Item

Abstract

Purpose/Introduction: Proton magnetic resonance spectroscopy (MRS) enables the detection of several metabolites within the human body. The concentration of water is av thousand times higher than those of the metabolites. As a consequence, it introduces baseline distortions, eddy current and gradient modulation artifacts in the final spectrum [1–2]. Several techniques have been developed trying to suppress the water peak using special water suppression (WS) methods [3–4] etc. At low magnetic fields (i.e.\3 T) the performance of the aforementioned WS techniques can be quite good. However, at higher magnetic fields this is a difficult task due to B1 + inhomogeneity. A solution to this problem is the development of VAPOR [5]. This method enables very good WS against B1 + ingomogeneity at the expense of time (*700 ms). This duration can prolong significantly the examination time, especially in case of time consuming sequences (e.g. multi slice MRSI). For this reason, the aim of this study was the development of a short duration water suppression using optimised flip angles (SODA) at 9.4 T. Subjects and Methods: An algorithm was developed trying to minimize the residual longitudinal magnetization for different number of CHESS [4] pulses, T1 relaxation times, B1 + inhomogeneity and time delays (Fig. 1). After the determination of the desired flip angles and time delays, an algorithm was written in order to find the best combination of gradient amplitudes for the suppression of the unwanted coherence pathways (Fig. 2; after n pulses, 3n coherence pathways are produced). Afterwards, the final WS sequence was incorporated with a STEAM technique and tested on a phantom and a healthy volunteer. All the experiments were carried out using a 4 channel transceiver array coil connected to a whole body 9.4 Tesla SIEMENS scanner. Results: The phantom results demonstrated a suppression of 99.7 of the initial water peak (suppression factor = 449; Fig. 3). In addition, in vivo results showed that SODA scheme allows sufficient WS giving a suppression factor[1300. Discussion/Conclusion: In this study we demonstrated a short duration WS scheme using optimized flip angles (SODA) and gradients showing preliminary results of this study. This method has a lot of room for improvements such as the introduction of phase cycling during SODA and the replacement of the Gaussian pulses with pulses with better profiles.