Rapid detection of bacteria at an early stage is crucial to ensure food safety. Raman spectroscopy has been widely used in the detection and identification of foodborne bacteria as it is reliable, label-free, and easy to perform. However, Raman spectra for single-cell analysis are hindered by relatively low signal-to-noise ratio (SNR) and analytical speed. This study aimed to identify different foodborne bacteria using Raman spectroscopy at single-cell level and accelerate the detection using a conditional generative adversarial network (CGAN). Five major foodborne bacteria were investigated, including Campylobacter jejuni, Escherichia coli, Listeria monocytogenes, Staphylococcus aureus, and Salmonella enterica. A total of 1000 single-cell Raman spectra of these bacteria were collected using a homebuilt confocal Raman spectroscopic system with a 671-nm laser. A CGAN was developed to improve the SNR of single-cell Raman spectra and reduce the spectral collection time. Bacterial classification was performed using a convolutional neural network. SNR of single-cell Raman spectra increased from 3.35 to 20.27 after spectral recovery using CGAN and the processed Raman spectra recovered most spectral features. An identification accuracy of 94.9% was achieved for the classification of bacteria using CGAN-recovered single-cell Raman spectra, compared to 60.5% using unprocessed spectra. CGAN can accelerate data acquisition time by one order of magnitude (i.e., 30 s v.s. 3 s) by improving the SNR by a factor of ~ 6. To conclude, we proposed a rapid and reliable approach to identify foodborne bacteria using single-cell Raman spectroscopy combined with CGAN, providing a powerful tool for epidemiological surveillance of foodborne pathogens in the agri-food chain.

Rapid identification of foodborne bacteria using single-cell Raman spectroscopic analysis combined with a conditional generative adversarial network

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.