Detecting Bacterial Biofilms Using Fluorescence Hyperspectral Imaging and Various Discriminant Analyses

Area of Science:

• Food Science
• Microbiology
• Spectroscopy

Background:

• Bacterial biofilms on food processing surfaces pose a significant risk of foodborne illness.
• Effective hygiene management relies on the early detection of microbial contamination.

Purpose of the Study:

• To evaluate the feasibility of using fluorescence hyperspectral imaging for detecting Escherichia coli (E. coli) and Salmonella typhimurium (S. typhimurium) biofilms.
• To assess the accuracy of machine learning models in classifying bacterial presence on food processing materials.

Main Methods:

• E. coli and S. typhimurium were cultured on stainless steel and high-density polyethylene coupons.
• Fluorescence hyperspectral images were acquired using UV light (365 nm) in the 420-730 nm range.
• Discriminant analyses, including k-nearest neighbor (k-NN), were employed to classify bacterial contamination.

Main Results:

• Machine learning models, particularly k-NN, achieved over 90% specificity and sensitivity for detecting E. coli and S. typhimurium.
• The developed learning model accurately identified biofilms on the tested surfaces.
• Detection limits were 1-4 log CFU·cm⁻² for E. coli and 1-6 log CFU·cm⁻² for S. typhimurium.

Conclusions:

• Fluorescence hyperspectral imaging is a promising technique for the rapid, non-invasive detection of bacterial biofilms in agro-food processing environments.
• This method can enhance hygiene monitoring and contribute to improved food safety protocols.