Author:
Pedro B G,Bertemes-Filho P
Abstract
Abstract
Continuous glucose monitoring is essential to reduce the damages caused by diabetes and for choosing the right treatment approach. In most cases, non-invasive glucose measurement devices generate their results through statistical tools (e.g., artificial neural networks) with an error that increases the further away from the training sample the measurement is. An analytical model would contain only propagated errors. Impedance measurements of lactate ringer’s solutions with egg albumin containing different concentrations of sugar were performed to validate the model proposed for measuring glycemia in human blood using the electrical bioimpedance meter AD5933. The curve fitting showed errors lower than 1.5%. Chemical phenomena, such as reduced sugar, fructosamine and solvation, might explain the behaviors observed in the experiments. The results suggest that the relaxation coefficient has significant changes with the increase of sugar in the solutions. The findings encourage future research with bovine blood for a more realistic analytical model.
Subject
General Physics and Astronomy
Reference16 articles.
1. Diabetes Canada 2018 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada;Can J Diabetes,2018
2. Analytical Model for Blood Glucose Detection Using Electrical Impedance Spectroscopy;Pedro;Sensors [Internet]. MDPI AG,2020