EpiChange: Quantifying longitudinal changes after epilepsy surgery (360G-Wellcome-210109_Z_18_Z)

Resective surgery for epilepsy, where the part of the brain thought to cause seizures is removed, leads to seizure freedom in around 70% of patients 1 year post-surgery. This falls to around 50% at 5 years post-surgery. It is not fully understood why surgery only works initially for some patients, and why this falls over time post-operatively. Surgery has a substantial immediate impact on brain structure, however, the long-term impact of surgery on brain dynamics is poorly understood. In order to make progress in this area we will perform a retrospective analysis of longitudinally acquired electroencephalographic (EEG) data. EEG recordings were made pre-operatively, and post-operatively in 76 patients for up to 5 years. Using univariate and multivariate data analysis, in conjunction with machine learning, we will learn how brain dynamics change after surgery, and if this change relates to outcome. Crucially, we will attempt to identify which factors in brain dynamics correlate with seizure relapse, even years after surgery. If successful, this will pave the way to a larger project where changes can be reverse engineered to give predictions of post-operative decline using pre-operative data. Long-term, this research has implications for other disorders involving longitudinal decline following structural brain damage.