Recognition of Secondary Tonic-Colonic Epilepsy Seizures using Machine Learning

The most common form of epilepsy is erratic seizures brought on by excessive neuronal activity that is not efficiently communicated to the receiving end of the neuron. This kind of epilepsy is known as secondary tonic-clonic epilepsy when the brain stem and subcortical brain areas are directly implicated in the malfunctioning of the network. Since some individuals may not show any symptoms at all, diagnosing tonic-clonic epilepsy may be difficult in the medical field. Our study analyses patient scans and uses machine learning and deep learning methods to detect epilepsy early on. The random forest is performing better with an accuracy of 96.45% than the other implemented algorithms as decision tree and logistic regression on the Epileptic Seizure Recognition dataset. Hence recognizing rapid neuronal hyperactivity enables authorized medical professionals to intervene promptly.