In-vivo proton magnetic resonance spectroscopy (MRS) boasts the capability to provide quantifiable metabolite information (spectra) acquired from living tissue, an established procedure for the target metabolites containing uncoupled spin groups. Unfortunately, the majority of the metabolite spin systems found in-vivo comprise systems of scalar-coupled spins that, unlike the uncoupled groups, contribute complex multiplets of peaks with elaborate responses to the pulse sequences used to observe them. In addition to these complex signal modulations, the similarity of several of the proton spin systems found in-vivo results in significant spectral overlap, making unique identification problematic. Furthermore, metabolite quantification is hampered by a limited signal to noise ratio, inherent to NMR techniques, necessitating an optimization of the metabolite signal yield. The objective of this thesis was to evaluate and optimize the response of several of the metabolite spin systems found in-vivo to the realistic NMR pulse sequences used to observe them.

To provide a framework in which the most demanding strongly-coupled spin systems could be treated, including the influence of time-dependent radio frequency pulses, a numerical method of evaluation that incorporated the density matrix representation of the spin systems was developed. The Hamiltonian used included, in addition to the Zeeman interaction, the rf pulses, the gradient pulses, the chemical shielding interaction, as well as the scalar and dipolar coupling interactions.

The response of several scalar-coupled spin systems found in-vivo, including lactate, glutamate and glutamine, as well the aspartyl group of N-acetylaspartate, to a variety of in-vivo pulse sequences was calculated. A complete single voxel multiple quantum filter sequence was developed for the observation of glutamate, providing discrimination from the nearly identical glutamine resonances. The response of the coupled-spin systems to the single voxel PRESS and STEAM sequences were also calculated, with a focus on the characterization and optimization of those responses for the purposes of quantification. Finally, the influence of the direct dipole-dipole interaction, on the responses of both the Cr / PCr spins found in muscle and a concentrated pool of water spins, to in-vivo pulse sequences, were calculated. The calculated metabolite responses were in excellent agreement to both phantom and in-vivo results at 3 T.