Content area
This dissertation quantifies changes in brain network structure following traumatic brain injury (TBI) using diffusion magnetic resonance imaging (dMRI). It builds structural brain networks from rodent and human data to explore: 1) Differences in brain network reconstruction: The study finds that different methods for building brain networks can lead to different understandings of how the brain reorganizes after injury. For example, we see differences in standard network measures, such as the degree distribution, depending on if end-to-end or pass-through networks are created. These differences highlight how the choice of method can change research findings. 2) Hidden lesions detection in rodent models: The research successfully uses tractography networks to find and measure hidden lesions in rodent models, showing a reduction in fiber density on the injured side compared to the uninjured side. This approach enhances the detection of structural damage that is not always seen with usual imaging methods. 3) Variability in human brain network evolution post-TBI: The study shows differences in how brain network structures change after TBI among individuals. Some people's networks recover or restructure well, while others get worse over time. This shows different ways people respond to injury and recovery. In conclusion, this dissertation shows that how we model brain networks can greatly impact what we learn from them, especially in TBI cases. It stresses the importance of using the appropriate network building methods for accurate assessment and suggests moving towards methods that consider differences within groups. This work suggests a move towards personalized analyses, calling for new methods that are appropriately tailored to meet the needs of TBI patients.