Functional MRI methodology for the Basal Ganglia -- a comparative study into paradigms and analytical methods

The basal ganglia (BG) are structures in central areas of the brain and form
the central part in a network. They interconnect with cortical areas of the
brain and play an important role in the initiation, alteration or inhibition of
movements. Furthermore, the BG play are involved in the adaptation to novel
circumstances, learning of a new cognitive or motor skill, and reward actions.
The BG network of interconnected structures consists of different nuclei, such
as the Globus Pallidus (GP), Substantia Nigra (SN), Putamen, Caudate Nucleus
(CN), Sub Thalamic Nucleus (STN) and the Thalamus. Malfunctioning of this
network can lead to movement disorders such as Parkinson*s disease or dystonia.

Functional magnetic resonance imaging (fMRI) is a technique to measure brain
activity. It allows visualization of activity in the deeper areas of the brain
such as the BG. It relies on local differences in image intensities of
repetitive functional scans caused by cognitive processes. These functional
scans are known as EPI scans and are taken every two to three seconds. In
addition to the EPI scans, an anatomical scan is performed, known as a T1 scan.

In a functional MRI experiment, there is no pre-fixed protocol and the choice
of the different parameters is determined by the researcher. The most important
choices are the tasks the subject has to perform while in the scanner
(paradigm) and the processing of fMRI data sets afterwards.

fMRI of the BG is more difficult than functional imaging of cortical areas as
there are three reasons the functional signal in the BG is reduced:

1. The BG are relatively small, compared to the typical voxel size used in fMRI
studies.
2. The BG have more iron deposits, which interferes with the contrast used with
fMRI.
3. Tissues that surround the BG, such as the ventricles, deform the magnetic
field. This leads to deformation of the anatomy measured by the functional
scans 6.

The aim of our study is to investigate the effect of different paradigms and
analysis methods on the final result of functional MRI BG data. The second aim
is to determine the most optimal paradigm and analysis method for further
research in patients with movement disorders.

Scanner details
A high resolution of 128x128 will be used. This will allow a smaller voxel
size, to compensate for the small size of the BG. With a special sequence, a
magnetic field map will be measured. With this map, a correction for magnetic
field can be performed which will compensate for the magnetic field in
homogeneity caused by the ventricles surrounding the BG.

Paradigms
Four different paradigms, each chosen with the goal of generating activation
patterns in the BG, will be performed. Each paradigm lasts for about 15
minutes.

1. Serial Reaction Time (SRT). This is a cognitive task which measures the
neural correlates of implicit learning. A previous study showed that using this
task yields activation in the BG.

2. Go-NoGo. The Go-noGo is a motor inhibition task. Local Field Potential (LFP)
recordings with intra-cerebral electrodes inside the STN and GP show a
different electrical response when subjects have to inhibit a motor task. We
will translate the same tasks to a functional MRI setting.

3. 'Cognitive' Switch Task This is a cognitive task which measures the neural
correlates of switching between two tasks, namely identifying colors and
identifying letters. A previous study showed activation patterns in the BG.

4. Fingertap-Switch Task. This includes different types of motor tasks of the
hands. The activation pattern will be studied during each task, but also during
the switching periods between the different motor tasks.

nvt