Image Processing Reference
In-Depth Information
when voxel sizes below 0.4 cm
3
are measured (58). When an absolute quantifi-
cation of the signals should be performed, SVS sequences are preferred, because
the PSF effects in CSI sequences make the quantification difficult.
11.4
DIFFERENCES IN SEQUENCES FOR
MEASUREMENTS WITH NONPROTON NUCLEI
The measurement techniques for volume selection in single-volume spectros-
copy sequences described in chapter 11.2
are widely used in
in vivo
H-MRS.
Spectroscopy measurements can also be performed using other nuclei such as
phosphorus-31, carbon-13, sodium-23, and fluorine-19. Because the resonance
frequencies of these nuclei are different from those of protons, specific coils
for the application of RF pulses and for signal receiving are necessary. A simple
spatial selection of the acquired signal is possible by using small surface coils,
which acquire only signals from the volume near the coil. In combination with
a CSI measurement, good spatial resolution is obtainable (44), but difficulties
arise from the spatial inhomogeneous coil sensitivity. The measurement with
surface coils is, however, limited to volumes of interest in the proximity of the
surfaces of volunteers or patients. If small volumes within the body have to be
examined, a volume coil similar to the head coil has to be used and an image
data set is necessary for the selection of the appropriate voxel. Therefore, so-
called double-tuned coils are desirable, which allow measurements at the res-
onance frequencies of protons and the specific nucleus of interest. Although in
principle the same measurement techniques used in H-MRS can be used for
measurements with other nuclei (so-called heteronuclear spectroscopy), for
some applications other methods have to be used. In measurements of phos-
phorus nuclei, e.g., signal contributions from resonances with very low T2
values should be examined in many applications. This is not possible if echo
times longer than 20 msec are used as is usual for applications of PRESS and
STEAM. For such applications, the ISIS technique (3) can be used as an
alternative. In this measurement technique, up to three slice-selective inversion
pulses are used as preparation pulses prior to a nonselective excitation, which
is followed directly by the data acquisition (
Figure 11.23
). The delay time
between the excitation and the beginning of the data acquisition can be shorter
than 1 msec in this case. In such a measurement, signals from the whole volume
excited by the 90
pulse are acquired. The selection of signals from chosen
parts of this volume requires several measurements with different combinations
of preparation pulses. For a selection of a cuboid, eight measurements are
necessary, and the signal from the VOI can be calculated from a series of
successive subtractions. This procedure has the disadvantage that small changes
between the single measurement, e.g., patient movements, can lead to large
errors in the spectrum calculation. Therefore, the applications of the ISIS
technique are limited to those cases in which the use of single-shot techniques
such as STEAM and PRESS is not possible.
°
Search WWH ::
Custom Search