A Seizure Prediction Model Based On A Fractional-Order Neural Network

Epilepsy is a common neurological disorder, and its prediction has important implications for the treatment and quality of life of patients. In this paper, we propose a seizure prediction model based on a fractional order neural network. The dynamic behavior of the fractional Hopfield neural network simulated the neuronal activity of the hippocampus and prefrontal cortex and combined the features of real EEG signals (EEG) data. Numerical simulations show that fractional neural networks can effectively capture the memory effects and long-range dependence of neuronal activity, providing a new theoretical tool for seizure prediction.