Fig. 1

A displays a schematic of the EEG signal, detailing how waveforms vary in different states of brain activity. The figure also shows the differences in EEG waves between normal and abnormal states; B is a chart of time-domain and spectral analysis, with a focus on signal intensity in alpha and beta waves; C presents a channel correlation analysis in different patients’ epilepsy conditions, showing the correlation of EEG signals between different brain region channels during epileptic seizures in various patients. This is represented through color coding and line thickness to indicate the strength of correlation between channels, and how this correlation changes from normal to seizure states; D depicts the process of dimension reduction using principal component analysis (PCA), explaining how PCA is utilized for dimension reduction in EEG data. It shows the transformation of data from a high-dimensional space to a low-dimensional space, and the key information preserved in this process; E presents our model structure, illustrating the architecture used for analyzing EEG signals, including various processing layers, network architecture, and its outputs; Finally, F illustrates the implementation of the classification process, showing how the model classifies EEG signals, such as distinguishing between normal and abnormal signals