Surveillance of pathogenic bacteria on a food matrix using machine-learning-enabled paper chromogenic arrays

Authors: JIA, ZHEN - University Of Florida; Luo, Yaguang - Sunny; WANG, DAYANG - University Of Massachusetts; HOLLIDAYA, EMMA - University Of Massachusetts; SHARMA, ARNAV - University Of Massachusetts; GREENE, MADISON - University Of Massachusetts; ROCHEE, MICHELLE - University Of Massachusetts; THOMPSON-WITRICKA, KATHERINE - University Of Florida; FLOCKF, GENEVIEVE - Us Army Natick Center; PEARLSTEING, ARNE - University Of Illinois; YU, HENGYONG - University Of Massachusetts; ZHANG, BOCE - University Of Florida

Submitted to: Biosensors and Bioelectronics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/1/2024
Publication Date: 1/2/2024
Citation: Jia, Z., Luo, Y., Wang, D., Hollidaya, E., Sharma, A., Greene, M.M., Rochee, M.R., Thompson-Witricka, K., Flockf, G., Pearlsteing, A.J., Yu, H., Zhang, B. 2024. Surveillance of pathogenic bacteria on a food matrix using machine-learning-enabled paper chromogenic arrays. Biosensors and Bioelectronics. 248. Article e115999. https://doi.org/10.1016/j.bios.2024.115999.
DOI: https://doi.org/10.1016/j.bios.2024.115999

Interpretive Summary: Pathogen contamination on food products is a significant public health concern. Current real-time monitoring systems for foodborne human pathogens often require tedious operations and destructive sampling. In this current study, we report significant improvements to a paper sensor - machine learning (PCA-ML) methodology by streamlining dye selection, sensor fabrication, database construction, and machine learning and validation. Research findings benefit pathogen detection method developers, the food industry and consumers. Early detection and subsequent removal of contaminated foods from the supply chain will save lives and support the sustained industry growth.

Technical Abstract: Global food systems can benefit significantly from real-time monitoring of microbial food safety, a task for which tedious operations, destructive sampling, and inability to monitor multiple pathogens in real time remain challenging. We report significant improvements to a paper chromogenic array sensor - machine learning (PCA-ML) methodology sensing concentrations of volatile organic compounds (VOCs) emitted on a species-specific basis by pathogens, by streamlining dye selection, sensor fabrication, database construction, and machine learning and validation. This approach enables noncontact, time-dependent, simultaneous monitoring of multiple foodborne human pathogens (Listeria monocytogenes, Salmonella, and E. coli O157:H7) at levels as low as 1 log CFU/g) with over 90% accuracy. We provide theoretical results showing that chromogenic response, including limits of detection, depends on time integrals of VOC concentrations. We also discuss the potential for implementing PCA-ML in the food supply chain for different food matrices and pathogens, with species- and strain-specific identification.