Classification of physical activities and sedentary behavior using raw data of 3D hip acceleration
The purpose of this study was to develop and validate an algorithm for classifying physical activity (PA) classes and sedentary behavior (SED) from raw acceleration signal measured from hip. Twenty-two adult volunteers completed a pre-defined set of contr
- PDF / 867,274 Bytes
- 4 Pages / 595.276 x 790.866 pts Page_size
- 27 Downloads / 195 Views
esearch Unit of Medical Imaging, Physics and Technology (MIPT), University of Oulu, Oulu, Finland 2 Polar Electro Oy, Kempele, Finland 3 Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland 4 Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland
Abstract—The purpose of this study was to develop and validate an algorithm for classifying physical activity (PA) classes and sedentary behavior (SED) from raw acceleration signal measured from hip. Twenty-two adult volunteers completed a pre-defined set of controlled and supervised activities. The activities included nine daily PAs. The participants performed PA trials while wearing a hip-worn 3D accelerometer. Indirect calorimetry was used for measuring energy expenditure. The raw acceleration data were used for training and testing a prediction model in MATLAB environment. The prediction model was built using bagged trees classifier and the most suitable extracted features (mean, maximum, minimum, zero crossing rate, and mean amplitude deviation) were selected using a sequential forward selection method. Leave-one-out cross validation was used for validation. Activities were classified as lying, sitting, light PA (standing, table wiping, floor cleaning, slow walking), moderate PA (fast walking) and vigorous PA (soccer and jogging). The oxygen consumption data were used for estimating the intensity of measured PA. Total accuracy of the prediction model was 96.5%. Mean sensitivity of the model was 95.5% (SD 3.5) and mean specificity 99.1% (SD 0.5). Based on the results PA types can be classified from raw data of the hipworn 3D accelerometer using supervised machine learning techniques with a high sensitivity and specificity. The developed algorithm has a potential for objective evaluations of PA and SED. Keywords— physical activity, sedentary behavior, accelerometer, machine learning, raw data
I. INTRODUCTION
Physical activity (PA) enhances health outcomes by reducing the risk of morbidity and mortality [1]. On the other hand, excessive time spent in sedentary behavior (SED) and especially prolonged periods of sitting are independent risk factors for chronic diseases despite the amount of PA [2-3]. Therefore, objective methods to measure PA and the patterns of SED are required to improve understanding of their health effects. There are several types of PA monitors and analysis methods available [4]. The output of different monitors is dependent on the used signal processing methods and the thresholds © Springer Nature Singapore Pte Ltd. 2018 H. Eskola et al. (eds.), EMBEC & NBC 2017, IFMBE Proceedings 65, DOI: 10.1007/978-981-10-5122-7_218
of PA levels. As a result, different types of accelerometers are not interchangeable with each other when measuring PA in free-living conditions. [5] Researchers have proposed the use of raw acceleration signal instead of some specific unit e.g. counts [6]. However, there is still missing consensus of the accurate analysis technique [4, 6-7]. Mean amplitude deviation (MAD) has b
Data Loading...
