Type C hepatic encephalopathy (HE) is a severe neuropsychiatric complication of chronic liver disease, for which the prognosis is poor in the absence of liver transplantation. Cirrhosis in type C HE leads to a toxic accumulation of ammonia in the blood, which will eventually travel to the brain and adversely affect its structure and function. However, the biochemical mechanisms underpinning neurological and cognitive dysfunctions are intertwined and still incompletely understood. First, it remains unclear how brain cells morphology is affected by the ammonia-induced glutamine increase and osmotic stress in HE. In the bile-duct ligated (BDL) rat model of type C HE, alterations of neurons and astrocytes' shape have been observed ex vivo by histology, but these observations were until now not replicated in vivo. Magnetic resonance (MR) spectroscopy (MRS) at ultra-high field is a powerful tool to probe metabolism in vivo, and can, with the insertion of diffusion gradients, in addition be sensitized to probe cell-specific microstructure. In this thesis, diffusion-weighted MR spectroscopy (dMRS) and imaging (dMRI) experiments were conducted at 14.1T in the developing brain of the BDL rat model of type C HE. The acquisition was optimized to measure the cerebellum, a challenging brain region due to motion and the presence of fat, but of particular vulnerability in HE. Analysed jointly through cell-specific biophysical modelling, dMRS and dMRI probed faster metabolite diffusivities and faster intra-neurite/intra-axon water diffusivity in cerebellar white and grey matter of BDL rats compared to control rats. These observations point towards an alteration of cell density and/or of neurite network complexity and reorient the debate from the restrictive hypothesis of astrocytes swelling to the wider one of multi-cellular microstructure alterations in type C HE. The dMRS acquisition was further optimized with the implementation of a new sequence, DW-SPECIAL. The latter improved the detection and subsequent estimation of the diffusion properties of strongly J-coupled metabolites such as glutamine, of particular interest in the study of HE. A post-processing denoising technique based on the Marchenko Pastur principal component analysis method (MP-PCA) was also tested on simulated, rodent and human dMRS data. MP-PCA denoising yielded both valuable and adverse features specific to the nature of the input data, an effect for which a detailed description was provided and which should be carefully considered. Second, conflicting results on brain energy metabolism alterations in type C HE have been previously reported. Positron emission tomography (PET) is an imaging modality that enables the study of glucose uptake, following the conversion of fluorodeoxyglucose (FDG) in the first steps of the glycolysis in vivo. In this thesis, a new preclinical FDG PET methodology was implemented to compute quantitative 3D maps of the regional cerebral metabolic rate of glucose (CMRglc) from a labelling steady-state PET image of the brain and an image-derived input function. A 2-fold lower CMRglc brain glucose uptake was observed in the hippocampus and cerebellum of the BDL rats. Combined with MRS, it provided for the first time local and quantitative information on both brain glucose uptake and neurometabolic profile alterations in a ratmodel of type C HE. The quantitative approach also showed its strength when comparing groups of animals with divergent physiology.