Detection of epileptic seizure disorder using EEG signals

Advances in cognitive neuroscience and brain mapping have begun to give us the capacity to communicate directly with the human brain. This capability is enabled by the use of sensors that can track some of the physical processes in the brain. BCI is a communication system that can obtain signals and incorporate real-time seizure detection/prediction, as well as therapies including electrical stimulation to suppress seizures. Epilepsy is a brain dishevelment in which the peripheral nervous system malfunctions, resulting in loss of consciousness and convulsions. Electroencephalography (EEG), which detects ionic currents flowing between neural tissues and tracks variations in input voltages through electrodes across the patient's scalp, is an important tool for measuring brain activity of the epileptic patients. Identification of EEGs necessitates a close inspection by a practitioner as well as a significant amount of time and effort. This article discusses an automatic grouping of EEG signals for the identification of epileptic seizures using machine learning and deep learning methods based on wavelet transform and mathematical pattern classification. The model is divided into preprocessing of the data, feature extraction with the use of wavelet transforms, and quadratic classification using ML and DL classifiers. The extracted features were fed to different classifiers such as naive Bayes, K-nearest neighbors, artificial neural networks, and SVM. As there is automatic feature extraction in deep learning, the models such as CNN and RNN are used for classification. The presented methods will give you a brief explanation and classification of epileptic seizure detection and the problems faced by the clinicians. Future success can be determined by the advancement of electrode arrays or microelectrodes, as well as improvements in the early detection and variable-encoding algorithms.