Document Type

Thesis

Date of Award

3-2025

School/College

College of Science, Engineering, and Technology (COSET)

Degree Name

MS in Computer Science

Committee Chairperson

Dr. Muhammad Khan

Committee Member 1

Dr. Alaaeldin Sleem

Committee Member 2

Dr. Khaled Kamel

Committee Member 3

Dr. Momoh Yakubu

Abstract

Understanding the connectivity of neurons within the human brain is essential for advancing our knowledge of cognitive processes, sensory perception, and neurological disorders. This thesis leverages the Connectome Workbench, an open-source, advanced visualization and discovery tool developed as part of the Human Connectome Project (HCP), to model and analyze functional connectivity in the human brain. Specifically, this study investigates the connectivity patterns between the primary visual cortex (V1) and higher-order visual regions (V2, V3) across different age groups using resting-state functional MRI (rs-fMRI) data. The study hypothesizes that younger adults (18–35 years) will exhibit stronger functional connectivity between V1 and higher-order visual regions compared to older adults (36+ years), reflecting more efficient visual information processing. To test this, the study employs advanced computational algorithms to ensure precise and meaningful connectivity analyses. Brain images are registered to a common anatomical space using affine and non-linear transformations, while spherical registration aligns cortical surfaces for accurate cross-subject comparisons. Independent component analysis (ICA) is applied to extract intrinsic connectivity networks, enhancing the identification of functionally coherent regions. Data from the HCP is analyzed using graph-based connectivity models, where brain regions are represented as nodes and functional connections as edges in an adjacency matrix. The use of sparse matrix representations optimizes computational efficiency, allowing the study to process high-resolution visual cortex parcellations. Connectivity patterns are further examined using graph-theoretical measures, such as centrality metrics to identify critical hub regions, modularity analysis (via Louvain clustering) to detect functional sub-networks, and shortest-path algorithms to assess network efficiency. These computational techniques provide a robust framework for quantifying interregional communication and assessing age-related differences in functional connectivity. Results reveal significant differences in connectivity patterns between age groups, supporting the hypothesis that age-related decline in visual cortex connectivity may contribute to decreased visual processing efficiency. The findings enhance our understanding of age-related neural changes and provide insights for developing targeted interventions to preserve visual function in aging populations. The integration of advanced neuroimaging tools, such as the Connectome Workbench, demonstrates the power of surface-based and graph-theoretical analysis in capturing fine-grained details of cortical connectivity, setting a foundation for future research in computational neuroscience, neurodegenerative disease modeling, and visual system aging.

Download

Share

COinS