The World Health Organization states that 50 million people worldwide suffer from epilepsy. Detecting epileptic seizures is commonly done by use of an electroencephalography (EEG) though interpretation of wave patterns relies on neurologists. In order to reduce the potential of human error and implement early detection, Amiri et al. investigated the use of deep learning models to detect epileptic events. To do this, a discrete wavelet transform (DWT) was used to extract particular frequency bands from the plethora of EEG signals in order to create a 1-dimensional vector. Using the compatibility between a 1D Convolutional Neural Network (CNN) and Long Short-Term memory Network (LSTM), features and patterns were able to be identified as seizures with high accuracy in three different data sets. The three publicly available data sets of EEGs were: University of Bonn (BONN), Children’s Hospital Boston-Massachusetts Institute of Technology (CHB-MIT), and TUH EEG Seizure Corpus (TUSZ). The accuracy amongst each data set using the AI model was 97.24%, 96.94%, and 94.32% respectively.