Exploring Multi-Class Stress Detection Using Deep Neural Networks

This study uses Heart rate variability (HRV) as a biomarker to analyze the complex link between ongoing stress and its possible physical effects. To deal with difficult situations, people naturally feel stressed but when it lasts for a long time, stress can very badly affect the mental health. Depression, nervousness, and trouble sleeping are all common illnesses that can be caused by stress. Several advanced machine learning methods are used in this study to create a multi-class stress recognition model. The proposed stress detection model has achieved high accuracy, reaching 96.87% in stress identification using the SWELL-KW dataset. Leveraging this dataset, the study explores HRV traits as potential stress biomarkers, employing advanced machine learning methods including 1D CNN, 3D CNN, LSTM, GRU, and combinations such as LSTM + GRU + RNN. The main aim of this research work is to enhance the accuracy and utility of stress detection methods by investigating HRV traits as possible stress biomarkers. A key part of knowing how human bodies react to stress is being able to tell the difference between heart rate and HRV, especially when looking at the changes in RR intervals. Finally, the results of this study could help us learn more about how to recognize stress and come up with better ways to treat mental health problems that are caused by stress.