Objective measures, sensors and computational techniques for stress recognition and classification: a survey

Area of Science:

• Biomedical Engineering
• Computer Science
• Psychology

Background:

• Stress is a significant societal concern impacting individual well-being and societal functions.
• Stress research offers benefits in personal performance, learning, productivity, and public health.
• Developing methods for regular, unobtrusive stress monitoring is crucial for at-risk populations.

Purpose of the Study:

• To survey sensors for measuring stress, focusing on non-invasive and unobtrusive options.
• To investigate computational techniques for modeling and analyzing stress data.
• To introduce and define the concept of 'computed stress' for regular monitoring.

Main Methods:

• Review of existing literature on stress measurement sensors.
• Analysis of computational modeling techniques including Bayesian networks, artificial neural networks, and support vector machines.
• Focus on sensors that enable continuous monitoring without impeding daily activities.

Main Results:

• Identification of various non-invasive and unobtrusive sensors suitable for continuous stress monitoring.
• Overview of computational techniques for sensor fusion and automated stress data analysis.
• Discussion of the potential for 'computed stress' monitoring in diverse applications.

Conclusions:

• Non-invasive sensors and advanced computational models are key to effective stress monitoring.
• Further research in computational stress modeling can enhance individual and societal well-being.
• The concept of 'computed stress' offers a new avenue for proactive stress management.