My act: an automatic detection of daily physical activity and calorie expenditure using smart phones

Abstract

Several studies show that sedentary behavior poses dangerous signs to health. Patients requiring information of their physical activities can use a tri-axial accelerometer electronic device designed to attach to the waist of users. However, due to the low availability of devices, the applications were limited only to those patients who required special attention and who were to be closely monitored under a controlled environment. Many researchers have recently turned to smart phones to explore the built-in sensors of these phones to obtain accurate physical activity data. Today smart phones are ubiquitous and are equipped with a tri-axial accelerometer. It will be very useful if an activity detection system is implemented in these smart phones. Such system will aid patients with special needs in a special control environment and also others who want to know their daily activities for their own different purposes. This study discusses the use of smart phones in tracking human daily physical activities, divided into four: sleeping, resting, walking, and running. It was shown that the classification resulted in good accuracy in most cases and with reasonable battery consumption. Each activity performed over a period of time was also converted to energy expenditure using MET data published in the compendium of physical activities tracking guide. The following relevant information was also provided: duration of each activity, and step counts and distance obtained while walking and running. This tool could be used to automatically provide information on user’s daily pattern of physical activities. Tests of battery consumption were also shown. Several tests of performance were shown, and although the application depended on factors such as users, phones, phone location, duty cycles, walking patterns, and application settings, it worked very well in most situations after proper calibration to suit the user. Keywords: activity classification; accelerometer; smart phones; pedometer; rehabilitation; energy expenditure (Published: 17 February 2014) Citation: Journal of Assistive, Rehabilitative & Therapeutic Technologies 2014, 2 : 23187 - http://dx.doi.org/10.3402/jartt.v2.23187