Epileptic Seizure Detection from EEG: Targeting Specific Brain Regions with Machine Learning Approaches

The electroencephalogram (EEG) signal plays a vital role in diagnosing epilepsy. The EEG recordings of the ambulatory recording systems generate very lengthy data, and the detection of epileptic activity requires an expert to analyze the entire length of the EEG data. The traditional methods of analysis being tedious, many automated diagnostic systems for epilepsy have emerged in recent years. This study aims to determine the most effective brain region in seizure detection so that the use of unnecessary electrodes and other time-consuming analyses can be reduced. Data from three distinct brain regions-central, frontal, and occipital are utilized. The denoised EEG signals are utilized to extract the time domain, frequency domain, and entropy-based features. Later, the features are used to develop machine learning algorithms for classifying the signals. Upon analysis with different combinations of features, the best accuracy has been achieved from the frontal region of the brain using the boosted tree algorithm with an accuracy of 83.5%. This approach leverages the diverse information provided by each region to enhance detection reliability. In the future, further analysis can be done targeting the frontal region for improving the performance.