Although in vivo human MRS has existed only since the early 1980's, in vitro NMR spectroscopy has been performed in chemistry laboratories since the 1950's and is now quite sophisticated. The world's most powerful NMR spectrometer currently in operation is a 23.5T unit in Lyon, France at the Centre Européen de Résonance Magnétique Nucléaire (CNRS). The brief video (right) shows how a much smaller laboratory NMR spectrometer operates.
|
|
Advanced Discussion (show/hide)»
The statement that readout gradients are not used in MRS studies is not quite correct. While readout gradients are not typically used in clinical MRS protocols, there are special fast chemical shift imaging techniques where they are employed. These include Echo-Planar Spectroscopic Imaging (EPSI) and Spiral Spectroscopic Imaging, as well as multiple other fast chemical shift imaging methods. In these cases care is taken to keep track of the spatial and spectral components present in the signal, such as by scanning only one line of k-space at a time, or by adjusting pulse timing intervals or gradient areas between acquisitions.
Another feature distinguishing MRS and MRI is that the data sampling window is much longer in the former. In MRS sampling times may be several hundred milliseconds to allow for frequency discrimination between closely spaced spectral lines.
In MRI water and fat protons are responsible for nearly all of the recorded signal in each voxel. In MRS, tissue water and fat are often viewed as background contaminants that obscure the small signals from metabolites of interest. These metabolites are typically present in millimolar (mM) concentrations, a thousand times smaller than those of water and fat. Accordingly water and fat signals need to be suppressed and/or excluded from the voxel of interest. Methods for accomplishing this are provided in later Q&A's.
Because MRS localization methods are less efficient than frequency encoding and metabolite concentrations are low, some sacrifice of spatial resolution is required when obtaining spectra instead of images. In practice, MRS studies typically use voxel sizes of 1 to 5 cm³, whereas voxels for imaging are a thousand times smaller (1−5 mm³). This is not necessarily bad, however, as larger voxels permit detection of low concentration metabolites with acceptable signal-to-noise in reasonable imaging times.
Bertoldo D, Watcharakorn A, Castillo M. Brain proton magnetic resonance spectroscopy. Introduction and overview. Neuroimag Clin N Am 2013; 23:359-380.
Ernst RR, Anderson WA. Application of Fourier transform spectroscopy to magnetic resonance. Rev Sci Instrum 1966; 37:93-102. (Famous paper showing how Fourier transform of the MR time domain signal yields a frequency spectrum)
Hajek M, Dezortova M. Introduction to clinical in vivo MR spectroscopy. Eur J Radiol 2008; 67:185-193.
Pohmann R, von Kienlin M, Haase A. Theoretical evaluation and comparison of fast chemical shift imaging methods. J Magn Reson 1997; 129:145-160. (Exception to the rule above that readout gradients can't be played during MRS studies).
Posse S, Otazo R, Dager SR, Alger J. MR spectroscopic imaging: principles and recent advances. J Magn Reson Imaging 2013; 37:1301-1325.
Skoch A, Jiru F, Bunke J. Spectroscopic imaging: basic principles. Eur J Radiol 2008; 67:230-239.