Characterisation and Classification of Foodborne Bacteria Using Reflectance FTIR Microscopic Imaging

Abstract

This work investigates the application of reflectance Fourier transform infrared (FTIR) microscopic imaging for rapid, and non-invasive detection and classification between Bacillus subtilis and Escherichia coli cell suspensions dried onto metallic substrates (stainless steel (STS) and aluminium (Al) slides) in the optical density (OD) concentration range of 0.001 to 10. Results showed that reflectance FTIR of samples with OD lower than 0.1 did not present an acceptable spectral signal to enable classification. Two modelling strategies were devised to evaluate model performance, transferability and consistency among concentration levels. Modelling strategy 1 involves training the model with half of the sample set, consisting of all concentrations, and applying it to the remaining half. Using this approach, for the STS substrate, the best model was achieved using support vector machine (SVM) classification, providing an accuracy of 96% and Matthews correlation coefficient (MCC) of 0.93 for the independent test set. For the Al substrate, the best SVM model produced an accuracy and MCC of 91% and 0.82, respectively. Furthermore, the aforementioned best model built from one substrate was transferred to predict the bacterial samples deposited on the other substrate. Results revealed an acceptable predictive ability when transferring the STS model to samples on Al (accuracy = 82%). However, the Al model could not be adapted to bacterial samples deposited on STS (accuracy = 57%). For modelling strategy 2, models were developed using one concentration level and tested on the other concentrations for each substrate. Results proved that models built from samples with moderate (1 OD) concentration can be adapted to other concentrations with good model generalization. Prediction maps revealed the heterogeneous distribution of biomolecules due to the coffee ring effect. This work demonstrated the feasibility of applying FTIR to characterise spectroscopic fingerprints of dry bacterial cells on substrates of relevance for food processing.

Combining AI Tools with Non-Destructive Technologies for Crop-Based Food Safety: A Comprehensive Review

“…The forward feature selection (FFS) method was used to reduce features in spectra space, followed by three linear classifiers, namely the linear discriminant classifier (LDC), logarithmic linear classifier (LOGLC), and quadratic discriminant classifier (QDC). Also, it was coupled with smoothing and standard normal variate (SNV) for noise removal, followed by SVM and PLSDA to characterize and differentiate between Bacillus subtilis and Escherichia coli cell suspensions in food spoilage context [ 90 ]. The authors of [ 71 ] have also showed the benefits of using FTIR combined with five classification approaches, namely adaptive boosting (AdaBoost), random forests (RF), SVM, and multilayer perceptrons (MLPs) for the automated classification of contaminated maize.…”

“… A diagram representing the chronological (from 2001 to 2022) distribution of the resulting articles. Each data-point represents an article, the colors are according to which field of study, and the lines are related to the decision-making objective Mehl, 2001 [ 47 ]; Irudayaraj, 2002 [ 48 ]; Yang, 2003 [ 49 ]; Gupta, 2005 [ 50 ]; Gupta, 2006 [ 51 ]; He, 2008 [ 55 ]; A. Scarlatos, 2008 [ 54 ]; Siripatrawan, 2008 [ 56 ]; Stöckel, 2010 [ 57 ]; Günes, 2013 [ 60 ]; Shapaval, 2013 [ 59 ]; Geng, 2017 [ 66 ]; Y.Shen, 2017 [ 67 ]; Lasch, 2018 [ 68 ]; Guo, 2019 [ 72 ]; Kaushik, 2019 [ 69 ]; Liu, 2019 [ 74 ]; Öner, 2019 [ 71 ]; Sun, 2019 [ 75 ]; Wan-dan, 2019 [ 73 ]; Bertania, 2020 [ 85 ]; Le, 2020 [ 77 ]; Sahu, 2020 [ 80 ]; Shen, 2020 [ 84 ]; Wange, 2020 [ 78 ]; Weng, 2020 [ 79 ]; Wu, 2020 [ 83 ]; Gonzalez, 2021 [ 93 ]; Guo, 2021 [ 94 ]; Li, 2021 [ 95 ]; Magnus, 2021 [ 91 ]; Nie, 2021 [ 101 ]; Rahi, 2021 [ 86 ]; Ricci, 2021 [ 97 ]; Vakilian, 2021 [ 96 ]; Wang, 2021 [ 92 ]; Xu, 2021 [ 90 ]; Yan, 2021 [ 99 ]; Yin, 2021 [ 98 ]; Adejimi, 2022 [ 109 ]; Bowler, 2022 [ 107 ]; Cordovana, 2022 [ 110 ]; Kim, 2022 [ 108 ]; Manthou, 2022 [ 115 ]; Rady, 2022 [ …”

Optical Immunosensor for the Detection of Listeria monocytogenes in Food Matrixes

“…In general, PCR and DNA hybridization-based techniques are faster but need isolated genetic materials, specific instrumentation, trained personnel, and high cost, which is not suitable for clinical point-of-care usage. Nowadays, various and specific detection methods have been introduced for detecting LM in food samples, including loop-mediated isothermal amplification assays, matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectroscopy, and Fourier transform infrared (FT-IR) spectroscopy. − Though some remarkable advantages are reported, these methods lack various aspects such as calculating the number, biochemical characteristics, and colony identification, which are more time-consuming, and these immunological assays can suffer in specificity and hence false-positive results were noticed . Moreover, these methods have been lacking in additional aspects like the amount of waste produced, long enrichment time, quantity and the use of expensive chemicals, and specialized equipment …”

Exploring the identification of multiple bacteria on stainless steel using multi-scale spectral imaging from microscopic to macroscopic

“…Towards this goal, we have recently probed the capability of FTIR reflectance micro-spectroscopic imaging to detect and classify bacterial cells that were dried onto metallic surfaces, i.e. Aluminium, Stainless Steel 304 and 316 11 . In our previous work, we found that GP Bacillus subtilis and GN Escherichia coli could be reliably detected and distinguished from each other on both aluminium and stainless-steel surfaces at pre-application optical densities ranging from 10 to 0.1 OD 600 .…”