Home // BIOTECHNO 2018, The Tenth International Conference on Bioinformatics, Biocomputational Systems and Biotechnologies // View article

Automated Quantification of the Resistance of Epithelial Cell Layers from an Impedance Spectrum

Authors:
Thomas Schmid
Dorothee Günzel
Martin Bogdan

Keywords: Physiology; Epithelia; Impedance Spectroscopy; Feature Selection; Artificial Neural Networks; Random Forests.

Abstract:
Quantifying the permeability of intestinal epithelia is a central task in assessing tissue dysfunctions. This can be achieved conveniently by determining the electric resistance of the epithelial tissue by impedance spectroscopy. While in clinical practice this parameter is often estimated by manual extrapolation of discrete two-dimensional plots of impedance spectra, this approach is known to be particularly unprecise for spectra that deviate from a semicircular shape. Previous computational approaches achieved less then ten percent deviation from the known target value on average, but outliers exhibited significantly larger maximum deviations. Here, we show that systematic feature extraction and selection allow estimation of the epithelial resistance with less than one percent deviation from the known target value on average and less than ten percent at maximum. As a result of detailed modeling of cell culture lines and functional states, epithelial resistance for the cell lines HT-29/B6, IPEC-J2 and MDCK I can be quantified reliably under control conditions, as well as under influence of EGTA and nystatin.

Pages: 8 to 13

Copyright: Copyright (c) IARIA, 2018

Publication date: May 20, 2018

Published in: conference

ISSN: 2308-4383

ISBN: 978-1-61208-639-2

Location: Nice, France

Dates: from May 20, 2018 to May 24, 2018