Physiological Measurement and Modeling of Dengue and Dengue Haemorrhagic Fever using Bioelectrical Impedance and Advanced Signal Processing Techniques

摘要

This paper presents a physiological measurement and modeling of dengue fever (DF) and dengue haemorrhagic fever (DHF) using bioelectrical impedance analysis (BIA) and advanced signal processing (ASP) techniques. The development of novel non-invasive prognosis and diagnosis systems for DF and DHF using BIA and ASP techniques will be discussed. BIA and multivariate analysis have been used to model the haemoglobin (Hb) concentration in dengue patients. An improved Hb modeling in dengue patients using linear and nonlinear autoregressive moving average with exogenous input models give better accuracy. Whilst using the BIA, statistical analysis, Self organising map (SOM) and artificial neural network (ANN) have successfully diagnose, and classify severity of risk in DF and DHF patients. Furthermore, the use of artificial neural network (ANN) has non-invasively predicted the day of fever defervescence in DF and DHF patients.