Multi-modal stress detection

Abstract

Feeling stressed occurs when our body and mind respond to challenging situations, and is a common aspect of life. However, if stress persists for a long time, it may negatively impact our well-being and can lead to various diseases. Many people find it hard to identify their feelings and filling out questionnaires can take time. Hence, we’re trying to build a solution, to find out if a person is stressed or not based on their physiological signals. We do so by tracking EDA and Heart Rate Variability (HRV) signals of the person using a wrist-worn device. We can use these signals, process them, and apply machine learning models to predict the level of stress in a user. If the user is stressed, we’ll inform them about it, so that they can take appropriate actions in time.