Particle Filter-Based Diagnosis and Prognosis for Human Hydration States

• Published in: IEEE sensors letters Vol. 7; no. 9; pp. 1 - 4
• Main Authors: Niu, Guangxing, Bing, Sen, Zhang, Bin, Chiao, J.-C.
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.09.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Bayesian estimation; Catastrophic events; Chemical and biological sensors; Diagnosis; Dielectric properties; Estimation; human hydration; Hydration; particle filtering; Particle filters; Predictive models; Prognosis; Prognostics and health management; Resonant frequency; Sensors; State estimation; Water consumption
• ISSN: 2475-1472; 2475-1472
• DOI: 10.1109/LSENS.2023.3306984

A miniature microwave resonator to probe the dielectric properties in the dermis layer of the forearm skin was demonstrated in whole-body water consumption experiments to evaluate hydration processes. This letter furthers its applications by using a particle filter-based estimation and prediction method for human hydration states so the system can be used for continuous monitoring of the body. The hydration state is an important indicator of human health conditions. Accurate hydration estimation and prediction can help to reduce the risk of disease development and prevent catastrophic events. A Bayesian estimation method and a particle filtering-based human hydration state diagnosis and prognosis method were demonstrated and tested with experimental data from human subjects. Preliminary results showed that the proposed estimation approach has a satisfactory performance in accuracy between 82.1% and 93.8%. Our investigation found that different models may be needed for different physiological conditions.