The aim of this project is to automate the EEG signal classification task and the detection of epileptic seizures using machine learning and deep learning models. we used three different datasets, one coming from the University of Bonn in Germany the second is available on https://www.isip.piconepress.com/ , the last dataset comes from a private source so for reasons of confidentiality it would not be shared.
Dataset description: The complete data consists of five sets (A to E), each containing 100 instances in one channel. Sets A and B consist of EEG signals recorded by five healthy volunteers while in a relaxed and awake state, eyes open (A) and eyes closed (B), respectively. C, D and E sets were recorded in five patients. EEG signals in group D were collected in the epilepgenic zone. Set C was recorded from hippocampus formation of the opposite hemisphere of the brain. The C and D assemblies consist of EEG signals measured during crisis-free (interictal) intervals, while EEG signals in the E-package were only recorded during the activity of the crisis (ictal). Each file is a record of brain activity for 23.6 seconds. The time series is sampled in 4097 data points. Each data point is the value of the EEG record at a different time. So we have a total of 500 individuals with each 4097 data points for 23.6 seconds.
Download link: https://www.ukbonn.org/epileptologie/ag-lehnertz-downloads/
This is a subset of TUEG that contains 100 subjects epilepsy and 100 subjects without epilepsy, as determined by a certified neurologist. The data was developed in collaboration with a number of partners including NIH. \n
Isip dataset: The TUH EEG Epilepsy Corpus (TUEP) download link:
https://www.isip.piconepress.com/projects/tuh_eeg/html/downloads.shtml