Enhancing the Classification of Eye Bacteria Using Bagging to Multilayer Perceptron and Decision Tree

Enhancing the Classification of Eye Bacteria Using Bagging to Multilayer Perceptron and Decision Tree

Xu-Qin Li (University of Warwick, UK), Evor L. Hines (University of Warwick, UK), Mark S. Leeson (University of Warwick, UK) and D. D. Iliescu (University of Warwick, UK)
Copyright: © 2011 |Pages: 17
DOI: 10.4018/978-1-61520-915-6.ch011

Abstract

Eye bacteria are vital to the diagnosis of eye disease, which makes the classification of such bacteria necessary and important. This chapter aims to classify different kinds of eye bacteria after the data were collected by an Electronic Nose. First the Multi-layer perceptron (MLP) and decision tree (DT) were introduced as the algorithm and the base classifiers. After that, the bagging technique was introduced to both algorithms and showed that the accuracy of the MLP had been significantly improved. Moreover, bagging to the DT not only reduced the misclassification rate, but enabled DT to select the most important features, and thus, decreased the dimension of the data facilitating an enhanced training and testing process.
Chapter Preview
Top

Experiment And Data Collection

Instrumentation

The most common bacterial eye infection is conjunctivitis. Organisms such as Staphylococcus aureus, Haemophilus influenzae, Streptococcus pneumonia and Escherichia coli have been associated with this condition. Although the number of organisms responsible for infection of the eye is relatively small, the damage caused may be irreversible which makes rapid diagnosis essential. Techniques such as neural network based ENs, which can almost instantly detect and classify odorous volatile components enable the nature of the infection to be diagnosed as quickly as possible. The EN, which is able to mimic the human sense of smell has been the subject of much research at the University of Warwick over the past 20 years or so (Dutta, Hines, Gardner, & Boilot, 2002).

The EN used here was a Cyrano Sciences' Cyranose 320, currently it is used in diverse industries including petrochemical, chemical, medical, food, packaging and many more. For example, the diagnosis of disease often relies on invasive testing methods, subjecting patients to unpleasant procedures. A tool such as the Cyranose 320 will enable physicians and dentists to provide immediate, accurate diagnosis of chemical components and microorganisms in breath, wounds, and bodily fluid.

The Cyranose 320 is a portable system which consists of 32 individual polymer sensors blended with carbon black composite, configured as an array. It works by exposing this array of polymer composite sensors to the chemical components in a vapor or aromatic of volatile compounds they swell. When the sensors come in contact with the vapor, the polymer expands like a sponge, changing the conductivity of the carbon pathways and causing the resistance of the composites. The change in resistance is measured as the sensor signal and captured as the digital pattern representing the test smell (Gardner, Boilot, & Hines, 2005) and from that measurement the overall response to a particular sample is produced.

Complete Chapter List

Search this Book:
Reset