The classification of subfunctions of spatially defined brain regions within large-scale functional networks is an important matter of neuroscientific research. Although cytoarchitectonic measures can be applied to the neocortex to map functional-anatomical subregions, the cerebellar cortex is a challenge due to its uniform cytoarchitecture. Instead of histological approaches, localized cerebellar functions have been inferred using clinical lesion studies and brain activation studies. However, the significance of their findings is limited by a large inter-subject heterogeneity and the difficulty of interpreting unspecific coactivation. A promising approach is to understand the role of the cerebellum in brain function by mapping its relationship to the rest of the brain. Invasive animal studies using tracing methods already provided cerebellar subdivisions through polysynaptic cerebello-cerebral connectivity. In the current work, connectivity was measured using functional magnetic resonance imaging in humans, acquired during the resting-state. The synchronization of low-frequency fluctuations described by this data is used to characterize functional connectivity. Based on the well-studied primate sensorimotor system as well as human sensorimotor-related activation studies five cerebellar regions were selected (lobule I-IV, V, VI, VIIIa/b). By means of their connectivity patterns to the cerebrum, we investigated whether known and less well studied functional subdivisions could be described within and across these lobules. The dissertation consists of three parts: (i) cerebellar signal assessment (ii) cerebello-driven connectivity maps, (iii) cerebellar sublobular topography. We developed an algorithm which substantially improved the signal-to-noise ratio in anatomically defined cerebellar lobules by differentiating cerebellar and cerebral signals. In a correlation analysis, the mean time series of the refined cerebellar lobules revealed five cerebello-driven brain networks. The regions in those cerebral networks have been found to not only process sensorimotor information, but to also be involved in affective and cognitive control, language and executive functions. A subsequent partial correlation analysis of the mean cerebral time series with voxelwise cerebellar time series provided a systematic mapping of cerebral cconnectivity in cerebellar lobules. This functional sublobular topography indicated parallel networks, though also overlapping connectivity was found within and across lobules. In conclusion, this thesis provides – in respect to sensorimotor-related regions – the basis for functional subdivisions within anatomical cerebellar lobules using noninvasive imaging. Additionally, it introduces a method to improve extraction of relevant signal from anatomically delineated regions. Based on these results, a connectivity-based parcellation of the entire cerebellum as the potential to substantially expand our understanding of cerebellar functional organization, and enable functional localization of cerebellar damages.
PDF-Datei von FUDISS_thesis_000000099020
Falls Ihr Browser eine Datei nicht öffnen kann, die Datei zuerst herunterladen und dann öffnen.