumen as
having no coronary artery disease, minimal disease (stenosis < 50%), or
clinically significant disease (stenosis > 50%)
[2].
Radiographic Angiography
Conventional radiographic coronary angiography was performed in multiple
projections using standard techniques. Assessment of each coronary vessel and
visual estimation of the maximal percentage of reduction of the luminal
diameter for each lesion were performed by an experienced cardiologist who was
unaware of all clinical and coronary MR angiography data.
Statistics
All data are presented as mean ± SD. Comparisons among the sequences
with regard to quantitative results were made by a one-way analysis of
variance with pairwise comparison using the Scheffé test ( <
0.05, two-tailed). The overall diagnostic accuracy for the detection of
significant coronary artery disease was determined for coronary MR angiography
as compared with radiographic angiography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In total, images of 204 segments (51 segments with four sequences) were available for comparative analysis. Of these, 24 segments belonged to the right coronary system (3 patients x 2 segments x 4 sequences) and 180 segments belonged to the left coronary system (9 patients x 5 segments x 4 sequences). Twenty-four segments (12%) could not be assessed on MR angiography because they were not entirely included in the scan volume, and 12 segments (6%) were of poor quality (grade 1). Of the segments having poor image quality, four were of the mid left anterior descending coronary artery segment, three of the proximal left circumflex coronary artery segment, and seven of the mid left circumflex coronary artery segment. Thus, 167 segments (82%) were assessed on MR angiography; the proportions of segments for which images could be assessed ranged from 47% (for the mid left circumflex coronary artery segment) to 100% (for the left main coronary artery).
Table 2 summarizes the results of the comparison of the four sequences with regard to subjective image quality, scanning times, signal-to-noise ratio, and contrast-to-noise ratio. Overall image quality was 2.7 ± 0.6, with no significant differences among the four sequences. Examples of left coronary artery MR angiograms having comparable image quality across the four sequences are shown in Figure 1A, 1B, 1C, 1D.
|
|
|
|
|
The spiral sequence and the balanced fast field-echo sequence showed an approximate twofold improvement in signal-to-noise ratio when compared with the cartesian turbo field-echo sequence (p < 0.001) and an almost two-fold improvement in contrast-to-noise ratio (p < 0.01 and p < 0.05). Differences in signal-to-noise ratio and contrast-to-noise ratio between the spiral and the balanced fast field-echo sequence were not significant. As a result of black-blood properties of the fast spin-echo sequence, no signal-to-noise ratio was calculated for this sequence; and the contrast-to-noise ratio was significantly lower than for the spiral and balanced fast field-echo sequences (p < 0.05). The difference in contrast-to-noise ratio between the turbo field-echo and the fast spin-echo sequences was not significant (p = 0.99).
The ratio of signal to noise per voxel size per scanning duration was significantly higher for the spiral than for the turbo field-echo and balanced fast field-echo sequences (p < 0.001 for both). Differences in this ratio between the turbo field-echo and balanced fast field-echo sequence were not significant.
The average scanning time for each sequence ranged from 5 min 10 sec
(± 18 sec) for the spiral sequence to 10 min 28 sec (± 35 sec)
for the fast spin-echo sequence. The average navigator efficiency was similar
(
50%) for all four sequences.
Data for vessel diameter and vessel sharpness are summarized in Table 3. Vessel diameter for spiral MR angiography sequences was significantly smaller than for fast spin-echo (p < 0.01) and balanced fast field-echo (p = 0.01) sequences. Smaller average diameters for the spiral MR angiography sequence compared with the turbo field-echo sequence were not significant (p = 0.6). Also, no significant differences in the overall average vessel diameter were recorded among the fast spin-echo, balanced fast field-echo, and turbo field-echo sequences.
|
Vessel sharpness was significantly greater in the images acquired with the balanced fast field-echo sequence than in the images acquired with the other techniques (p < 0.001). No significant differences were observed between turbo field-echo and spiral sequences (p = 0.98).
Data regarding detection of coronary artery disease are displayed in Table 4. Most (all but one) patients with coronary artery disease had only moderate disease (50–70% diameter stenosis). The overall prevalence of disease in the study population was 27%. In the comparison with conventional radiographic angiography, the accuracy for detection of disease using MR angiography was similar for all approaches and ranged from 74% (turbo field-echo) to 79% (spiral). Figure 2A, 2B, 2C, 2D, 2E, 2F shows significant (< 50%) stenosis of the proximal left anterior descending coronary artery as revealed on all four sequences.
|
|
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Quantitative Evaluation
Navigator performance and scanning time.—As expected, no
significant differences were seen in navigator efficiency for the four
sequences. The spiral sequence showed the most favorable average scanning
time.
Signal-to-noise ratio and contrast-to-noise ratio.—In our objective comparison, we found an approximate twofold increase in the signal-to-noise ratio and an almost twofold increase in the contrast-to-noise ratio for the balanced fast field-echo and the spiral sequences when compared with the turbo field-echo sequence. The fast spin-echo sequence showed a similar contrast-to-noise ratio when compared with the turbo field-echo sequence.
This finding is in good agreement with the findings of Bornert et al. [11], who found an 80% increase in signal-to-noise ratio and a 60% increase in contrast-to-noise ratio with the double interleaf spiral sequence.
