Design and implementation of stress detection system

• Main Author: Chen, Mei Jie
• Other Authors: Vidya Sudarshan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/166310
• Institution: Nanyang Technological University

This study aimed to investigate the body’s representation of stress through ECG signals. By studying stress detection methods through signal processing methods and various forms of machine learning, the study contributed significantly to the eventual development of an application that can prompt stress management techniques for the user when elevated stress levels are detected. This is a significant step in smart wellbeing management applications as it can be integrated into everyday life to prompt intervention the moment one is feeling stressed. A total of 8 participants took part in a variety of stress tests while having their ECG signals recorded throughout the process. This data was processed using both hardware and software methods before it was used to train and test the CNN and VGG19 models. The VGG19 model performed better, achieving a testing accuracy of 88.16%. Stress management techniques were also researched upon to create a tiered intervention scheme. Application mock-ups were completed to demonstrate how these intervention methods could be delivered to the user.