The Human Connection: A Graph Theory Approach to Depression and fMRI Resting State Network Analyses
Open Access
- Author:
- Meeks, Kathleen
- Graduate Program:
- Neuroscience
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 26, 2023
- Committee Members:
- Charles Geier, Thesis Advisor/Co-Advisor
Frank Gerard Hillary, Committee Member
Sonia Cavigelli, Professor in Charge/Director of Graduate Studies
Dahlia Mukherjee, Special Signatory
Zachary Fisher, Committee Member - Keywords:
- Anhedonia
Graph Theory
Depression
Functional Connectivity
Resting-State fMRI - Abstract:
- Anhedonia, or the inability to feel pleasure, is a key feature of major depressive disorder and is associated with both treatment resistance and suicide attempts. Previous resting state fMRI analyses have identified networks with both increased and decreased functional connectivity in patients with major depressive disorder compared to healthy controls. However, this is one of the first studies to use graph theory metrics to analyze resting state data as it specifically relates to anhedonia in depressed and healthy participants. Ten minutes of fMRI resting state data was preprocessed for each of the 45 participants—25 with major depressive disorder and 20 healthy controls. Graphs and metrics were then calculated at both the group level and individual level for 139 brain regions derived from the Power 264 parcellation. We further characterized groups and individuals using both isolated networks and clinically relevant network pairs. We found that average path length was significantly lower for MDD overall as well as within and between the default mode network and task-positive networks. Mean degree significantly predicted depression symptom severity and modularity predicted anhedonia symptom severity. These findings highlight the potential of biological graph metrics to objectively quantify brain activity in specific regions that reflects the severity of clinical symptoms. Such a tool not only clarifies the pathophysiology of anhedonia, but may also inform precision diagnosis, treatment selection and outcome prediction for mood disorders.