The work in this thesis is focused on the study of neuroinflammation with molecular magnetic resonance imaging (MRI) methods. Neuroinflammation is a response of the central nervous system to pathological insult and it is present in many neurological disorders, such as Alzheimer’s disease. Being able to image neuroinflammation non-invasively with MRI techniques would have an important clinical value for diagnosis and assessment of therapy effectiveness. The aim of this work is to develop and validate an MR biomarker of neuroinflammation using MR Spectroscopy (MRS) and chemical exchange saturation transfer imaging (CEST). First, intravenous administration of lipopolysaccharide (LPS) is used as a mild inflammatory stimulus in wild type mice and in a mouse model of Alzheimer’s disease (AD). Elevated levels of the osmolyte myo-inositol, measured with MRS and microglia activation are found in AD mice after LPS administration. Due to the inherent low spatial resolution of MRS, a CEST MRI method is developed next. A myo-inositol CEST protocol is optimised, using Matlab simulations based on the Bloch-McConnell equations for a three pool model, in order to maximize the contrast and to estimate the amount of signal that can be expected in vivo. In vitro and in vivo tests are presented and a fast CEST sequence is developed, while the experimental difficulties and limitations of the technique are discussed. A CEST protocol is finally applied to evaluate the metabolite response to an LPS inflammatory challenge using MRS and histology as validation. A correlation is described between CEST and MRS myo-inositol levels, as well as between CEST and microglia concentration (Iba1 immunostaining), which highlight the potential of CEST as a non-invasive in vivo neuroinflammatory biomarker.