BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
First published online August 6, 2007
British Journal of Radiology (2007) 80, 668-673
© 2007 British Institute of Radiology
doi: 10.1259/bjr/88996134
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lutterbey, G
Right arrow Articles by Gieseke, J
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lutterbey, G
Right arrow Articles by Gieseke, J

Full paper

Clinical evaluation of a speed optimized T2 weighted fast spin echo sequence at 3.0 T using variable flip angle refocusing, half-Fourier acquisition and parallel imaging

G Lutterbey, MD 1 M P Wattjes, MD 1 J Kandyba, MD 1 M Harzheim, MD 2 M V Falkenhausen, MD 1 N Morakkabati, MD 1 H Schild, MD 1 and J Gieseke, PhD 3

1 Department of Radiology, 2 Department of Neurology, University of Bonn, Sigmund-Freud-Strasse 25, 53105 Bonn, Germany, 3 Philips Medical Systems, Best, the Netherlands

Correspondence: Dr Goetz Lutterbey, Radiologische Universitätsklinik, Universität Bonn, Sigmund-Freud-Strasse 25, D-53105 Bonn, Germany. E-mail: goetz.lutterbey{at}ukb.uni-bonn.de


   Abstract
Top
 Abstract
Introduction
 Patients and methods
 Results
 Discussion
 References
 
This paper aims to demonstrate the capabilities of a speed optimized T2 weighted single-shot turbo spin echo sequence, using parallel imaging, variable flip angle refocusing and half-Fourier acquisition (FAS-TSE), in comparison with a standard TSE (sTSE) sequence in patients with suspected multiple sclerosis (MS). 33 patients presenting with a clinically isolated syndrome (CIS) suggestive of MS were prospectively examined on a 3.0 T MR system using FAS-TSE and a sTSE sequence. The FAS-TSE (scan time 11 s) and the sTSE (scan time 122 s) were compared regarding lesion detectability, lesion contrast, grey/white matter contrast, overall image quality and artefacts. Scanning parameters affecting image contrast and spatial resolution were kept identical. 208 lesions were detected using the sTSE sequence compared with 183 lesions (88%) using the FAS-TSE. The FAS-TSE was rated inferior regarding lesion contrast. The mean value/range/standard deviation of the lesion/white matter contrast were 0.26/0.06–0.49/0.089, respectively, with the sTSE vs 0.21/0.04–0.40/0.081 with the FAS-TSE. The FAS-TSE was rated inferior regarding overall image quality, but superior regarding motion artefacts. The grey/white matter contrast was qualitatively judged as comparable for both sequences. FAS-TSE provides sufficient T2-SE contrast and diagnostic image quality for whole brain studies in 11 s. It is suited to reduce motion artefacts in restless patients and for fast acquisition of additional scanning planes.


   Introduction
Top
 Abstract
 Introduction
Patients and methods
 Results
 Discussion
 References
 
Routine MRI is still a time-consuming imaging technique. Economic considerations as well as anxious, painful and restless patients make high-speed MRI desirable.

Currently, a temporal resolution of one image per second is achievable using fast gradient echo techniques (GRE) such as echo planar imaging (EPI). But at high field strengths such as 3.0 T, the fast T2* weighted (T2W) GRE are strongly influenced by susceptibility artefacts, especially if high spatial resolution is required. Moreover, many different types of GRE sequences are provided by the manufacturers with different appearances of image contrast.

T2W spin echo (SE) sequences have a more standardized image appearance and are less sensitive to susceptibility effects, but need a relatively long scan time compared with GRE. Thus, fast SE techniques (e.g. turbo spin echo, TSE) are widely used to accelerate T2W imaging. Generally, TSE techniques require high radiofrequency (RF) power, because many refocusing pulses are necessary to build the echo train.

Doubling the field strength (e.g. from 1.5 to 3.0 T) results in a fourfold increase in RF power; consequently, the specific absorption rate (SAR) limits are exceeded earlier in the highfield environment. Therefore, the signal gain from highfield MRI cannot be easily transferred into faster TSE sequences, because the repetition time (TR) and the echo spacing cannot be chosen as short as technically possible [1–3].

