Physical activity recognition using a wearable accelerometer : new perspectives for energy expenditure assessment and health promotion

A physically active lifestyle is important for living a healthy life. However, defining how much physical activity is necessary for promoting health requires objective and valid instruments for monitoring physical activity and the related physiological outcomes. This thesis showed that accurate recognition of physical activity types, like walking, cycling, and running was feasible using a wearable accelerometer. This method improved daily estimations of energy-expenditure, and it showed that the cycling-, walking- and sitting daily duration determined the subjective activity level. Furthermore, specific activity types characterizing the behavior of obese subjects played a key role for protecting against cardiovascular diseases.