Submitted to: Near Infrared Spectroscopy International Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: June 13, 2013
Publication Date: November 5, 2013
Citation: Park, B., Seo, Y., Yoon, S.C., Windham, W.R., Lawrence, K.C., Hinton Jr, A. 2013. Classification of gram-positive and gram-negative foodborne pathogenic bacteria with hyperspectral microscope imaging. Near Infrared Spectroscopy International Conference Proceedings. 744-748. Interpretive Summary: Among many other detection methods, traditional culture-based methods remain the most reliable and accurate “gold standard” techniques for pathogen detection. These methods involve the culturing of an inoculum to amplify the microbial cell numbers followed by plating on a selective or differential media to generate colonies that can be detected based on their distinct colony morphologies. Culture-based methods are very sensitive with good specificity; and also can give both colony count estimations and qualitative information of the microorganisms present in the food samples. However, culture-based methods are labor intensive and take at least 2~3 days for the microorganisms to multiply to visible colonies for a presumptive positive result. Another challenge is unwanted background microflora grow together with target microorganisms on agar media. However, this limitation could be improved if combined with optical detection methods. Specifically, the optical method with hyperspectral microscope imaging is a good candidate for real-time, in-situ foodborne pathogen detection with less colony biomass. The objective of this research is to develop new optical method to identify foodborne pathogens with acousto-optic tunable filter-based hyperspectral microscope imaging technology. More specifically, classification models to identify gram-positive and gram-negative foodborne pathogenic bacteria were developed for rapid detection of foodborne pathogens.
Technical Abstract: Optical method with hyperspectral microscope imaging (HMI) has potential for identification of foodborne pathogenic bacteria from microcolonies rapidly with a cell level. A HMI system that provides both spatial and spectral information could be an effective tool for analyzing spectral characteristics of foodborne pathogenic bacteria. The most important task for hyperspectral microscope imaging technique is to immobilize live cells on glass slide until the scanning has been completed, because if live cells move during scan, all spectral data from hypercube could be invalid. For the acousto-optic tunable filter (AOTF)-based HMI method, it takes approximately 45 seconds to acquire up to 89 contiguous spectral images at 250 ms exposure time. From this study, we have successfully developed the method to acquire quality hyperspectral microscopic images from various gram-negative and gram-positive bacteria live cells. Support vector machine (SVM) algorithm was able to classify two species (Salmonella vs. Staphylococcus) with 100% accuracy. Also, Salmonella Typhimurium serotype could be identified with 97.6% accuracy from four other serotypes when SVM algorithm was applied.