Comprehensive Review on Stress Detection using EEG Signals and Machine Learning Techniques.

Main Article Content

Amol S. Chaudhari, Hemang Shrivastava

Abstract

Stress significantly impacts mental and physical health, making accurate detection essential for enhancing productivity and well-being. Electroencephalography (EEG) has emerged as a pivotal tool for stress analysis, offering a non-invasive, cost-effective, and temporally precise method to evaluate brain activity. This paper reviews advancements in stress detection using machine learning (ML) and deep learning (DL) techniques applied to EEG data. Traditional stress detection methods, such as self-reports and subjective surveys, face limitations in reliability and scalability. In contrast, ML and DL models, including Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNNs), and hybrid architectures, demonstrate superior feature extraction and classification accuracy. Studies have reported classification accuracies exceeding 95% using advanced signal processing and hybrid frameworks. However, challenges remain in optimizing computational efficiency, reducing algorithmic complexity, and validating models on diverse, real-time datasets. This review highlights the potential of innovative feature extraction techniques and hybrid ML-DL models to address these gaps, paving the way for robust, scalable, and real-time stress detection systems. By overcoming current limitations, future research can significantly contribute to mental health management and preventive care strategies. This paper provides a comprehensive overview of recent advancements and outlines future directions in EEG-based stress detection research.

Article Details

Section
Articles