A context-aware fitness guide system for exercise optimization in U-health

In this paper, exercise management systems have been introduced, which are generally used to optimize exercise. They create a proper exercise program via an exercise prescription based on the personal physical status of the user. However, exercise programs, generally created at intervals of two weeks to three months, are static and cannot reflect the user's exercise goals, which change dynamically. This paper proposes context-aware exercise architecture (CAEA), which provides an exercise program via a dynamic exercise prescription based on awareness of the user's status. We use sensors of a U-health environment and implement CAEA as an intelligent fitness guide (IFG) system. The IFG system selectively receives necessary parameters as input according to the user's exercise goals. Based on the changes in the user's exercise type, frequency, and intensity, the system creates an exercise program via an exercise optimization algorithm. In this paper, to show the exercise efficiency using the IFG system, we compared a noncontrol group to a control group. An eight-week study was performed comparing the changes of body weight in the two study groups. The study showed that the control group using the IFG system approached the desired body weight 2.57% more closely than the noncontrol group. Since IFG provides a real-time exercise program for users via an exercise optimization algorithm, it enables the user to perform effective and stable exercise according to the user's physical status.