The evaluation of foodborne pathogenic bacteria using near infrared (NIR) hyperspectral imaging and multivariate image analysis

Date
2018-12
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: Near-infrared (NIR) hyperspectral imaging (HSI) and multivariate image analysis (MIA) was investigated for its potential as a rapid analytical method for the identification of foodborne pathogenic bacteria and distinguishing between the various genera and species used. NIR hyperspectral images of Bacillus cereus, Escherichia coli, Salmonella enteritidis, Staphylococcus aureus and a non-pathogenic bacterium, Staphylococcus epidermidis were acquired using a sisuChema SWIR (short wave infrared) hyperspectral pushbroom imaging system with a spectral range of 920-2514 nm. Hyperspectral images of streaked out (on Luria Bertani agar) bacteria were acquired after 20, 40 and 60 h growth (37 °C). Principal component analysis (PCA) was applied to mean-centered data, and was used to remove background and bad pixels from images. To investigate the possibility of distinguishing between bacteria, standard normal variate (SNV) correction and the Savitzky-Golay technique (2nd derivative, 3rd order polynomial; 25 point smoothing) was applied to data of growth plates imaged after 20 h. PCA score plots, score images and loading line plots were then evaluated. Bacteria were divided into 3 groups which were merged into mosaics. One group contained bacteria which appeared similar in colour (white) on the growth media (B. cereus, E. coli and S. enteritidis), another contained 3 Gram positive bacteria (B. cereus, S. aureus and S. epidermidis) and the third contained 2 species of the same genera (S. aureus and S. epidermidis). On the cleaned images, PCA score plots illustrated distinct chemical differences between colonies which appeared similar in colour on growth media. It was possible to distinguish B. cereus from E. coli and S. enteritidis along PC1 (58.1 % SS) and between E. coli and S. enteritidis in the direction of PC2 (7.75 % SS). S. epidermidis was separated from B. cereus and S. aureus along PC1 (37.5% SS) and was attributed to variation in amino acid and carbohydrate content. Two clusters were evident in the PC1 vs. PC2 PCA score plot for the images of S. aureus and S. epidermidis, thus permitting distinction between species. Differentiation between genera, Gram positive and Gram negative bacteria and pathogenic and non-pathogenic species was possible using NIR hyperspectral imaging. Partial least squares discriminant analysis (PLS-DA) was used as the supervised classification method to differentiate and predict the types of bacteria present. For models built and tested using the 20 h growth plates, the best predictions were made of B. cereus and the two Staphylococcus species, where results ranged from 82.0-99.96% correctly predicted pixels. However the poorest predictions were made of E. coli and S. enteritidis where results ranged from 2.34-53.9. To improve on these results, the effect of colony age has on prediction accuracies were investigated, while keeping rapidity in mind. PLS-DA models were built on standard normal variate (SNV) treated data for 20, 40 and 60 h growth plates and tested on a second set of plates. Predictions were improved – the highest overall prediction accuracies, where test plates required the least amount of growth time, was found with models built after 60 h growth and tested using 20 h growth plates. Predictions for bacteria differentiation within these models ranged from 83.1 to 98.8 % correctly predicted pixels.
