Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (FTIR) and Artificial Neural Networks Applied to Investigate Quantitative Changes of Selected Soluble Biomarkers, Correlated with H. pylori Infection in Children and Presumable Consequent Delayed Growth

Abstract

Helicobacter pylori infections causing gastroduodenal disorders are a common medical problem. The aim of this study was to determine the specific motives of infrared spectroscopy (IR) spectra of sera from H. pylori-infected and uninfected children applied to investigate quantitatively-selected soluble biomarkers correlated with H. pylori infection in children and presumable consequent delayed growth. Sera from 41 children infected with H. pylori (Hp(+)) and 43 uninfected (Hp(−)) under the care of the Polish Mother’s Hospital in Lodz, Poland, were analyzed. The H. pylori status was confirmed by gastroscopy, 13C urea breath testing, and anti-H. pylori IgG antibodies. Infrared spectra were measured using an FTIR/FT-NIR Spectrum 400 spectrometer (PerkinElmer). The IR spectrum was measured in the wavenumber range 3000–750 cm−1 and subjected to mathematical calculation of the first derivative. Based on the chi-square test, 10 wavenumbers of spectra correlating with H. pylori infection were selected for use in designing an artificial neural network. Ten parts of the IR spectra correlating with H. pylori infection were identified in the W2 and W3 windows associated mainly with proteins and the W4 window related to nucleic acids and hydrocarbons. Artificial neural networks for H. pylori infection were developed based on chemometric data. By mathematical modeling, children were classified towards H. pylori infection in conjunction with elevated levels of selected biomarkers in serum potentially related to growth retardation. The study concludes that IR spectroscopy and artificial neural networks may help to confirm H. pylori-driven growth disorders in children.