Use of fuzzy-logic-inspired features to improve bacterial recognition through classifier fusion

Escherichia coli O157:H7 has been found to cause serious health problems. Traditional methods to identify the organism are quite slow, pulsed-held gel electrophoresis (PFGE) images contain "banding pattern" information which can be used to recognize the bacteria. A fuzzy logic rule-based system is used as a guide to find a good feature set for the recognition of E. coli O157:H7. While the fuzzy rule-based system achieved good recognition, the human inspired features used in the rules were incorporated into a multiple neural network fusion approach which gave excellent separation of the target bacteria. The fuzzy integral was utilized in the fusion of neural networks trained with different feature sets to reach an almost perfect classification rate of E. coli O157:H7 PFGE patterns made available for the experiments.