Real-Time EEG Dynamics in Females: A Comparative Analysis of Eyes-Open and Eyes-Closed Resting States

Electroencephalography-based detection of cognitive states in elderly populations remains underexplored, particularly in relation to resting-state conditions such as eyes open (EO) and eyes closed (EC). This study investigates and compares EEG signal differences between these two common resting conditions using elderly females (ages 60–80) data sourced from the OpenNeuro platform because they actively affect patterns of brain activity and are essential for evaluating resting-state brain function, particularly when comparing populations or monitoring the course of disease. After applying a band-pass filter (0.5–45 Hz), we extracted 20 statistical features from each EEG segment signal. Multiple classifiers were applied with random forest achieving the highest accuracy: 97% and an EO recall of 99%. outperforming SVM, Logistic Regression, and 1D CNN. These findings suggest that statistical feature analysis and changes in alpha and beta power are promising biomarkers for distinguishing brain activity patterns in this older female population. Our proposed method leverages statistical feature extraction and machine learning to enable detection. The proposed approach allows detection with relatively lower training execution time of 2.2 seconds and highest classification accuracy, demonstrating its potential for clinical and cognitive monitoring in elderly populations.