A Biofeedback-based Physical Activity Advisory System

Physical inactivity, a phenomenon on the rise in numerous countries, has gained global attention because of its negative effects on humans' physical wellness. It represents a stumbling block in the way of living a healthy lifestyle. Recent statistics of World Health Organization (WHO) ranked physical inactivity as the fourth leading risk factors for adults' mortality all over the world [1]. Also, physical inactivity is considered as one of the most prominent contributing factors in several severe diseases such as breast and colon cancer, diabetes and many heartrelated diseases [1]. Therefore, improving daily physical activity levels is an urgent societal goal in order to tackle the physical inactivity problem. Achieving such challenging goal requires addressing the factors that affect adults’ physical activity. In fact, there are many factors that lead to physical inactivity such as the busy lifestyle, lack of awareness regarding required physical activity levels and other environmental factors. Physical activity advisory systems can be seen as a promising solution for the inactivity problem. In order to enhance their effectiveness, these systems must take into account most of the factors previously mentioned. In this thesis, we aim to provide a method to promote the increase of daily physical activity levels by leveraging biofeedback and context awareness features. In order to achieve this purpose, we design and develop an algorithm that provides a user with personalized physical activity advice. This advice increases the user's awareness through the use of calories expenditure. To add a context awareness component to our algorithm, we propose an extension of the Ubiquitous Biofeedback (UB) Model [2]. We believe that combining the biofeedback feature with context awareness component would make the system sensitive to the user’s status and thus increase the chances of her or him following it. This advice represents the daily-recommended amount of physical activity for maintaining healthy lifestyle according to [3, 4]and other international organizations' recommendations. In order to prove the concept of the proposed algorithm and extended UB Model, we design and develop a system called "CAB". It is a context aware biofeedback system that tracks user's physical movement and estimates the amount of calories burnt to provide the user with a personalized physical activity advice that considers user's current status, preferences and surrounding environmental context. The system utilizes a biofeedback sensor and a smart phone in order to provide the personalized advice that is