When comparing the balanced fast field-echo and turbo field-echo sequences, the difference in spatial resolution must be considered.
When divided by the voxel size and scan duration, the signal-to-noise ratio was not significantly different for the turbo field-echo and balanced fast field-echo sequences. When comparing the balanced fast field-echo and the spiral sequences, the scanning time must also be considered. Although it had a similar signal-to-noise ratio, the spiral sequence had three times greater in-plane spatial resolution and a scanning time shorter by a factor of 1.3, which resulted in a significantly better signal-to-noise per voxel size per scanning time ratio.
Quantitative comparisons in signal-to-noise ratio and contrast-to-noise ratio are difficult because of the variability of voxel size and bandwidth among the sequences. However, making all relevant parameters equal for all sequences would result in compromised image quality. For example, lowering the bandwidth of balanced fast field-echo to that of turbo field-echo would result in a substantial increase in TE and TR, which, in turn, would yield amplified motion artifacts and thus compromised image quality. Similarly, improving the spatial resolution in balanced fast field-echo sequences also results in an increase in TE and TR, with the same consequences.
Vessel diameter and sharpness.—The smaller vessel diameter found in the spiral sequence when compared with the cartesian turbo field-echo sequence is in contrast to the findings of Bornert et al. [11], who found similar average diameters. One potential explanation includes the prolonged acquisition window of the turbo field-echo sequence in our study (56 msec for spiral vs 115 msec for turbo field-echo). With prolonged acquisition windows, more blurring may be expected because of residual cardiac motion, resulting in larger apparent diameters.
The balanced fast field-echo sequence showed the best vessel sharpness, which may be attributed to the significantly lower spatial resolution and water–fat shift artifacts and opposed-phase artifacts at the borders of the vessels. The resulting signal-attenuated rim parallel to the vessels has previously been described as an advantage for vessel definition [8]. However, the impact of these artifacts on quantification of stenoses remains to be investigated in a larger clinical trial. The use of a fat saturation prepulse could reduce these artifacts. Another improvement of the balanced fast field-echo sequence would be the introduction of a T2 prepulse for contrast enhancement [7, 21].
Qualitative Evaluation
For qualitative assessment of the different MR techniques, a consensus
interpretation was performed, and image quality and the accuracy for the
detection of significant coronary artery disease were determined in comparison
with the gold standard, radiographic angiography. Image quality ranged from
good to very good and did not show significant differences among the
sequences.
The prevalence of significant coronary artery disease in our study was low (24–27%), a bias related to the recruitment of stable patients who had undergone diagnostic radiographic angiography but no percutaneous intervention. Thus, the examined patient collective was most demanding for an MR angiography assessment, because all stenoses but one were only moderate (50–70%) in severity. Therefore, our patient collective may not represent the typical patient who undergoes cardiac catheterization. As a consequence, further improvements in accuracy may be expected for all sequences in a more balanced patient collective.
The turbo field-echo sequence is currently the only sequence that has previously been evaluated in a larger patient group [2]. Its overall accuracy in detecting or excluding coronary artery disease was similar in that study to our findings.
Limitations of our study are that sublingual isosorbide dinitrite was not administered before MR angiography and the lack of quantitative data for radiographic angiography. We specifically chose not to administer isosorbide dinitrite because of the concern that its effects would not be uniform throughout acquisition of all four sequences. In addition, interpretation of the sequence accuracy data is limited because the data are based on a small population with a relatively low prevalence of disease.
The favorable results obtained for the spiral sequence in comparison with the cartesian turbo field-echo sequence have several possible explanations. Beside the higher signal-to-noise ratio and contrast-to-noise ratio and the increased vessel sharpness, spiral imaging is less sensitive to flow and motion artifacts because of its inherent first gradient moment nulling [22]. Although an enhanced signal-to-noise ratio was found for the balanced fast field-echo sequence, this enhanced signal may not easily be traded for enhanced spatial resolution because doing so would result in prolonged TR and TE and therefore more flow artifacts on the balanced fast field-echo images.
Conclusion
Excellent visualization of the proximal coronary arterial system was
obtained with all four free-breathing coronary MR angiography techniques. The
balanced fast field-echo and spiral sequences seem favorable, having an almost
twofold improvement in signal-to-noise ratio and contrast-to-noise ratio and
significantly improved vessel sharpness. An advantage for the spiral sequence
with respect to overall scanning time was also identified.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Kefer, E. Coche, G. Legros, A. Pasquet, C. Grandin, B. E. Van Beers, J.-L. Vanoverschelde, and B. L. Gerber Head-to-Head Comparison of Three-Dimensional Navigator-Gated Magnetic Resonance Imaging and 16-Slice Computed Tomography to Detect Coronary Artery Stenosis in Patients J. Am. Coll. Cardiol., July 5, 2005; 46(1): 92 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ozgun, A. Hoffmeier, M. Kouwenhoven, R. M. Botnar, M. Stuber, H. H. Scheld, W. J. Manning, W. Heindel, and D. Maintz Comparison of 3D Segmented Gradient-Echo and Steady-State Free Precession Coronary MRI Sequences in Patients with Coronary Artery Disease Am. J. Roentgenol., July 1, 2005; 185(1): 103 - 109. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
