Authors: Hao Gong; Bin Chen; Xu Zhang; Charles C. Tseng
Addresses: Department of Biomedical Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA; Department of Electrical and Computer Engineering, Purdue University Calumet, Hammond, IN 46323, USA ' Department of Electrical and Computer Engineering, Purdue University Calumet, Hammond, IN 46323, USA ' Department of Biological Sciences, Purdue University Calumet, Hammond, IN 46323, USA ' Department of Biological Sciences, Purdue University Calumet, Hammond, IN 46323, USA
Abstract: The colony-based laser scatter imaging provides a convincing solution to microbial source tracking. The optical scattering patterns of bacterial colonies are tightly correlated to the corresponding growth patterns. This relationship is manifested as the development of optical scattering patterns with the increment of colony size. An investigation was conducted into this relationship and the optimal range of colony size for improving the accuracy of microbial source tracking technique. All the bacterial samples from five host species were cultivated under the same conditions. The optical scattering patterns were recorded for the average colony diameter from 0.1 mm to 1.5 mm, using a bench top laser imaging system. Gabor wavelet was utilised to encode image signatures. Fuzzy-C-means was employed to cluster the colony patterns from the same host species. The experimental results demonstrate that the optimal range of the colony diameters is 0.8-1.0 mm. The corresponding identification rate of microbial source tracking is >80%.
Keywords: microbial source tracking; laser imaging; colony size optimisation; laser scatter imaging; Gabor wavelet; image signatures; fuzzy C-means; clustering; optical scattering patterns; bacterial colonies.
International Journal of Computational Biology and Drug Design, 2013 Vol.6 No.3, pp.234 - 243
Published online: 29 Jul 2013 *Full-text access for editors Access for subscribers Purchase this article Comment on this article