Home // International Journal On Advances in Internet Technology, volume 11, numbers 3 and 4, 2018 // View article
Authors:
Toon De Pessemier
Enias Cailliau
Luc Martens
Keywords: Activity Recognition; Wearable; Health Information; Recommendation; Personalization
Abstract:
Wearables are often equipped with an accelerometer and heart rate sensor. However, the accuracy of the generated heart rate measurements is still unclear. This paper evaluates heart rate measurements during various physical activities performed by test users and compares three types of wearable devices: the specialized sports device with chest strap, the fitness tracker, and the smart watch. Consistent heart rate measurements are obtained with all wearables during activities that require no or very little wrist movements, such as sitting in a chair, cycling, walking, or even squat exercises. In contrast, wearables worn around the wrist (smart watches and fitness trackers) and sport devices worn around the chest measure significantly different heart rates during activities that require a lot of movement of the wrist, such as dumbbell biceps curl and push up exercises. These movements of the user's wrist were measured using the accelerometer of the wearable, and allow the detection of repetitions of a physical activity with a typical movement pattern, such as a dumbbell biceps curl. Based on accelerometer and heart rate data, a user profile is created for a rule-based filter to generate personal recommendations for physical activities. A mobile app demonstrates that heart rate measurements and activity recognition can be used to assist and guide users during workouts.
Pages: 136 to 146
Copyright: Copyright (c) to authors, 2018. Used with permission.
Publication date: December 30, 2018
Published in: journal
ISSN: 1942-2652