The detection of patients at risk of gastrointestinal toxicity during pelvic radiotherapy by electronic nose and FAIMS: a pilot study.

James A Covington,Linda Wedlake,Karna D Bardhan,Chuka U Nwokolo,Ramesh P Arasaradnam,Nathalie Ouaret,Jervoise Andreyev,Matthew G Thomas

doi:10.3390/s121013002

Abstract

It is well known that the electronic nose can be used to identify differences between human health and disease for a range of disorders. We present a pilot study to investigate if the electronic nose and a newer technology, FAIMS (Field Asymmetric Ion Mobility Spectrometry), can be used to identify and help inform the treatment pathway for patients receiving pelvic radiotherapy, which frequently causes gastrointestinal side-effects, severe in some. From a larger group, 23 radiotherapy patients were selected where half had the highest levels of toxicity and the others the lowest. Stool samples were obtained before and four weeks after radiotherapy and the volatiles and gases emitted analysed by both methods; these chemicals are products of fermentation caused by gut microflora. Principal component analysis of the electronic nose data and wavelet transform followed by Fisher discriminant analysis of FAIMS data indicated that it was possible to separate patients after treatment by their toxicity levels. More interestingly, differences were also identified in their pre-treatment samples. We believe these patterns arise from differences in gut microflora where some combinations of bacteria result to give this olfactory signature. In the future our approach may result in a technique that will help identify patients at “high risk” even before radiation treatment is started.

Highlights

The electronic nose (e-nose) was first conceived in the early 1980s [1] and has undergone continuous refinement ever since
The patients were of a similar age in both groups and had a similar dose of radiotherapy, including fractionation and duration of treatment
This was done as we have previously noticed sample depletion with running samples using a dynamic sampling approach, the sensors are exposed to different gas profiles between tests

Summary

Introduction

The electronic nose (e-nose) was first conceived in the early 1980s [1] and has undergone continuous refinement ever since. The possibility of using the electronic nose for the identification and monitoring of disease has shown considerable promise. The detection of gas phase bio-markers from human biological output, be it breath, sweat, blood, urine or faecal matter, has been shown to identify a disease state. E-nose technology is close to real-time, can be operated without special gases, at room temperature and pressure, is non-invasive and could be produced at a financially acceptable cost for the medical profession. The range of diseases that e-nose technology has been applied to is considerable, ranging from lung cancer, brain cancer and melanoma to inflammatory bowel disease and even schizophrenia [4,5,6,7,8,9]

Objectives

Methods

Results

Discussion

Conclusion