| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 Department of Radiology, Saint Barnabas Medical Center, 94 Old Short Hills
Rd., Livingston, NJ 07039.
2 Clinical Science, Philips Medical Systems, Cleveland, OH 44143.
Received January 9, 2004;
accepted after revision May 24, 2004.
Address correspondence to P. D. Friedman.
Introduction
 Top Introduction Materials and MethodsResultsDiscussionReferences |
|---|
The SENSE technique is readily used in multiple MRI applications, such as body imaging, pediatric imaging, angiography, dedicated abdominal studies, orthopedics, and cardiac imaging. One of the few areas of MRI in which parallel imaging is not being used sufficiently is breast MRI. Research about the use of SENSE imaging for breast examinations is evolving, and there are expectations within the radiology community that the SENSE technique will have a major impact on diagnostic sensitivity [1]. These expectations are mainly because of the advantages that SENSE imaging presents in the form of both higher spatial resolution for morphology and higher temporal resolution for dynamic contrast enhancement. These features better show findings for diagnosis than non-SENSE-imaging techniques within an acceptable scanning time.
We describe the use of SENSE imaging for breast MRI studies in a community-based radiology practice that performs approximately 125 breast MRI examinations per month. To our knowledge, this is the largest ongoing application of SENSE imaging for breast examinations. Approximately 1,300 patients at our institution have undergone scanning with this technique since January 2002. Our main focus in this article is to describe our initial and ongoing experiences with the SENSE technique in imaging the breast.
Breast MRI with the SENSE technique results in images with higher temporal and spatial resolution than images obtained without the SENSE technique. High spatial resolution is essential for better visualization of morphology and the characterization of lesions. High temporal resolution is essential for better contrast uptake statistics and consequently for the characterization of lesions [1]. We have been able to use shorter scanning times that, in turn, have improved patient comfort, generated increased patient throughput, and resulted in lower costs. We illustrate how the SENSE technique can aid in improving MRI evaluation of the breast.
|
TopIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Non-fat-saturated T1-weighted and fat-saturated T2-weighted images were acquired before the administration of contrast medium. All the images were acquired in a conventional bilateral fashion in a transverse orientation. Both the T1- and T2-weighted protocols were based on turbo spin-echo protocols with an in-plane resolution of 1.0 x 1.2 mm, 30 slices of 5-mm thickness, and scanning times of 51 and 42 sec, respectively.
The fat saturation of the T2-weighted images was achieved with binomial spatial–spectral pulses of the combination 1331. Axial single-shot diffusion imaging also was performed; two b values were used (0, 800) with an in-plane resolution of 2.2 x 2.8 mm with 20 slices of 5-mm thickness. In all three sequences, SENSE was used either to reduce scanning time for turbo spin-echo imaging 50% or to reduce the Echo Planar Imaging (EPI) readout duration for diffusion imaging.
The 3D dynamic contrast enhancement acquisition was performed in three different orders—namely, high temporal and low spatial resolution, moderate temporal and spatial resolution, and low temporal and high spatial resolution. A single dose of 30 mL of IV contrast material (gadodiamide, Omniscan, Amersham Health) was administered. The patients were split into three groups to follow one of three bilateral dynamic imaging protocols. However, in all three of the protocols, the dynamic contrast enhancement scanning time was maintained so that it did not exceed 5 min. It is standard procedure in breast MRI to acquire dynamic information for approximately 4–5 min [4].
The protocol details, which are listed in Table 1, show the scanning parameters used for each of the three protocols. Sequence 1 shows high temporal and lower spatial resolution with no fat saturation; sequence 2 shows higher spatial and lower temporal resolution with fat saturation; and finally, sequence 3 shows an optimal spatial and temporal resolution. The non-fat-saturation protocol had limitations mainly because subtractions must be performed. The higher temporal resolution indicated for sequence 1 had to compromise on lower spatial resolution, which limited the detection of submillimeter lesions. The higher spatial resolution with lower temporal resolution (sequence 3) presented limits on the contrast uptake curves. The compromise between spatial and temporal resolution with the inclusion of fat saturation to avoid subtractions is preferred because it provides sufficient time for uptake analysis and an ability to detect submillimeter lesions.
