An intelligent rapid odour recognition model in discrimination of Helicobacter pylori and other gastroesophageal isolates in vitro

Two series of experiments are reported which result in the discrimination b etween Helicobacter pylori and other bacterial gastroesophageal isolates us ing a newly developed odour generating system, an electronic nose and a hyb rid intelligent odour recognition system. In the first series of experiment s, after 5 h of growth (37 degreesC), 53 volatile 'sniffs' were collected o ver the headspace of complex broth cultures of the following clinical isola tes: Staphylococcus aureus, Klebsiella sp., H. pylori, Enterococcus faecali s (10(7) ml(-1)), Mixed infection (Proteus mirabilis, Escherichia coli, and E. faecalis 3 x 10(6) mi each) and sterile cultures. Fifty-six normalised variables were extracted from 14 conductive polymer sensor responses and an alysed by a 3-layer back propagation neural network (NN). The NN prediction rate achieved was 98% and the test data (37.7% of all data) was recognised correctly. Successful clustering of bacterial classes was also achieved by discriminant analysis (DA) of a normalised subset of sensor data. Cross-va lidation identified correctly seven 'unknown' samples. In the second series of experiments after 150 min of microaerobic growth at 37 degreesC, 24 vol atile samples were collected over the headspace of H. pylori cultures in en riched (HPP) and normal (HP) media and 11 samples over sterile (N) cultures . Forty-eight sensor parameters were extracted from 12 sensor responses and analysed by a 3-layer NN previously optimised by a genetic algorithm (GA). GA-NN analysis achieved a 94% prediction rate or 'unknown' data. Additiona lly the 'genetically' selected 16 input neurones were used to perform DA-cr oss validation that showed a clear clustering of three groups and reclassif ied correctly nine 'sniffs'. It is concluded that the most important factor s that govern the performance of an intelligent bacterial odour detection s ystem are: (a) an odour generation mechanism, (b) a rapid odour delivery sy stem similar to the mammalian olfactory system, (c) a gas sensor array of h igh reproducibility and (d) a hybrid intelligent model (expert system) whic h will enable the parallel use of GA-NNs and multivariate techniques. (C) 1 999 Elsevier Science S.A. All rights reserved.