One-third of epilepsy patients are drug-resistant patients. 30-40% of the patients who undergo resection surgery still are not rid of their seizures. We hypothesize that structure and function correlations will demonstrate vital network edge connections in seizure activity. Using this information, we are working toward utilizing computational diffusion models to predict seizure spread from a predetermined onset zone and better understanding the brain's connectivity. This will enable more precision in the determination of the focal point of seizures for surgery, enabling more effective and favorable resection outcomes.
My personal contributions to this project include
1
Streamlining and optimizing algorithm that processes structural and functional adjacency matrices and calculates structure-function correlation
2
Machine-learning seizure propagation detection algorithm using univariate channels (per electrode channel) to better understand how seizures diffuse throughout the brain.
