Psychological stress detection using deep convolutional neural networks

Abstract

Many psychological motives and life incidences are answerable for inflicting psychological stress. It’s the primary reason for inflicting many cardiovascular diseases. This paper presents a study on psychological stress detection with the aid of processing the Electrocardiogram (ECG) recordings using Convolutional Neural Networks (CNN) as a classification approach. The main purpose of this study was to trace students under stress during their oral exam. A dataset of ECG recordings of 130 students was taken during the oral exam. A customized CNN is designed for stress recognition, and it has achieved 97.22% and 93.10% stress detection accuracy for filtered and noisy datasets, respectively.