Classification of Field Asymmetric Ion Mobility Spectrometry Data for Detection of Bowel Bacteria

Julian W. Gardner, James McIntosh, Natalie Ouaret, Peter Gold, Chuka Nwokolo, Karna Bardhan, Ramesh Arasaradnam, and James Covington

Keywords

Pattern recogniton, Data processing, Chemical sensors

Abstract

Urine samples taken from patients before and after bowel cleansing, previously analysed with an e-nose, have been analysed using an Owlstone Nanotech Lonestar device based upon field asymmetric ion mobility spectrometry (FAIMS). Clinical samples have been studied and chemical headspace classified as a crucial first step towards our understanding of more complex microflora populations. Artificial neural networking techniques have been combined with dimensionality reduction and feature selection methods with an accuracy of up to 94%.

Important Links:



Go Back