A Comprehensive Survey on Computing Services to Detect Stress and Anxiety Using Smart Devices

Stress and anxiety are part of a modern person's life, which is why they contribute to severe psychological and somatic disorders. The general objective of this work is to develop a live healthcare platform for stress and anxiety recognition and monitoring using machine learning algorithms from wearable biometric data or phone sensors. It is a scalable and privacy guaranteed solution to monitor continuously by processing real-time data only, through the integration of cloud computing with edge computing. Using machine learning, signals like heart rate variability, electrodermal activity, and accelerometry are analyzed to accurately detect signs of stress and anxiety. It is an app that provides users with personalized feedback and interventions to help enhance their mental health. Our evaluation emphasizes the accuracy of detection, scalability, and the performance of real-time processing on the platform, while also addressing challenges related to privacy and computational efficiency. The findings indicate that the platform can be a valuable resource for managing stress and anxiety in real time, highlighting the growing role of wearable technology in the healthcare sector.