Related Experiment Videos
Spectroscopic quantification of bacteria using artificial neural networks.
Mathala J Gupta1, Joseph Irudayaraj, Chitrita Debroy
1Department of Agricultural and Biological Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802, USA.
Journal of Food Protection
|November 24, 2004
Summary
Fourier transform-infrared spectroscopy and artificial neural networks accurately identified foodborne pathogens. This method achieved 100% species classification and 90-100% strain classification accuracy, outperforming traditional methods.
Area of Science:
- Analytical Chemistry
- Microbiology
- Computational Biology
Background:
- Accurate identification of foodborne pathogens is crucial for public health and food safety.
- Traditional methods for pathogen detection can be time-consuming and labor-intensive.
- Spectroscopic techniques combined with machine learning offer potential for rapid and accurate microbial analysis.
Purpose of the Study:
- To evaluate the efficacy of Fourier transform-infrared (FTIR) spectroscopy coupled with artificial neural networks (ANNs) for identifying and classifying foodborne pathogens.
- To compare the performance of the FTIR-ANN method against the traditional plate count method.
Main Methods:
- Bacterial species (Enterococcus faecium, Salmonella Enteritidis, Bacillus cereus, Yersinia enterocolitica, Shigella boydii) and Escherichia coli strains were prepared in phosphate-buffered saline at concentrations from 10^9 to 10^3 CFU/ml.
- Fourier transform-infrared spectra were acquired for each sample.
- Artificial neural networks were trained using the spectral data for classification and validated with independent samples.
Main Results:
- The concentration-based classification of bacterial species using FTIR and ANNs achieved 100% accuracy.
- Strain-based classification of Escherichia coli demonstrated high accuracy, ranging from 90% to 100%.
Conclusions:
- Fourier transform-infrared spectroscopy combined with artificial neural networks provides a highly accurate and efficient method for identifying and classifying foodborne pathogens.
- This technique shows significant promise as a rapid alternative to traditional microbiological methods in food safety applications.