Real -Time Electroencephalography (EEG)-Based Detection of Cognitive Fatigue in Human–Machine Interaction Systems: A Biomedical Engineering Approach

Cognitive fatigue impairs human performance in complex human–machine interaction (HMI) systems, leading to reduced efficiency and potential safety risks. This study presents a biomedical engineering approach for real-time monitoring of cognitive fatigue using electroencephalography (EEG) signals. EEG data were recorded from 30 participants performing sustained attention tasks in simulated HMI environments. Signals were preprocessed using band-pass filtering (0.5–45 Hz) and independent component analysis (ICA) for artifact removal. Fatigue-related features, including theta (4–7 Hz) and alpha (8–13 Hz) band power ratios were extracted. Using a support vector machine (SVM) classifier, cognitive fatigue was detected with an accuracy of 92%, sensitivity of 90%, and specificity of 94% in real-time monitoring. The results indicate a significant increase in theta/alpha ratios correlating with self-reported fatigue scores (Pearson’s r = 0.78, p < 0.01). This framework demonstrates the feasibility of EEG-based adaptive HMI systems for enhancing operator performance, safety, and overall well-being.