Rapid discrimination of four Salmonella enterica serovars: A performance comparison between benchtop and handheld Raman spectrometers

Abstract

Foodborne illnesses, particularly those caused by Salmonella enterica with its extensive array of over 2600 serovars, present a significant public health challenge. Therefore, prompt and precise identification of S. enterica serovars is essential for clinical relevance, which facilitates the understanding of S. enterica transmission routes and the determination of outbreak sources. Classical serotyping methods via molecular subtyping and genomic markers currently suffer from various limitations, such as labour intensiveness, time consumption, etc. Therefore, there is a pressing need to develop new diagnostic techniques. Surface-enhanced Raman spectroscopy (SERS) is a non-invasive diagnostic technique that can generate Raman spectra, based on which rapid and accurate discrimination of bacterial pathogens could be achieved. To generate SERS spectra, a Raman spectrometer is needed to detect and collect signals, which are divided into two types: the expensive benchtop spectrometer and the inexpensive handheld spectrometer. In this study, we compared the performance of two Raman spectrometers to discriminate four closely associated S. enterica serovars, that is, S. enterica subsp. enterica serovar dublin, enteritidis, typhi and typhimurium. Six machine learning algorithms were applied to analyse these SERS spectra. The support vector machine (SVM) model showed the highest accuracy for both handheld (99.97%) and benchtop (99.38%) Raman spectrometers. This study demonstrated that handheld Raman spectrometers achieved similar prediction accuracy as benchtop spectrometers when combined with machine learning models, providing an effective solution for rapid, accurate and cost-effective identification of closely associated S. enterica serovars.

Keywords: Raman spectrometer; Raman spectrum; Salmonella serovar; characteristic peaks; label‐free SERS; machine learning algorithm.

FIGURE 1

The average and deconvoluted SERS spectra of four S. enterica serovars collected from two Raman spectrometers. (A) Average SERS spectra of the four Salmonella strains under the benchtop Raman spectrometer. (B) Average SERS spectra of the four Salmonella strains under the handheld Raman spectrometer. (C) Deconvoluted SERS spectra of the four Salmonella strains under the benchtop Raman spectrometer. (D) Deconvoluted SERS spectra of the four Salmonella strains under the handheld Raman spectrometer. The X‐axis represents Raman shifts in the 530–1800 cm⁻¹ range, while the Y‐axis represents the relative Raman intensity. a.u. means artificial unit and has no real meaning.

FIGURE 2

OPLS‐DA analysis of S. enterica SERS spectra between the two spectrometers. (A) OPLS‐DA clustering analysis of raw data measured by the benchtop spectrometer. (B) OPLS‐DA clustering analysis of standardized data measured by the benchtop spectrometer. (C) OPLS‐DA clustering analysis of raw data measured by the handheld spectrometer. (D) OPLS‐DA clustering analysis of standardized data measured by the handheld spectrometer.

FIGURE 3

Confusion matrices for the SVM models in the analysis of S. enterica SERS spectra generated from benchtop and handheld Raman spectrometers, respectively. (A) Confusion matrix for the SVM model analysis of S. enterica SERS spectra from the benchtop Raman spectrometer. An average prediction accuracy of 99.5% was achieved. (B) Confusion matrix for the SVM model analysis of S. enterica SERS spectra from the handheld Raman spectrometer. An average prediction accuracy of 100% was achieved. Numbers in the confusion matrix represent the percentage of correctly classified (on the diagonal) or misclassified (off the diagonal) spectra, respectively.