Public Health Initiative: Enhancing Stress Evaluation using EEG Data

Abstract

Study proposes an innovative public health that uses EEG data processing to improve stress assessment. Stress is a widespread health issue that affects people in a variety of demographics. Stress evaluation techniques based on traditional self-reporting are biased and subjective. EEG records patterns of brain activity linked to stress reactions and offers an objective measurement. In this study, we proposed Social Spider Advanced Bidirectional Long ShortTerm Memory (SS-ABiLSTM), an innovative method for improving stress evaluation by integration of EEG data. Initially, EEG data is gathered, and then preprocessed the EEG data using min-max normalization to remove inaccurate records from a dataset. To extract features from the preprocessed data using the Discrete Wavelet Transform (DWT), this reliably extracts frequency-domain information from the data. The most discriminative characteristics for stress assessment to select feature using Recursive Feature Elimination (RFE). By combining these methods, stress assessment models that utilize EEG data have high accuracy. Utilizing the application of SS-ABiLSTM, that addresses stress prevention and management while also improving the ability to interpret EEG data in public health. The proposed methods are to evaluate F1-score (90.1%), precision (88.2%), recall (91.8%), and accuracy (98.97%). The accuracy of the proposed technique is superior to the existing techniques. Finally by using EEG data for stress evaluation in public health shows potential for more individualized care and better mental health results.