MR Spectroscopy (MRS) and spectroscopic Imaging (MRSI):
Compared to 7T or 9.4T MR fields commonly used in preclinical imaging, where one dimensional MRS is limited to the accurate quantification of more than 12 metabolites, more than 30 potential biomarkers, including a few neurotransmitters, can be detected at higher fields. Further, the use of very high fields leads to high quality spectra that opens the door for dynamic spectroscopy and for accurate biochemical information of metabolites with low concentrations (~1mM), from a volume of interest as small as a few µL in a few minutes of acquisition time. To this end, our goal is to take advantage of the diverse spectroscopic acquisition types, with the development of multi-pulse, multi-dimensional spatial, spectral, and ultra-fast spectroscopy for the quantification of intrinsic biomarkers in the study of small animal brain and hepatic metabolism. The multi2D-dimensional acquisition is expected to enable a better separation of macromolecules and metabolites, facilitating biomarker identification. Quantitative algorithms dedicated to the fitting of 2D MRS signal are evaluated in terms of metabolite quantification accuracy. Our goal is to provide, for a given static field strength, accurate concentration measurements of the maximum number of neurotransmitters and neurochemical compounds such as glutamate, glutamine, GABA, scyllo-inositol, aspartate, taurine, N-acetyl-aspartyl-glutamate, glucose and several amino acids. The knowledge of the metabolite profile will enable a better understanding of the complex mechanisms of pathological processes. It will help the diagnosis and monitoring of conventional and novel therapies.
New MR encoding concept:
NMR signal generation, acquisition and processing is a 3D+t process but since its discovery, all these steps have been handled using a 2D complex algebra. A 3D approach of signal modeling acquisition and processing can lead to innovative NMR information types and totally new ways of NMR image acquisition. For this purpose, a hypercomplex algebra such as bi-quaternionic or Clifford algebra is mandatory and should lead to a new NMR signal formalism. Our goal is to completely drop the complex algebra in the description of Bloch's equation and its inversion. This approach is conceptually much cleaner than all the previous descriptions used so far and is of a great help in the comprehension and modeling of radio-frequency pulses. Geometric algebra is also insightful in the description of the properties of the signal acquired.