|
After dynamic imaging of every patient, a 3D fast-field echo sequence with fat suppression was performed. A fat-selective binomial spatial and spectral saturation pulse (Proset 1331) was incorporated for fat suppression with an in-plane resolution of 0.85 mm2 and 175 slices of 0.75-mm thickness. The scanning time was 4 min. A SENSE factor of 3 (2 along phase [P] and 1.5 along slice [S] direction) was used also. Maximum intensity projections were generated from this sequence in addition to sagittal multiplanar reformations.
|
TopIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Figure 1A, 1B shows an example of our initial dynamic protocol without SENSE imaging. Only unilateral imaging was available, which resulted in longer examination times and increased contrast costs. In addition, there is a noticeable difference in the resolution of these examinations in comparison with our current examinations.
|
|
Figures 2A and 2B show representative images of the breast obtained with the SENSE technique from the dynamic contrast enhancement protocol that we currently use. This protocol results in images with an optimal blend of spatial and temporal resolution.
|
|
The postisotropic high-resolution acquisition yields a 1-mm3 voxel that is useful for creating multiplanar reformations in either the sagittal or coronal plane without losing resolution and without spending extra time to reacquire any data in those planes. Figures 2C, 2D, 2E show a representative high-resolution axial image, multiplanar reformation image, and maximum-intensity-projection image.
|
|
|
|
TopIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Faster imaging techniques that yield the high temporal resolution necessary for effective contrast-uptake analysis for lesion characterization need to be combined with high spatial resolution. SENSE imaging provides both the necessary high spatial and temporal resolution. In doing so, it addresses a fundamental dilemma confronting radiologists who interpret breast MRI as to which lesions should be biopsied and which should be followed up. The major advantage of SENSE breast MRI is in the elimination of dynamic subtractions, which are prone to error caused by motion, thus making breast MR interpretations unreliable. The SENSE technique, combined with 3D dynamic contrast enhancement and fat saturation, eliminates the necessity for any subtractions. Because of a higher SENSE factor (2 along the phase and 1.5 along slice direction), we were able to keep a high temporal resolution of less than 1 min and also maintain a high spatial resolution.
Before SENSE imaging, our initial non-SENSE unilateral imaging examination took approximately 30 min, required two patient visits, two contrast injections, and subtraction images. Temporal resolution was greater than 1 min. We briefly explored bilateral non-SENSE imaging, which took approximately 45 min to achieve the spatial and temporal resolution similar to SENSE imaging. With SENSE imaging, the increased spatial resolution enabled us to better identify the borders of lesions and helped us better characterize lesions. This increased ability significantly reduced the uncertainty that sometimes accompanies decisions based on conventional breast MRI regarding which lesions should be biopsied and which should be followed up. Higher SENSE factors also resulted in reduced artifact from breathing and cardiac motion. Because of the reduction in motion artifact, the axilla was better visualized.
In conclusion, SENSE imaging has resulted in a faster and more robust protocol that yields the required spatial and temporal resolution for better lesion delineation. The increased speed in data acquisition is reflected in a higher patient throughput that has produced monetary gains for the MRI facility. SENSE imaging also generates true bilateral imaging by using the same single bolus of contrast material for both breasts, resulting in increased patient comfort. Bilateral imaging is helpful for evaluating symmetry, restricting the examination to a single patient visit, and enabling a single radiologist evaluating both breasts to issue a more concise report.
The SENSE technique is worth trying; it does not require the use of a physicist or extra technologist training. Reproducible results with 22 min to scan both breasts are attainable. Time–intensity curves and 3D reformatted images can be performed on a workstation (View Forum Work Station, Philips Medical Systems) or at the MR console. We have found that use of the SENSE technique has resulted in fewer call-backs for cardiac motion artifact and incomplete visualization of the axilla.
Breast MRI generally is considered an expensive examination technique with a relatively low specificity [2]. However, with continued improvements in technique aimed at increasing patient throughput, reducing costs, and improving diagnostic specificity, breast MRI will become even more beneficial and, possibly, become an effective screening tool.
|
TopIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. J. Kim, E.-K. Kim, J. Y. Kwak, B.-W. Park, S.-I. Kim, and K. K. Oh Bilateral Synchronous Breast Cancer in an Asian Population: Mammographic and Sonographic Characteristics, Detection Methods, and Staging Am. J. Roentgenol., January 1, 2008; 190(1): 208 - 213. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. D. Friedman, S. V. Swaminathan, K. Herman, and L. Kalisher Breast MRI: the importance of bilateral imaging. Am. J. Roentgenol., August 1, 2006; 187(2): 345 - 349. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
