Many different regions of the brain are involved in appetite control. A full understanding
of their function and interaction requires studying neuronal activity at high resolution
simultaneously in space and time. Two Magnetic Resonance Imaging (MRI) methods can
potentially achieve this goal. Manganese-Enhanced (MEMRI) uses the accumulation of
administered Mn[2+], which is paramagnetic (hence MRI visible) and taken up by active neurons
through voltage-gated Ca[2+] channels during action potentials. Haemodynamic methods use one
or more of many MRI-visible changes that occur to circulating blood in a brain region when it
changes activity. These include blood-oxygenation level dependent (BOLD) and cerebral blood
volume weighted (CBV) MRI. The aim of this project was to further develop, adapt and then use
these methods to study the effects on neuronal activity of stimuli related to appetite and energy
balance. The majority of work went towards adapting MEMRI for this. Amongst many tested
changes, improvements were made to the MRI acquisition protocol (specifically using fast spin
echo rather than spin-echo acquisition) to make it more sensitive to Mn-induced signal changes,
increase spatial coverage from partial to whole brain and rostro-caudal spatial resolution from
1 to 0.4mm, all while maintaining the same temporal resolution. Most importantly, the
neuroimaging analysis framework used in haemodynamic functional MRI was adapted for
use with MEMRI. This included the adaptation of spatial normalization software to handle
Mn-sensitive T[1]-weighted images dominated by non-brain tissue rather than brain dominated
T[2]/T*[2]-weighted images, and the generation of a signal change model for use in GLM. This
enabled much more objective, reproducible and less laborious data analysis than with previous
hand drawn ROIs. Attempts were made to use BOLD- and CBV-fMRI to study the effects
of potent, appetite-modulating gut hormones on appetite, though these failed to produce a
response.