Diagnostic Features of 30 T MR Imaging

The main differences compared with lower field strength systems are: (1) changes in tissue contrast, (2) increased magnetic susceptibility, and (3) increased chemical shift.

Changes in Tissue Contrast

Whereas proton density is clearly a magnetic field-independent parameter, T1 and T2 relaxation times are field-dependent, usually increasing and, respectively, decreasing at higher magnetic fields.

The rates at which excited protons relax are a function of magnetic field strength. In general, the longitudinal or spin-lattice relaxation rate (R1 = 1/T1) of certain tissues (e.g. semisolids) decreases with field strength, whereas the transverse relaxation rate (R2 = 1/T2) is scarcely affected [5,19]. This leads to a 25-40% increase in T1 relaxation time in tissues on passing from 1.5 T to 3.0 T fields. By contrast, for biological fluids (like CSF and blood) the longitudinal rate is weakly affected, so that R1 and R2 are basically equal at both field intensities. As a result, given the same TR, T1 weighting is greater at 3.0 T than at 1.5 T. GoodT1 contrast can also be achieved using relatively long TR.

The increase in T1 relaxation time is a major drawback in high-field SE T1 imaging, as the reduction in T1 differences between tissues results in loss of contrast (Fig. 3.4).

T1 lengthening with increasing magnetic field also applies to blood. Blood T1 is insensitive to the level of oxygenation and increases linearly with field strength. This is particularly valuable in MR angiography (especially in time-of-flight sequences), as it involves greater background suppression due to a lower R1 of stationary tissues and a greater flow enhancement given the broadly constant R1 of blood, thus yielding better vessel-tissue contrast (Fig. 3.5) [20, 21].

Fig. 3.4. T1 imaging: SET1 sequence acquired at 3.0 T (TR/TE 500/min, slice thickness 5 mm, FOV 24, matrix 320x224,1 NEX, 2:00) (a) and 1.5 T (TR/TE 500/min, slice thickness 5 mm, FOV 24 cm, matrix 256x 192, 1 NEX, 1:23) (b). The increased T1 involves reduced contrast at 3.0 T with respect to 1.5 T

Fig. 3.5. Normal arterial circulation: 3D TOF study at 3.0 T (a) and 1.5T (b). Vascular conspicuity and background are greater in a (TOF SPGR TE/TE/FA 30/min/20, slice thickness 1.4 mm, matrix 512x320, ZIP 1024, ZIP 2, 1 NEX, 50 locations per slab, 6:18) than in b (TOF SPGR TE/TE/FA 45/4/20, slice thickness 1.4 mm, FOV 24 x 18, matrix 512 x 224,1 NEX, 50 locations per slab, 6:22) resulting in greater vessel-tissue contrast

Fig. 3.5. Normal arterial circulation: 3D TOF study at 3.0 T (a) and 1.5T (b). Vascular conspicuity and background are greater in a (TOF SPGR TE/TE/FA 30/min/20, slice thickness 1.4 mm, matrix 512x320, ZIP 1024, ZIP 2, 1 NEX, 50 locations per slab, 6:18) than in b (TOF SPGR TE/TE/FA 45/4/20, slice thickness 1.4 mm, FOV 24 x 18, matrix 512 x 224,1 NEX, 50 locations per slab, 6:22) resulting in greater vessel-tissue contrast

The long T1 is also useful in perfusion studies, as As regards brain tissue T2 relaxation time, it dimin-spin tags persist over a longer time, boosting sensiti- ishes with increasing magnetic field, albeit not linearly. vity. In fact, T2 contrast is largely unaffected by field strength, probably due to the action of other mechanisms (like exchange and/or diffusion during gradient activation).

