Date of Award

Summer 8-31-2017

Document Type

Thesis

Degree Name

Master of Science in Biomedical Engineering - (M.S.)

Department

Biomedical Engineering

First Advisor

Bharat Biswal

Second Advisor

Sergei Adamovich

Third Advisor

Xiaobo Li

Fourth Advisor

Xin Di

Abstract

Resting State fMRI is used extensively for analysing brain regions. fMRI is also used to determine task induced activations from cerebral networks. This study involves both rest and task activation data and implements statistical methods applied on Blood Oxygenation Level Dependent (BOLD) signals. The conventional task based designs are very specific and may not reflect day to day activities. The two task conditions in this study are watching a movie called INSCAPE with low cognitive load and segments of contextual Hollywood movies under continuous and natural conditions.

The aim is to investigate individual and group level Functional Connectivity changes between REST and the two conditions - INSCAPE and MOVIE. The hypothesis is that the functional networks of INSCAPE emulate Resting State networks more closely than traditional MOVIES if the consistent group level effect is regressed out. Voxel wise regression is used on individual subjects across sessions and between subjects per session. Network Wise Correlation Coefficient Maps were also generated using Power‟s 264 ROIs. Group level concatenated (cICA), for all conditions, and tensor ICA (tICA) maps for only natural viewing conditions were also generated. A paired t-test was performed between INSCAPE vs. REST and MOVIE vs. REST conditions. The results show that ROIS with significant differences were considerably lower for INSCAPE than MOVIE for an FDR corrected arbitrary threshold of p < 0.0005.

Share

COinS