Several techniques provide a reduction in RF deposition and scanning time, including the parallel imaging technique (e.g. SENSE = SENSitivity encoding [4]), half-Fourier acquisition and also variable refocusing techniques such as hyperecho [5–7]. If used in combination, these techniques will substantially reduce the signal-to-noise ratio (SNR). Therefore, 3.0 T imaging combined with the techniques mentioned above will demonstrate synergistic effects, because the high magnetic field might compensate for the loss of SNR and, simultaneously, these techniques reduce the RF power needed for spatial encoding.

The aim of this study is to compare the diagnostic value of a single-shot T2W-TSE sequence including a stepwise reduced refocusing flip angle (in the following text: flip angle sweep = FAS), SENSE and half-Fourier acquisition with a standard T2W-TSE (sTSE) sequence in patients with suspected multiple sclerosis (MS). We chose this study population as a model because the expected abundance of small white matter lesions will help to discover the limits of the sequence. The sequence parameters we used were chosen for maximum effect on scan time to demonstrate the capability of this technique.


   Patients and methods
Top
 Abstract
 Introduction
 Patients and methods
Results
 Discussion
 References
 
Patients presenting with a clinically isolated syndrome (CIS) were prospectively included in the study. The inclusion criteria were:

  1. Clinical presentation of a first CIS suggestive of MS as defined by the International Panel on MS diagnosis [7].
  2. Time between onset of CIS and the MRI examination within 1 month.
  3. No prior vascular, immunological, malignant or infectious central nervous system (CNS) diseases in the past medical history.

The study design was approved by our institutional ethics committee. Written consent was obtained from all patients.

The examinations were performed using a 3.0 T whole-body scanner (3.0 T Intera; Philips Medical Systems, Best, the Netherlands) using an eight element head coil (InvivoTM; Gainesville, FL) suitable for parallel imaging. The gradient amplitude was 30 mT m–1, the rise time 0.2 ms resulting in a slew rate of 150 T m–1 s–1.

The complete imaging protocol included a T2W-FLAIR (fluid-attenuated inversion recovery) sequence, T1-SE sequence before and after intravenous injection of gadolinium–diethylenetriamine pentaacetic acid (DTPA; 0.1 mmol kg–1 body weight; Magnevist®, Schering AG, Berlin, Germany), T2W-TSE and a speed optimized FAS-TSE in two orientations. The T2W-TSE sequences were the object of our study; the other sequences were performed regarding clinical issues and verification of lesions in some cases.

In a pilot study [8], parameters of this FAS-TSE were evaluated by testing different SENSE factors, echo times (TE) and refocusing angles. The parameters with the best compromise of SNR, T2 contrast and scan time were chosen for our comparative study.

Both T2W-TSE sequences (sequence parameters listed in Table 1Go) were performed with a 5 mm slice thickness in transverse and with 2 mm slice thickness in the sagittal orientation. The geometric and contrast parameters were kept identical for both sequences. In the sagittal sequences, minor modifications (lowering the turbo factor and two signal acquisitions (number of excitations (NEX)) were necessary to compensate for the signal loss from thinner slices. Considering our experience from the pilot study, we applied a frequency selective fat suppression technique (SPAIR = SPectrally selective Attenuated Inversion Recovery) to reduce the visibility of the scalp, because structures from the edges of the image can appear in the centre of the image. This infolding arises from parallel imaging when high SENSE factors are used, because the truly measured field of view (FoV) is reduced and spatial misregistration can occur.


View this table:
[in this window]
[in a new window]

  		Table 1. MRI sequence parameters

 
The results of the standard T2W-TSE (sTSE) – serving as reference – were compared with the FAS-TSE sequence.

Image analysis
All images were transferred to a workstation (ViewForum 2003 Release 3.1, Philips Medical Systems, Best, the Netherlands), and all pulse sequences were analysed by two experienced radiologists in consensus. The FAS-TSE images were evaluated first, because the sTSE served as reference.

Both readers were blinded to the clinical presentations and the results of the clinical testing.

Lesionwise analysis
The detected lesions were counted and classified according to the anatomical location in juxtacortical, periventricular, callosal, infratentorial and other deep white matter, as defined in other studies. These studies defined a minimum lesion size of 3 mm to avoid confusion with unspecific white matter abnormalities and Virchow–Robin spaces [9, 10].