T2* is also shorter at 3.0 T. This maybe useful when studying the deoxyhaemoglobin-containing vascula-ture (i.e. the venous system) and can also yield brain tissue contrast, for instance between grey and white matter. Indeed, the shorter T2* relaxation times enhance the differences in the T2* of different tissues, resulting in greater distortion and in a larger number of artefacts due to signal loss. Indeed, changes in R2* are sensitive to changes in B0 and are therefore related to differences in field susceptibility and homogeneity. Optimal B0 field homogeneity, improved high-order coil shimming and broader reconstruction band-widths are essential to improve the signal in such circumstances.

Increased Magnetic Susceptibility

Susceptibility effects increase with field strength. Increased magnetic susceptibility enhances the BOLD (blood oxygenation level dependent) effect, making clinical BOLD imaging more practical and informative as a result. BOLD contrast is the result of small magnetic susceptibility effects also responsible for the MR signal changes caused by changes in blood oxygenation under 3 %.

3.0 T fMR images reflect greater signal changes (up to 7%) (Figs. 3.6-3.8) [22-25].

The increased static MR signal can be used to reduce the cortical volume needed for signal averaging to

Fig. 3.7. Visual fMRI at 3.0 T: the enhanced BOLD effect maps additional areas at the millimetre and submillimetre levels (arrows)
Fig. 3.6. Motor fMRI with bilateral finger tapping: study at 3.0 T (a) and 1.5T (b). The enhanced sensitivity to blood oxygenation and reduced background noise of the 3.0 T system (a) affords more reliable localization of motor areas compared with the 1.5 T imager (b)

intensity O.S - 2 sec 5 sec 10 sec Time

Fig. 3.8. Visual fMRI: the signal level obtained with the 1.5 T system is 1-3% (green) compared with 2-5% with the 3.0 T system (red)

intensity O.S - 2 sec 5 sec 10 sec Time

Fig. 3.8. Visual fMRI: the signal level obtained with the 1.5 T system is 1-3% (green) compared with 2-5% with the 3.0 T system (red)

achieve sufficient SNR. Another, potentially more important advantage of the higher B0 is that as the field strength increases, the field gradient around the capillaries becomes larger and extends further into the parenchyma, thus involving a greater amount ofbrain tissue in producing the functional signal. In field gradients, the magnetization vectors inside voxels attain different phases and the effective transverse relaxation time T2* decreases as a result. Concurrently, the shortened T2* of blood at high B0 reduces the relative contribution from the large veins. So the weighting of capillary signal contributions becomes more significant and the functional signal is more closely coupled with neuronal activity [26].

The BOLD contrast increases as a result, becoming more sensitive to the susceptibility effects in and around the smaller vessels rather than in and around larger draining vessels. GE-EPI are usually preferred to SE-EPI sequences because they afford better discrimination of small from large vessels. SE-EPI are the sequences of choice for the study of small vessels on which the BOLD effect depends with the increase in magnetic field strength. In addition, because SE-EPI sequences are less sensitive to magnetic susceptibility effects, they are preferred for fMRI in areas where differences in susceptibility at the air-tissue interface (e.g. paranasal sinuses) are more difficult to depict at high-field strengths, resulting in marked SNR reductions.

The greater sensitivity of high-strength magnetic fields to magnetic susceptibility can also be exploited to boost the sensitivity of contrast-enhanced perfusion studies and to improve the sensitivity of FSE sequences to haemorrhagic lesions (Figs. 3.9-3.11) [19, 27, 28].

However, 3.0 T MR systems have two potential disadvantages with respect to 1.5 T scanners. Firstly, EPI sequences are usually noisier due to switching of the usually more powerful gradients, even more so when EPI sequences are acquired with GE sequences, because using SE sequence refocalization prevails over phase dispersion, thereby reducing the noise. Secondly, high-field systems are characterized by a larger number of magnetic susceptibility artefacts, especially when us-

Fig. 3.9. Arachnoid cysts complicated by haemorrhage in subacute-chronic stage after surgery. Study with GRE T2 sequences at 3.0 T (TR/TE/FA 525/9.8/20, slice thickness 5 mm, FOV 24 X 18, matrix 512 X 224,2 NEX, 2:60) (a) and 1.5 T (TR/TE/FA 500/15/20, slice thickness 5 mm, FOV 24 X18, matrix 320 X 192,2 NEX, 2:28) (b). The greater sensitivity of 3.0 T MRI to the effects of magnetic susceptibility affords better depiction of haemosiderin deposits

