**2.5 Animal model**

128 Advances in the Biology, Imaging and Therapies for Glioblastoma

using spin-echo sequences. However, their filling factor is relatively low, which can penalize the S/N. This sensitivity problem can be partially mitigated by using cross-polarized antennae. This type of volumetric antenna is the easiest to use, and was used for the images

Surface antennae can also be used, offering an excellent quality factor combined with a very high filling factor. This type of antenna therefore provides a higher S/N than the antennae

However, due to their configuration, the flip angle applied varies depending on the depth of the zone to be imaged within the sample, making the use of spin-echo sequences impossible. In addition, with this type of antenna, contrast can vary depending on the depth of the zone

Finally, in clinical imaging, and increasingly for small-animal imaging, volumetric emission antennae coupled to an array of surface reception antennae are the most commonly used. This setup offers the advantages of the two types of antennae described above: excellent B1

Fig. 2. a) Cross-polarized 25-mm-diameter volumetric antenna; b) 16-mm-diameter surface emission/reception antenna; c) & d) antenna system combining volumetric emission (80 mm

These three types of antennae are exclusively used for mouse brain imaging and were tuned

Finally, quality imaging in small animals requires a specific component to position and monitor the animal. This component must allow the animal to be maintained in a stable and

emission homogeneity associated with an excellent signal-to-noise ratio.

presented in this chapter.

observed within the sample.

in diameter) and phased-array reception.

**2.4 Animal handling/monitoring** 

to 200 MHz in the experiments described here.

described previously.

U87 human brain tumor cells were implanted in nude mice (18-20 g, n = 20, Charles Rivers, L'Arbresle, France) by stereotactic injection into the striatum. Mice were anesthetized with isoflurane (1.5% in air) and secured in the stereotactic apparatus (Stoelting Europe, Dublin, Ireland). The scalp was cleaned with Betadine (MEDA Pharma, Paris, France) and the skull was exposed by midline scalp excision. A small hole (0.5 mm in diameter) was then drilled 0.1 mm posterior and 2.3 mm left to the bregma. Five hundred thousand U87 cells dissolved in 2 mL of Minimum Essential Medium were injected using a 10-mL Hamilton syringe into the left hemisphere at a depth of 3 mm below the brain's surface. On withdrawal of the injection needle, the hole in the skull was sealed with bone wax and the scalp was sutured.

3D TrueFISP MRI Provides Accurate Longitudinal Measurements of Glioma Volumes in Mice 131


Simulations of the state of magnetization with TrueFISP, gradient-echo and spin-echo sequences were performed in Igor Pro using well-known equations (Eq. [4], [5] and [6)]. They were performed as a function of flip angle α, TR, TE or the T1 and T2 values of tissues. At 4.7 T, the T1 and T2 values of brain was equal to 1,295 ms and 53 ms, respectively, and

A 1/6.32 correction factor was applied to 2D RARE (Eq. [6]), corresponding to the voxel size, the number of averages and the number of k-space samples by comparison with 3D

 (6) S/N was simulated for the brain and tumor, and contrast was evaluated according to

The signal-to-noise ratio was evaluated for both tumor and brain tissues. For the TrueFISP sequence, the maximal signal was obtained with the shortest possible TE and a flip angle between approximately 15° and 35°, as shown in Figs. 5ab. The slight S/N difference between the healthy brain and the tumor is also observable. The gradient-echo sequence generates a much lower S/N compared with the TrueFISP sequence for all flip angles.

(7)

(4)

(5)

time: 2 min 34 sec x4 = 10 min 16 sec; transverse orientation.

The following sequence parameters are used:

for tumors, 1,525 and 72 ms, respectively [8].


**3.3 Theory** 




Eq. [7].

TrueFISP acquisition.

with TE = 2.5 ms and T2\* = 25 ms