AFRIKAANSE OPSOMMING: Naby-infrarooi (NIR) hiperspektrale beelding en meeveranderlike beeld ontleding (MIA) is ondersoek vir die potensiaal daarvan om te dien as a ‘n vinnige analitiese metode vir die identifisering van wat voedselbederf patogene bakteriese veroorsaak. Hierdie methods kan ook gebruik word om te onderskei tussen die verskillende genera en species wat voorkom. NIR hiperspektrale beelde van Bacillus cereus, Escherichia coli, Salmonella enteritidis, Staphylococcus aureus en ‘n nie-patogeniese bakterie, Staphylococcus epidermidis is verkry deur gebruik te maak van ‘n sisuChema SWIR (kort golf infrarooi) hiperspectrale beelding stelsel met spectrale reikwydte van 920-2498 nm. HIperspektrale beelde van uitgestreepte (op Luria Bertani agar) bakterieë is verky ná 20, 40 en 60 uur groei tyd (37 ° C). Hoof komponent analise (HKA) is toegepas op middelgesentreerde data, en is gebruik om sodoende agtergrond en dooie piksels te verwyder uit die beeld. Die standaard normale variasie (SNV) restelling en die Savitzky-Golay tegniek (2de afgeleide, 3de orde polinoom, 25 punt gladding) is toegepas op data van groeiplate wat na 20 uur afgeneem word, om die moontlikheid om tussen bakterieë te onderskei. HKA telling grafieke, telling beelde en lading lyne plotte is geëvalueer. Bakterieë is in 3 groepe verdeel wat in mosaïek saamgevoeg is. Een groep bevat bakterieë wat soortgelyk in kleur voorkom (wit) op die groeimedia (B. cereus, E. coli en S. enteritidis) 3 Gram –positiewe bakterieë (B. cereus, S. aureus en S. epidermidis), en die derde groep 2 spesies van dieselfde genera (S. aureus en S. epidermidis). Op diie gekorrigeerde beelde het HK telling grafieke met afsonderlike chemise verskille tussen kolonies geïllustreer wat op die groeimedia soordgelyk gelyk het. Deur die groepering van datapunte te bestudeer, kon daar onderskei word tussen B. cereus en E. coli, asook B. cereus en S. enteritidis en tussen E. coli en S. enteritidis in die rigting van HK 2 (7.75% SS). Om hierde observasie te maak is daar na die HK1 (58.1 % SS) as verwys op die tellings grafiek. Die aminosuur en kolidraatinhoud het did moontlik gemaak om tussen B. cereus en S. aureus op PC1 (37.5 % SS) te onderskei. Twee groeperings was duidelik in die HK1 vs HK2 HKA-telling grafiek vir die beeld van S. aureus en S. epidermidis, wat dus onderskied tussen spesies toelaat. Differensiasie tussen genera, Gram-positiewe en Gram-negatiewe bakterieë asook patogeniese en nie-patogeniese spesies was moontlik met behulp van NIR hiperspektrale beelding. Parsiële kleinste kwadrate diskriminant analise (PKW-DA) is gebruik as die supervised klassifikasie method om die tipes bakterieë teenwoordig te onderskei en te voorspel. Vir modelle wat gebou en getoets is met behulp van die 20 h groeiplaatjies, is die beste voorspelling gemaak van B. cereus en die twee Staphyloccocus species,, waar die resultate gewisselhet vanaf wat aanbetref 82.0-99.6 % korrek voorspelde piksels. Die swakste voorspellings is egte gemaak van E. coli en S. enteritidis, waar die resultate gewissel het vanaf 2.34-53.9. Om te verbeter op hierdie resultate, is die effek van kolonie ouderdom op voorspellingsakkuraatheid ondersoek, terwyl die tyd wat dit nee mom die toets aft e handel in gedagte gehou word. PKW-DA modelle is gebou op SNV behandelde data vir 20, 40 en 60 h groeiplaatjies en getoets op ‘n tweede stek plate as bevestiging. Voorspellings is op die volgende wyse verbeter: die hoogste algehele voorspellings akkuraatheid, waar toetsplate die minste hoeveelheid groeityd benodig het, is gevind met modelle gebou ná 60 uur groei en getoets met 20 h groeiplaatjies. Voorspellings vir bakteriedifferensiasie binne hierdie modelle het gewissel van 83,1 tot 98,8% korrek voorspelde pixels.
Description
Thesis (MSc)--Stellenbosch University, 2018.
Keywords
Multivariate analysis, Hyperspectral imaging, Imaging, hyperspectral, Near infrared spectroscopy, Foodborne diseases – Analysis, Pathogenic bacteria – Analysis, Food -- Bacteriology, UCTD
Citation