Output details
11 - Computer Science and Informatics
University of Westminster
Artificial Odor Discrimination System using electronic nose and neural networks for the identification of urinary tract infection
<24>Originality: This paper presents a novel extended normalized radial basis function model trained with the expectation maximization algorithm. The model incorporated a “split and merge” technique to dynamically build its structure. The fusion of multiple classifiers dedicated to specific feature parameters was also developed based on fuzzy integral principles.
Significance: The software was utilized to detect in vivo urinary tract infections clinical samples with the aid of an electronic nose.
Rigour: This study has shown the potential of using point-of-care devices based on gas-sensors. Results published in a leading peer-reviewed journal.