Text

theses

Resting state functional MRI (fMRI) studies have demonstrated temporal correlation across physically distant voxels (or regions) in functionally related regions and that they are dominated by low frequency fluctuations in the range of approximately 0.01 – 0.1 Hz. While these studies have been widely replicated, due to hardware limitation the sampling rate of an fMRI machine has been limited to about 1 data point every 2 seconds resulting in a Nyquist sampling rate of 0.25 Hz, have focused on fMRI signal <025 Hz. Yet various electrophysiological measurements like EEG, LFP, and MEG acquire data at much faster rate at up to 200 times points every second and study neuronal fluctuations in range from 1 ~ 100 Hz. In addition, to be limited by the lower sampling rate of fMRI, resting state fMRI studies, are primarily focused on sub segment (0.01-0.1 Hz) of the whole frequency bands (0-0.25 HZ), due to. The goal of the current dissertation is to utilize recent advancements in fMRI signal acquisition techniques, which can acquire 1 data point in 0.5 seconds, to study functional integration between brain regions in during resting state fMRI in higher frequency BOLD fluctuations. In order to achieve this goal, we obtained resting state fMRI data acquired from healthy subjects at higher sampling frequency of 1.5 Hz as well as resting state fMRI data acquired from schizophrenic patients at sampling frequency of 0.5 Hz from open-access data repositories. Using this open access fMRI data, we performed three distinct studies to investigate frequency specific differences in resting state functional connectivity. In the first study, we quantified RSFC across five different frequency bands. We implemented two of the most widely used methods: independent component analysis and seed based correlation to estimate RSFC across frequency bands. Commonly known RSNs such as the default mode, the fronto-parietal, the dorsal attention and the visual networks were consistently observed at multiple frequency bands. Significant inter-hemispheric connectivity was observed between a seed and its contralateral brain region across all frequency bands, though overall spatial extent of seed based correlation maps decreased in slow-2 and slow-1 frequency bands. These results suggest that functional integration between brain regions at rest occurs over multiple frequency bands and RSFC is a multi-band phenomenon. These results also suggest further investigation of BOLD signal in multiple frequency bands and related changes in whole brain network topologies. In lieu of the results from the first study, in the second study we investigated changes in whole brain network topologies associated with changes in frequency bands based RSFC. We performed graph theory analysis on whole brain RSFC in five distinct frequency bands to study the whole brain network architecture. We observed significant differences in local connectivity properties across frequency bands and corresponding changes in network hubs, modularity and small-world network index. The brain network topologies at all the frequency bands showed small-world topologies though, RSFC network at slow-4 and slow-5 networks showed significantly higher small-world indices compared to that of slow-1 and slow-2 networks. Lastly, due to differential power distribution of BOLD signal across resting state networks observed during the first project, we studied changes in BOLD signal power in clinical populations. In this regard, we studied disruption of BOLD signal power in various frequency bands in schizophrenia. We observed significant increase in frontal cortex power in psychosis patients compared to healthy controls across slow-2, slow-3 and slow-4 and opposite effect was observed in posterior brain regions, where controls showed increased BOLD signal power compared to psychosis patients. By performing these three coherent studies, we investigated frequency specific changes in RSFC and their disruption in psychosis patients, implying neurocognitive importance of resting state BOLD signal in higher frequency bands (>0.1 Hz).

ETD_6927

doi:10.7282/T3WH2S2H

Ph.D.

Includes bibliographical references

by Suril Gohel

ETD doctoral

The author owns the copyright to this work.