Development of Patch-Type Sensor Module for Energy Expenditure Estimation

Abstract

For ubiquitous healthcare system, in-situ monitoring of energy expenditure is one of essential requirement. Ubiquitous healthcare means monitoring anyone, in everywhere without any dependence on time and location. Excess nutrient and energy imbalance are considered major important cause of chronic disease. Either of them is base on the individual energy consumption and caloric intake. In this dissertation, accurate quantification of physical activity energy expenditure has been evaluated by the wireless patch-type sensor module. The patch-type sensor has been designed for real-time monitoring heart rate (HR) and movement index (MI). The major components of the system are the sensor board, the rubber case, and the communication module. The relation test for HR yields an error rate within 2% compared with CASE system (GE, USA). And the correlation coefficient between the average MI and conventionally used agility point score was found to be in the range of 0.8-0.9. The Zigbee telecommunication system was used for communication over distances of more than 400 m in an open space. The energy expenditure was estimated by the embedded incremental network. The embedded incremental network includes linear regression (LR) and radial basis function network (RBFN) based on context-based fuzzy c-means (CFCM) clustering. This incremental network is constructed by building a collection of information granules through CFCM that is guided by the distribution of errors of the linear part of the LR model. The results showed significantly high accuracy compared with the conventional network structure (RMSE = 0.78).