Authors : M. Subadra , N.S. Marimuthu , R. Ravi and G.S. Vijayakumar
Abstract: A recent development in odour sensing technology and artificial intelligence, as E-nose system, has rediscovered the use of smell in clinical diagnosis. Nowadays, this technology is becoming an interesting alternative for medical point-of-care devices. This study analyses the possibility of E-nose for rapid and accurate detection of micro-organisms in normal sterile body fluids, to ensure correct chemotherapy. In this study, 75 samples of different kinds of bacteria which are the major cause for Sepsis and Urinary Tract Infection were inoculated in 20 mL growth medium and incubated for 2 h for volatile generation. These samples were analysed with an electronic nose. The instrument, equipped with 12 Metal Oxide Semiconductor (MOS) sensors, was used to generate a pattern of the volatile compounds present in the pathological samples. The sensor responses were evaluated by Principal Component Analysis (PCA) and Artificial Neural Network (ANN). Good results were obtained in the classification of bacterial samples by using a neural network model based on a multilayer perceptron that learned using a back propagation algorithm. The methodology is simple, rapid and the results suggest that the electronic nose could be a used as a tool for detection.
M. Subadra , N.S. Marimuthu , R. Ravi and G.S. Vijayakumar , 2007. Rapid Detection of Micro-Organisms in Sterile Body Fluids by Electronic Nose System . Asian Journal of Information Technology, 6: 1091-1097.