Classification of Ear, Nose, and Throat Bacteria Using a Neural-Network-Based Electronic Nose
- PDF / 1,486,026 Bytes
- 5 Pages / 612 x 792 pts (letter) Page_size
- 88 Downloads / 200 Views
Classification of Ear, Nose, and Throat Bacteria Using a Neural-NetworkBased Electronic Nose
Ritaban Dutta, Julian W. Gardner, and Evor L. Hines Abstract This article describes the use of an electronic nose (the Cyranose 320) to sense and identify three species of bacteria responsible for ear, nose, and throat (ENT) infections. Gathered data were a complex mixture of different chemical compounds. An innovative approach for classifying the bacteria data was investigated by using a combination of several clustering algorithms. The best results suggest that the three classes of bacteria examined can be predicted with up to 98% accuracy, allowing more precise diagnosis of ENT infection in patients. This type of bacteria data analysis and feature extraction is difficult, but it can be concluded that combined use of the analysis methods described here can solve the feature extraction problem with very complex data and enhance the performance of electronic noses. Keywords: bacteria analysis, electronic noses, ENT infections, fuzzy c-means, multilayer perceptron, principal component analysis, probabilistic neural networks, radial basis function networks, self-organizing map networks.
conductance between the carbon sensors. This, in turn, increases the total resistance of the film, which is monitored as the sensor signal.2 The responses from the sensors in the array form a response pattern or “smell print.” The sensor technology yields a distinct response signature for each vapor regardless of its complexity.
Swab Samples The ear, nose, and throat (ENT) bacteria experiments were conducted at the Heartland Hospital in Birmingham, United Kingdom, using the Cyranose 320 handheld electronic nose. Patients with different ENT infections were selected for medical experiments using the electronic nose.3 Three common ENT bacteria—Staphylococcus aureus, Pseudomonas aeruginosa, and mixed skin organisms (a group of microorganisms that commonly live on human skin and cause ENT infections in some cases)—were collected on swabs for the experiment. The polymer–carbon black composite sensor system in the electronic nose was used for sniffing all collected swab samples to gather bacteria information on 1000 ENT patients’ in-fections.3 With the assistance of two ENT specialists at Heartlands Hospital, swabs and blood samples were collected from the infected areas of the patients’ ear, nose, and throat regions for subsequent analysis. For comparison, swab samples were also collected from normal (well) patients who did not have bacterial infections. After collection, the four sets of swab samples were cultured in ISO S-type agar solution (a standard culture medium for bacteria) to assist the growth of the ENT bacteria present in the samples. Approximately 3 mg of agar solution containing cultured bacteria was poured into 15 ml vials with distilled water. The vial containing the bacterial solution was allowed to stand for 5–10 min to generate sufficient headspace from the bacterial solution for sniffing purposes.
Test Procedure Introduction
Data Loading...
