Detection and simultaneous identification of microorganisms from headspace samples using an electronic nose.

Abstract

The detection and simultaneous identification of a range of microorganisms by measuring the volatile compounds produced from plate cultures has been carried out using an electronic nose and a neural network classifier. Headspace samples were taken from static atmospheres formed from inoculated agar plates after a suitable growth period at 37°C and analysed using a standard 16 sensor array operating in transient flow mode. The response of the sensor array to water and to the control media in the absence of microbial growth was also determined, allowing greater discrimination of microbial volatiles. The response curves produced were processed using standard back propagation neural network techniques to provide identification. The overall classification rate for 12 different bacteria and one pathogenic yeast was 93.4%. Data for a sub-set of seven bacteria gave 100% classification using the same methods. In a second experiment three similar yeast cultures were compared and correctly classified at a level of 96.3% with no pre-processing to remove the sample signal generated by the media. Principal component analysis on selected data gave clear discrimination between water vapour and the test samples.