Assessing Brain Pathology Based on Large-Scale Interaction Patterns of Spontaneous Brain Activity

This thesis introduces two methodological advances in studying the large-scale organization of spontaneous activity of the human brain and in addition provides a translation of these methods to the clinical realm. First, a generic, data-driven framework for identifying alterations in brain wide functional connectivity is presented and tested in a patient population suffering from multiple sclerosis. A correlation between fMRI derived measures of functional connectivity and behavior is established. In the face of a strong cognitive decline and the severe disintegration of the central white matter, specific networks increased their functional connectivity. This observation challenges the prevailing view on how functional connectivity indicates the integrity of brain networks and holds important implications for resting state fMRI investigations of brain diseases as well as theoretical and modeling studies of large-scale cortical dynamics. In addition, the observed correlation between behavior and connectivity has a high relevance for medical settings as it may potentially serve as an objective proxy and biomarker for the degree of impact of multiple sclerosis on functional processing and brain network organization. Second, a novel method for measuring functional connectivity based on noninvasive electrophysiological data (M/EEG) is established. The method overcomes a fundamental problem for deriving measures of interaction from noninvasive electrophysiological data by a novel technique of phase orthogonalized power correlation. By applying the method to MEG recordings of spontaneous brain activity in a large cohort of healthy participants, a characterization of the large-scale organization of spontaneous neuronal oscillatory activity could be provided. Spontaneous fluctuations of oscillatory activity are spatially organized into frequency dependent and functionally specific correlation structures. Thus, the method provides a new tool for understanding large-scale brain organization based on neuronal oscillatory activity. The combined application of both methods to MEG recordings of spontaneous brain activity of a group of congenitally blind participants further highlights the strong potential of the approaches by demonstrating specific alterations in the organization of spontaneous oscillatory neuronal activity in the blind. Occipital areas of the blind are found to exhibit known oscillatory signatures of active functional processing. The phase of slow oscillatory processes (delta range ~2 Hz) predicted the amplitude of faster rhythms (gamma range ~90 Hz). In addition the formerly visual areas of the blind exhibited frequency specific connectivity (beta range ~25 Hz) with prefrontal sites. These findings suggest that formerly visual areas serve non-visual processing during unconstrained mental activity in the blind and that a specific occipito-prefrontal pathway may underlie the reintegration of the occipital processing resources into cortical networks. Overall, the application of the two novel methods to clinical populations demonstrates the unique access that spontaneous brain activity provides to pathophysiological principles of brain organization. The advances described in this thesis bridge methodological and conceptual gaps in the study of spontaneous activity of the human brain.

Author

• Hawellek, David J.

Identifier

hdl:10900/49717

OAI record oai:identifier

https://publikationen.uni-tuebingen.de/oai/request?verb=GetRecord&metadataPrefix=mets&identifier=oai%3Apublikationen.uni-tuebingen.de%3A10900%2F49717