Automated EEG Pathology Detection Based on Significant Feature Extraction and Selection

Neurological diseases are a significant health threat, often presenting through abnormalities in electroencephalogram (EEG) signals during seizures. In recent years, machine learning (ML) technologies have been explored as means of automated EEG pathology diagnosis. However, existing ML-based binary classification methods largely focus on extracting EEG-related features, which may lead to poor performance classifying by overlooking potentially redundant information. this paper, we propose novel Kruskal–Wallis (KW) test-based framework for detection. Our first divides data into frequency sub-bands using wavelet packet decomposition and then extracts statistical characteristics from each selected coefficient. Next, the piecewise aggregation approximation technique is used obtain aggregated feature vectors, followed KW test methodology select features. Finally, three ensemble classifiers, random forest, categorical boosting (CatBoost), light gradient machine, classify extracted features normal or abnormal classes. proposed achieves an accuracy 89.13%, F1-score 87.60%, G-mean 88.60%, respectively, outperforming other competing techniques same dataset, shows great promise