Principles of MR Spectroscopy and Chemical Shift Imaging - Advance Image Processing in Magnetic Resonance Imaging - page 369

into the transverse plane ( Figure 11.4g ). The net magnetization after this pulse

is zero because the direction of the magnetization vectors varies between the + x

and the x direction and, therefore, the magnetization contributions are cancelled

out. After the third pulse, a rephasing occurs similar to conventional spin-echo

−

sequences ( Figure 11.4h ) and, at the time TE + TM, all spins are again in the

same phase (Figure 11.4i) and build a measurable net magnetization vector
(Figure 11.4j). Due to the missing influence of that magnetization, which has
been dephased within the TM interval between the second and third pulses, the
amplitude of the sum vector at the beginning of the data acquisition is only half
of the amplitude that can be obtained in a spin-echo sequence with the same echo
time (4). The main advantage of the STEAM sequence compared to PRESS is a
reduced minimal echo time. The time between the second and the third RF pulses
is part of the echo time in spin-echo sequences. In STEAM sequences, however,
the relevant part of the magnetization vectors have only longitudinal magnetiza-
tion in this time and experience only T1 relaxation, but no T2 relaxation. Because
in human tissue T1 is usually much longer than T2, the signal loss in the time
between the second and third pulses is much less in STEAM sequences. This
sequence is therefore used if very short echo times should be realized. Another
advantage of the STEAM sequence is the avoidance of 180 pulses and, therefore,

the increased difficulties with the higher pulse amplitude in the center of the pulse
and with the nonideal slice profile can be avoided (5).

°

11.2.2 A SVS

RTIFACTS IN

One of the most important quality parameters in SVS is the volume selectivity of
the signal. The measurement sequence must avoid any signal from the regions
outside the volume of interest. This can be realized by the careful spoiling of
unwanted signals. Unwanted signals can originate from all positions in which
transverse magnetization is produced during the measurement sequence. Transverse
magnetization is produced at first within the whole slice excited by the first 90 °

pulse, but only the small part of the slice within the volume of interest should
contribute to the measured signal. The two following pulses are again excitation
pulses in STEAM sequences, and they lead to transverse magnetization throughout
the excited slices. The entire volume with unwanted transverse magnetization in a

STEAM measurement consists of three orthogonal slices and is shown in Figure 11.5 ,

whereas the volume of interest (from which the signal should originate) is the
intersection of the three slices. In order to obtain good spatial selectivity for the
acquired MRS signal, any signal contribution from the unwanted transverse mag-
netization must be strongly reduced. This can be realized by appropriate spoiling,
which leads to strong dephasing of the magnetization. Spoiling can be achieved by
the application of strong field gradients without RF pulses. The effect of a spoiler
gradient is a large distribution of phase angles of the transverse magnetization within
even small structures. This phase distribution leads to a cancellation of the signal
contributions from these structures. In the volume of interest, which is also affected
by the applied spoiler gradients, this signal cancellation can be avoided

by the