Contrast analysis
The lesions' contrast was measured by placing a region of interest (ROI) in lesions (L) ≥ 1 cm (to avoid partial volume effects) and copying the ROI to the correspondent white matter area (WM) of the other hemisphere (if not affected). This procedure was repeated in both sequences on the axial images only. The contrast was calculated from the measured signal intensities (si) using the formula: (siL–siWM)/(siL + siWM). Signal-to-noise (SNR) and contrast-to-noise (CNR) ratios could not be obtained, because we had to use SENSE, which is automatically combined with a homogeneity filter (called CLEAR). This filter suppresses the signal of the background noise and makes reliable measurements of SNR and CNR impossible.

The tissue contrast values of both sequences were matched lesionwise and statistically evaluated using the Wilcoxon test.

Qualitative analysis
Qualitative analysis was based on a five-point (1–5) rating scale and performed for:

Statistical differences between the two T2W sequences were assessed with the Wilcoxon test.

All statistical calculations were performed by the SPSS software package (SPSS Inc., Chicago, IL). p-values ≤0.05 were considered (expecting a better result of the sTSE) as statistically significant.


   Results
Top
 Abstract
 Introduction
 Patients and methods
 Results
Discussion
 References
 
33 patients (25 females, 8 males; age range 19–56 years, mean 34 years) were included in this study. White matter lesions were detected in 26 patients with the sTSE sequence, whereas lesions were found in 24 patients using the FAS-TSE sequence.

Lesionwise analysis
The sTSE sequence detected 208 lesions, whereas 183 (88%) were visible on the FAS-TSE images. The exact anatomical distribution is displayed in Table 2Go. The sTSE showed higher detection rates in all anatomical locations, especially in the juxtacortical and deep white matter region.


View this table:
[in this window]
[in a new window]

  		Table 2. Anatomical location of detected lesionsa

 
Contrast analysis
Contrast measurements as described above were performed in 89 lesions.

The mean value/range/standard deviation of the lesion/white matter contrast was 0.26/0.06–0.49/0.089, respectively, with the sTSE vs 0.21/0.04–0.40/0.081 with the FAS-TSE; the difference was statistically significant (p<0.01).

Qualitative analysis
The grey/white matter contrast was judged as "very good" using sTSE and FAS-TSE in all cases.

With respect to the criteria of lesion conspicuity and overall diagnostic image quality, all examinations were judged as "good or very good". The sTSE performed better in 18 out of 28 cases; consequently, equal results were found in 15 and 5 patients, respectively. The differences in the scores were statistically significant (p<0.01).

Concerning movement artefacts, we detected "no or minor artefacts" only. No difference was found in 19 patients. The FAS-TSE was ranked better in 14 cases, which was also a statistically significant result (p<0.01).


   Discussion
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
References
 
In the past few years, more highfield scanners (commonly operating at 3.0 T) have been installed, and the number of clinical examinations has increased. Owing to the higher SNR, highfield imaging should allow for faster scanning. It has to be taken into account that some physical conditions are altered at higher field strengths, such as the prolongation of the T1 and shortening of the T2 relaxation times. Tissue heating is probably the most important limiting factor, because RF power deposition increases with the square of field strength. This fact limits the advantages of a stronger gradient coil system and faster pulse switching, reducing the capabilities of highfield scanners [11–15].

Fast SE imaging techniques are based on an echo train of 180° RF refocusing pulses (number of pulses per echo train = turbo factor), enabling faster image acquisition but increasing RF deposition. As energy deposition is limited by SAR levels, the full potential of the TSE technique cannot be utilized.

Parallel imaging techniques, primarily used to reduce scan time, are highly valuable in highfield MRI, as they decrease the number of RF pulses, leading to a reduction in RF deposition.

The technique we used (SENSE [4]) takes information for spatial encoding from sensitivity profiles provided by the receiver coils and thus reduces the number of phase encoding steps.

Another approach to decreasing RF deposition is reduction in the 180° refocusing pulses. When using smaller refocusing angles, each pulse carries lower RF energy and, consequently, the number of pulses in a given TR can be increased. In consequence, higher turbo factors can be used resulting in a drastically shortened scan time. This technique is exactly described as hyperecho [5] but, to our knowledge, no clinical study has been performed to evaluate the capabilities of this technique.

