An activity recording system with a radial-basis-function-network-based energy expenditure regression algorithm

This paper presents an activity recording (AR) system and a radial-basis-function-network-based (RBFNB) energy expenditure regression algorithm. The AR system includes motion sensors and an electrocardiogram sensor which is composed of a set of sensor modules (accelerometers and electrocardiogram amplifying/filtering circuits), a MCU module (microcontroller), a wireless communication module (a RF transceiver and a Bluetooth® module), and a storage module (flash memory). A RBFNB energy expenditure regression algorithm consisting of the procedures of data collection, data preprocessing, feature selection, and construction of energy expenditure regression model, has been developed for constructing energy expenditure regression models. The sequential forward search and the sequential backward search were employed as the feature selection strategies, and a radial basis function network as the energy expenditure regression model in this study. Our experimental results exhibited that the proposed energy expenditure regression algorithm can achieve satisfactory energy expenditure estimation by combing appropriate feature selection technique with the regression models.