s-Track: An Intelligent Non-Invasive Stress Tracker for Internet-of-Medical Things

S.P. Mohanty
University of North Texas, Texas, United States

Keywords: Smart Healthcare, Internet of Medical Things (IoMT), Stress Detection, Borderline personality disorder, Deep Neural Network

We propose an IoT-enabled device that can monitor stress level of individuals using deep learning methods, and help doctors to issue advise on food, physical activity, and battle readiness. Psychological stress can be defined as a state of emotional or mental strain due to different circumstances. It is a feel of pressure which affects the physiological parameters in a person wherein the body behaves the way when it is under attack which again releases complex hormones and chemicals like adrenaline, cortisol, etc. Highly impulsive behaviors, inappropriate emotional reactions can lead to a personality disorder called borderline personality disorder (BPD). Stress affects the psychological parameters of the human body like mental and emotional disorders, depression, anxiety, panic attack, and phobias. Monitoring stress levels when a person engages in high intensity physical activities, remain a challenge. We propose the following research for s-Track: (1) Explore machine learning methods for automatic stress monitoring by analyzing body temperature, rate of motion and sweat during physical activity. (2) Explore novel methods for accurate stress level monitoring from the eating patterns. (3) Explore novel methods for automatic stress detection from photoplethysmographic (PPG) signal from the Heart Rate Variability (HRV).