In our FAS protocol, the refocusing pulse angle decreases from 180° to 60° in five steps at the start of the echo train, and the following pulses were applied with a 60° angle. The theoretical loss of signal is partially compensated by so-called stimulated echoes.

In contrast to hyperecho, our FAS sequence combines the reduced refocusing angle with the half-Fourier acquisition technique. This shortens the echo train further, resulting in reduced blurring and the capability of performing a single-shot acquisition. Using a half-Fourier factor of 0.65 means that 65% of the k-space lines in the direction of the y-axis were measured, and the remaining lines can be calculated on account of the symmetry of k-space.

Thus, with a combination of a SENSE factor of 4 and a half-Fourier factor of 0.65, a turbo factor of 43 is needed to encode a 256 matrix after one excitation pulse. The use of a single-shot multislice technique is another reason why motion artefacts can be reduced, because the acquisition of one slice needs approximately 0.5 s.

Our FAS-TSE protocol was designed for maximum speed and demonstrated an 11-fold reduction in scan time (compared with our standard T2-TSE) from 122 to 11 s for 24 slices. It is important to realize that all these techniques (FAS, SENSE and half-Fourier acquisition) reduce the SNR (Figure 1Go). We used the higher field strength only to compensate for this signal loss, because we wanted to achieve the maximum possible scan time reduction. This resulted in noisier images, which are sufficient for fast diagnostic brain imaging. Our approach is not suitable for high quality brain imaging, as shown by our results regarding small white matter lesions. But the FAS sequence can be easily adapted (e.g. lowering the SENSE factor, higher NSA) to individual requirements concerning scan speed and image quality.


Figure 1
View larger version (93K):
[in this window]
[in a new window]

  		Figure 1. Sagittal 2 mm thick images from a 47-year-old patient with a spinal cord lesion. The FAS-TSE (b) demonstrates a "frozen" aspect without pulsation artefacts in the brainstem, but a noisier image appearance (compared with the sTSE in (a)), clearly visible at the lower margin of the head coil masking the spinal lesion. The appearance of soft tissues is different, because fat saturation was applied with the FAS-TSE.

 
Tissue contrast may be influenced by applying lower refocusing angles but, with proper parameter settings, the usual T2 contrast is preserved [16, 17].

With the settings used in our study, the tissue contrasts of the FAS-TSE appeared similar to those of the sTSE, best demonstrated in the cortex where no infolding effects decrease the image quality (Figure 2Go). This is true for our patient group, but the contrast of other lesions such as tumour or infection might be different and will have to be evaluated in further studies.


Figure 2
View larger version (88K):
[in this window]
[in a new window]

  		Figure 2. These enlarged images from a 38-year-old female patient with a high periventricular lesion load display a similar contrast ((a) sTSE; (b) FAS-TSE) of grey and white matter, lesions and surrounding oedema.

 
Using the FAS-TSE, we detected 88% of all lesions with almost no difference between the periventricular and infratentorial region. In the juxtacortical and deep white matter region, the detection rates were slightly lower. This was partially caused by an infolding artefact near the midline (Figure 2Go), generally occurring with parallel imaging using high SENSE factors.

As motion artefacts were significantly reduced, in some cases, FAS-TSE was superior to the standard sequence (Figure 3Go). This effect is much more important in patients with lesser compliance than in our patients with suspected multiple sclerosis, serving as a test population. Therefore, the FAS-TSE technique should be considered especially for restless patients or when physiological movement impairs image quality (e.g. abdominal imaging). FAS-TSE could also be helpful when additional imaging planes are of interest but longer examination times are critical.


Figure 3
View larger version (65K):
[in this window]
[in a new window]

  		Figure 3. This example taken from a 21-year-old male patient suffering from a facial tic demonstrates the effect of unintended patient motion in the sTSE (a) compared with the FAS-TSE (b).

 
In conclusion, using the FAS-TSE technique presented in this study, it is possible to achieve a drastically reduced scan time for SE brain MRI, preserving the familiar imaging appearance and almost the diagnostic value of a standard T2W-TSE sequence.

