The aim of this study was to investigate the validity of the ActiSmile (AS), a physical activity feedback device. Twenty-five participants (age: 36.5 ± 14.3 y) performed nine different activities including standing, running, cycling, playing badminton, sweeping the floor with a broom, and several types of walking for 3 min 25 s each. Each participant received two AS devices, one attached at the hip (AShip) and one carried in the trouser pocket (ASpocket). The AS classifies each min as inactivity, walking, or running. Recognized activity time was compared to observed activity. Indirect calorimetry was used to validate the energy expenditure (EE) estimated by the AS. The AShip correctly recognized 100% of standing as inactivity, between 97 and 100% of different walking modes and 97% of running. The ASpocket correctly recognized 100% of standing as inactivity, between 88% and 96% of different walking modes, and 96% of running. Good estimates of EE were reached for moderate walking (bias: < 1.1 kcal/min). The AS is therefore valid for recognizing both walking at individual intensities and running. It provides good group estimates of the EE during moderate walking. Therefore, the AS has the accuracy to be an effective tool for intervention studies.
© Copyright 2013 International Journal of Computer Science in Sport. Sciendo. Alle Rechte vorbehalten.