Fig. 3.9. Arachnoid cysts complicated by haemorrhage in subacute-chronic stage after surgery. Study with GRE T2 sequences at 3.0 T (TR/TE/FA 525/9.8/20, slice thickness 5 mm, FOV 24 X 18, matrix 512 X 224,2 NEX, 2:60) (a) and 1.5 T (TR/TE/FA 500/15/20, slice thickness 5 mm, FOV 24 X18, matrix 320 X 192,2 NEX, 2:28) (b). The greater sensitivity of 3.0 T MRI to the effects of magnetic susceptibility affords better depiction of haemosiderin deposits

Fig. 3.10. Leptomeningeal haemosiderosis studied with FSE (a) and GRE (b). Despite the intrinsically lower sensitivity of FSE to magnetic susceptibility, the 3.0 T imager affords excellent depiction of the low haemosiderin signal, which is visualized even more clearly in the GRE sequence
Fig. 3.11. Multiple recent haemorrhagic foci studied with FSE (a) and FLAIR (b). Suppression of the CSF signal improves detection of the smaller lesions in the FLAIR sequence

ing GE and EPI sequences (mainly in the areas that most commonly generate artefacts, such as the temporal lobes), and from blurring when FSE sequences are performed, which often yield unsatisfactory images. In both cases, experience and use of correction systems (phase map correction) and techniques such as radial array orientation of fc-space data and propeller CD help dispel diagnostic doubts. Finally, by reducing effective echo spacing and TE at the expense of SNR, parallel imaging allows images to be obtained with an artefact incidence comparable to that of 1.5 T units [29].

Increased Chemical Shift

Chemical shift effects also increase with magnetic field strength. The increased chemical shift enhances spec-troscopic investigations via an increased chemical shift resolution, which affords wider separation of the absolute frequencies of different metabolites (Fig. 3.12) [30-32]. However, a larger number of artefacts is generated at the interfaces between tissues with different chemical bonds (fat/water) due to erroneous signal attribution within the reconstruction matrix, resulting in diagnostic limitations in standard anatomical imaging.

At 3.0 T, the water/fat chemical shift is around 440 Hz; this leads to the use of FSE sequences, in which more 180° pulses (greater echo train length), fewer TE and reduced slice thickness compensate for the susceptibility artefacts at the interfaces of tissues with different susceptibility constants. In addition, broader receiver bandwidths (32-125 kHz) are applied to keep the chemical shift within limits that do not impair diagnostic quality. Indeed, to reduce these artefacts to levels seen at 1.5 T, since the chemical shift frequency between water and fat doubles at 3.0 T, the receiver band width needs to be doubled too, with a consequent reduction in SNR by approximately 40 %. All other imaging parameters being equal, the images obtained at 3.0 T, which show a roughly 60 % improvement in SNR over those acquired at 1.5 T, will therefore have 20% better SNR. Attempts to enhance spatial resolution by increasing the in-plane matrix, reducing section thickness, or both will further compromise the SNR [33]. The reduced SNR associated with the receiver bandwidth is, however, partly compensated for by the higher SNR of 3.0 T systems and by last generation multichannel coils.

One alternative to reducing chemical shift artefacts that does not entail image degradation is fat saturation, which however will further reduce the number of sections that can be acquired because it increases the SAR. Another is water excitation, which does not increase the SAR significantly, improves image resolution and has an SNR similar to that at 1.5 T [33].

Relaxation Audio Sounds Lazy Summer Day

Relaxation Audio Sounds Lazy Summer Day

This is an audio all about guiding you to relaxation. This is a Relaxation Audio Sounds with sounds from Lazy Summer Day.

Get My Free MP3 Audio


Post a comment