Received for publication August 31, 2006. Revision received October 22, 2006. Accepted for publication October 25, 2006.


   References
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 

  1. Frayne R, Goodyear BG, Dickhoff P, Lanzon 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]
  2. Takahashi M, Uematsu H, Hatabu H. MR imaging at high magnetic fields. Eur J Radiol 2003;46:45–52.[CrossRef][Medline]
  3. Sicotte NL, Voskuhl RR, Bouvier S, Klutch R, Cohen MS, Muzziotta JC. Comparison of multiple sclerosis lesions at 1.5 and 3.0 T. Invest Radiol 2003;38:423–7.[CrossRef][Medline]
  4. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med 1999;42:952–62.[CrossRef][Medline]
  5. Hennig J, Scheffler K. Hyperechoes. Magn Reson Med 2001;46:6–12.[CrossRef][Medline]
  6. Hennig J, Weigel M, Scheffler K. Calculation of flip angles for echo trains with predefined amplitudes with the extended phase graph (EPG) algorithm: principles and applications to hyperecho and TRAPS sequences. Magn Reson Med 2004;51:68–80.[CrossRef][Medline]
  7. Busse RF. Reduced RF power without blurring: correcting for modulation of refocusing flip angle in FSE sequences. Magn Reson Med 2004;51:1031–8.[CrossRef][Medline]
  8. Gieseke J, Kuhl CK, Von Falkenhausen M, Schild HH. Ultra fast T2-weighted TSE sequences with flip angle sweep and SENSE at 3 T. Proc ISMRM 2004;11:74
  9. McDonald WI, Compston A, Edan G, Goodkin D, Hartung HP, Lublin FD, et al. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann Neurol 2001;50:121–7.[CrossRef][Medline]
  10. Barkhof F, Filippi M, van Waesberghe JH, Camp A, Miller DH, Ader HJ. Interobserver agreement for diagnostic MRI criteria in suspected multiple sclerosis. Neuroradiology 1999;41:347–50.[CrossRef][Medline]
  11. Wattjes MP, Harzheim M, Kuhl CK, Gieseke J, Schmidt S, Klotz L, et al. Does high-field image have an influence on the classification of patients with CIS according to current diagnostic MR imaging criteria for multiple sclerosis? AJNR Am J Neuroradiol 2006;27:1794–8.[Abstract/Free Full Text]
  12. Jacobs LD, Beck RW, Simon JH, Kinkel RP, Brownscheidle CM, Murray TJ, et al. Intramuscular interferon beta-1a therapy initiated during a first demyelinating event in multiple sclerosis. CHAMPS Study Group. N Engl J Med 2000;28:898–904.
  13. Heidemann RM, Griswold MA, Muller M, Breuer F, Blaimer M, Kiefer B, et al. Feasibilities and limitations of highfield parallel MRI. Radiologe 2004;44:49–55.[CrossRef][Medline]
  14. Baudendistel KT, Heverhagen JT, Knopp MV. Clinical MR at 3 Tesla: current status. Radiologe 2004;44:11–18.[CrossRef][Medline]
  15. De Vita E, Thomas DL, Roberts S, Parkes HG, Turner R, Kinchesh P, et al. High resolution MRI of the brain at 4.7 Tesla using fast spin echo imaging. Br J Radiol 2003;76:631–7.[Abstract/Free Full Text]
  16. Scarabino T, Nemore F, Giannatempo GM, Bertolino A, Di Sall F, Salvolimi U. 3.0 T magnetic resonance in neuroradiology. Eur J Radiol 2003;48:154–64.[CrossRef][Medline]
  17. Weigel M, Hennig J. Contrast behaviour and relaxation effects of conventional and hyperecho-turbo spin echo sequences at 1.5 and 3 T. Magn Reson Med 2006;55:826–35.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
C. K. Kuhl, F. Traber, and H. H. Schild
Whole-Body High-Field-Strength (3.0-T) MR Imaging in Clinical Practice * Part I. Technical Considerations and Clinical Applications
Radiology, March 1, 2008; 246(3): 675 - 696.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lutterbey, G
Right arrow Articles by Gieseke, J
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lutterbey, G
Right arrow Articles by Gieseke, J

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS