Applied Research

Classifying Stress Levels via Electroencephalography: A Systematic Literature Review

---

Abstract

The classification of stress levels using electroencephalography (EEG) is based on the understanding that prolonged or improperly managed stress can contribute to a variety of physical and mental health issues, including anxiety disorders, depression, sleep disturbances, and cognitive impairments. The goal of stress classification utilizing EEG is to identify and distinguish an individual’s stress level by examining their brain activity patterns. In this study, the systematic literature review (SLR) methodology was adopted, relying on secondary data derived from prior research published between 2014 and 2022, which are related to the utilization of EEG for stress level classification. The execution of the SLR adhered to the Preferred Reporting Item for Systematic Review and Meta-Analysis guidelines, encompassing three primary phases, namely, formulation of research questions and article search planning, sourcing, and scrutinizing reference materials, and synthesizing findings and drawing conclusions. Following this, a sequence of analytical processes was carried out, including spectral analysis, temporal analysis, pattern analysis, and machine learning techniques. The exploration was thorough searches within the library of the Scopus journal database (http://scopus.com), with the relevant datasets identified via http://figshare.com. Out of the 42 identified articles, distinct themes emerged: nine articles focused on feature selection, 19 articles delved into classification methods, 11 articles dealt with EEG signal processing, and three articles centered around stress detection robotics. Among these studies, 17 employed individual-specific data, whereas 25 utilized datasets. Notably, only five studies employed the hybrid feature selection algorithm, with one of them incorporating the feature-based common spatial pattern (FBCSP) algorithm, long short-term memory network (LSTM), and fast Fourier transform. The outcomes of the conducted analysis facilitated the creation of a comprehensive mind map, which encapsulated the findings of the SLR on EEG-based stress level classification. This application of the SLR approach proved effective in recognizing commonly employed classification methodologies, evaluating the caliber of preceding research, validating outcomes, and suggesting prospective research directions. The recommendations arising from this study pertain to the innovation and exploration of novel methodologies to further advance the field.

Keywords: Electroencephalography, Classification, Literature Review, Stress, Classification, Literature Review, Stress

How to Cite:

Setyorini, S., Zaeni, I. A. & Elmusyah, H., (2026) “Classifying Stress Levels via Electroencephalography: A Systematic Literature Review”, The Journal of Technology, Management, and Applied Engineering 1(1). doi: https://doi.org/10.31274/jtmae.16943