The prediction of bacteria type and culture growth phase by an electronic nose with a multi-layer perceptron network

Abstract

An investigation into the use of an electronic nose to predict the class and growth phase of two potentially pathogenic microorganisms , Eschericha coli (E. coli) and Staphylococcus aureus (S. aureus), has been performed. In order to do this we have developed an automated system to sample, with a high degree of reproducibility, the head space of bacterial cultures grown in a standard nutrient medium. Head spaces have been examined by using an array of six different metal oxide semiconducting gas sensors and classified by a multi-layer perceptron (MLP) with a back-propagation (BP) learning algorithm. The performance of 36 different pre-processing algorithms has been studied on the basis of nine different sensor parameters and four different normalization techniques. The best MLP was found to classify successfully 100% of the unknown S. aureus samples and 92% of the unknown E. coli samples, on the basis of a set of 360 training vectors and 360 test vectors taken from the lag, log and stationary growth phases. The real growth phase of the bacteria was determined from optical cell counts and was predicted from the head space samples with an accuracy of 81%. We conclude that these results show considerable promise in that the correct prediction of the type and growth phase of pathogenic bacteria may help both in the more rapid treatment of bacterial infections and in the more efficient testing of new antibiotic drugs.

Tunable conjugated polymers for bacterial differentiation

“…Biochemical fingerprints of whole cells have provided the basis for several new bacterial detection tools including mass spectrometry (MS), vibrational spectroscopy (Raman and infrared), electronic noses and light scattering analyses [66]. Pattern recognition techniques utilising the electronic nose (e-nose) in particular, have been proposed for use in many fields including microbial detection for medicine and food quality control [41,[73][74][75]. Here, we introduce a new strategy for bacterial discrimination based on bacterial adhesion properties which builds on this established body or work, but offers a new direction for research and application.…”

Wood production, wood technology, and biotechnological impacts

“…Most studies with electronic noses deal with the detection and classification of bacteria (Holmberg 1997, Gardner et al 1998), using sensorarrays consisting of six to nine metal oxide semiconductor gas sensors. Few reports are available for fungal detection: with an accuracy of 93% six spoilage fungi of meat (four Eurotium spp., each one Penicillium and Wallemia species) were classified on blood agar 24 hours after infestation and prior to visible growth, using an electronic nose consisting of 14 polymer sensors (Keshri et al 1998 (Keshri et al 2003).…”

Fast and robust gas identification system using an integrated gas sensor technology and Gaussian mixture models

“…Numerous preprocessing techniques have been proposed in the literature [4]. The most common procedure uses the steady state of the sensors' response as a feature vector and ignores the transient response [19]. A number of compression algorithms have been proposed to extract additional information from the transient response, resulting in improved selectivity and increased recognition accuracy [20].…”