Prediction model for mental and physical health condition using risk ratio EM

Abstract

Recently, mobile applications which provide health services at anytime and anywhere are on demand due to the growth of mobile wireless technologies. For the health service, an inspection service middleware is needed for monitoring health condition such as observing and analyzing EEG (electroencephalography), ECG (electrocardiography) and EMG (Electrocardiogram) waveforms from wearable ECG devices under the coverage of a wireless sensor network (WSN). For the inspection service middleware, we propose a new notion of prediction model based on risk ratio Expectation Maximization (EM) by monitoring real-time bio-signals. The prediction model can detect abnormal health condition by the monitoring system. In this paper, we explain the detail algorithms and results for these steps based on EM. There are the five modules as follows: (1) The measurement of bio-signals such as body temperature, EEG, ECG and EMG, (2) Object assessment from measurement wavelength, (3) Situation assessment from GPS in smart device, (4) Maximized health condition using risk ratio EM, (5) Knowledge update and decision making for healthy life.