Published Paper Details:

When we encounter difficult or taxing circumstances, or stressors, our bodies and minds go into a hyperarousal state, commonly known as stress. Chronic illnesses including anxiety, depression, and cardiovascular disease can develop in people who are exposed to various stressors for an extended period of time, particularly when these stressors happen at the same time. In order to avoid these long-term impacts, it is essential to recognise stress early on, and continuous monitoring provides a viable option. One possibility for personal stress monitoring is the use of wearable devices that can gather data continuously and in real-time, allowing users to keep tabs on their stress levels as the day progresses. This study provides an extensive analysis of several approaches to stress detection that make use of wearable sensors in conjunction with machine learning methodologies. Wearable sensors including electrocardiograms (ECGs), electroencephalograms (EEGs), and photoplethysmography (PPGs) are reviewed, along with the many settings when stress is tracked, such as while driving, learning, or working. Various stress detection approaches were examined in this review, with an emphasis on the stressors, procedures, outcomes, benefits, drawbacks, and difficulties encountered by each research. The area of stress monitoring is anticipated to benefit from these findings in the future. In addition, the study suggests a multimodal stress detection system that combines various kinds of wearable sensors with deep learning algorithms. This could be a great step towards better real-time stress detection that is both accurate and adaptive, which would be great for mental health management and healthcare prevention.