Detection by hyperspectral imaging of shiga toxin-producing Escherichia coli serogroups O26, O45, O103, O111, O121, and O145 on rainbow agar

Classification of two species of Gram-positive bacteria through hyperspectral microscopy coupled with machine learning

Gram stain is one of the most common techniques used to visualize bacteria under microscopy and classify bacteria into two large groups (Gram-positive and Gram-negative). However, such an inaccurate classification is unfavorable for bacterial research. For instance, soil-rhizosphere bacteria, Bacillus megaterium (B. megaterium) and Bacillus cereus (B. cereus) have different effects on plants, nonetheless, they are both Gram-positive and difficult to be differentiated. Here, we present a method to precisely classify Gram-positive bacteria via hyperspectral microscopy. The pH-value differences in the intracellular environment of various types of bacteria can lead to different ionization of the auxochrome of crystal violet (CV) molecules during the Gram stain process. Consequently, there is a subtle difference in the absorption peak of Gram-stained bacteria. Harnessing hyperspectral microscopy can capture this subtle difference and enable precise classification. Besides the spectral features, the spatial features were also used to improve the quality of bacterial identification. The results show that the classification accuracy of two species of Gram-positive bacteria, B. megaterium and B. cereus, is up to 98.06%. We believe this method can be used for other Gram-positive bacteria and Gram-negative bacteria, realizing a more elaborate classification for Gram-stained bacteria.

Unified Classification of Bacterial Colonies on Different Agar Media Based on Hyperspectral Imaging and Machine Learning

A universal method by considering different types of culture media can enable convenient classification of bacterial species. The study combined hyperspectral technology and versatile chemometric algorithms to achieve the rapid and non-destructive classification of three kinds of bacterial colonies (Escherichia coli, Staphylococcus aureus and Salmonella) cultured on three kinds of agar media (Luria–Bertani agar (LA), plate count agar (PA) and tryptone soy agar (TSA)). Based on the extracted spectral data, partial least squares discriminant analysis (PLS-DA) and support vector machine (SVM) were employed to established classification models. The parameters of SVM models were optimized by comparing genetic algorithm (GA), particle swarm optimization (PSO) and grasshopper optimization algorithm (GOA). The best classification model was GOA-SVM, where the overall correct classification rates (OCCRs) for calibration and prediction of the full-wavelength GOA-SVM model were 99.45% and 98.82%, respectively, and the Kappa coefficient for prediction was 0.98. For further investigation, the CARS, SPA and GA wavelength selection methods were used to establish GOA-SVM simplified model, where CARS-GOA-SVM was optimal in model accuracy and stability with the corresponding OCCRs for calibration and prediction and the Kappa coefficients of 99.45%, 98.73% and 0.98, respectively. The above results demonstrated that it was feasible to classify bacterial colonies on different agar media and the unified model provided a continent and accurate way for bacterial classification.

Rapid Identification of Campylobacter Strains Cultured Under Aerobic Incubation Using Hyperspectral Microscope Imaging

ABSTRACT

Campylobacter is an organism of concern for food safety and is one of the leading causes of foodborne bacterial gastroenteritis. This pathogen can be found in broiler chickens, and the level of allowable contamination of processed poultry is regulated by federal agency guidelines. Traditional methods for detecting and isolating this pathogen from broiler chicken carcasses require time, expensive reagents, and artificially generated microaerophilic atmospheres. An aerobic medium that simplifies the procedure and reduces the expense of culturing Campylobacter has been recently described, and Campylobacter can be grown in this medium in containers that are incubated aerobically. Hyperspectral microscopic imaging (HMI) has been proposed for early and rapid detection of pathogens at the cellular level. The objective of the present study was to utilize HMI to compare differences between Campylobacter cultures grown under artificially produced microaerobic atmospheres and cultures grown in aerobic medium. Hyperspectral microscopic images of three Campylobacter strains were collected cultures grown for 48 h microaerophilically and for 24 and 48 h aerobically, and a quadratic discriminant analysis was used to characterize the bacterial variability. Microaerobically cultured bacteria were detected with 98.7% accuracy, whereas detection accuracy of cultures grown in the novel medium was slightly reduced (-4.8 and -3.2% for 24 and 48 h, respectfully). The Mahalanobis distance multivariate metric was applied to quantify strain variability under all three treatment conditions. Across all strains and treatments, little cluster variation was present (4.22 to 4.42), indicating the consistency of the images collected from the three strains. The classification and spectral consistency was similar for cultures incubated in the aerobic medium for 24 h and cultures grown for 48 h under microaerobic conditions.

HIGHLIGHTS

Identifying non-O157 Shiga toxin-producing Escherichia coli (STEC) using deep learning methods with hyperspectral microscope images

Non-O157 Shiga toxin-producing Escherichia coli (STEC) serogroups such as O26, O45, O103, O111, O121 and O145 often cause illness to people in the United States and the conventional identification of these "Big-Six" are complex. The label-free hyperspectral microscope imaging (HMI) method, which provides spectral "fingerprints" information of bacterial cells, was employed to classify serogroups at the cellular level. In spectral analysis, principal component analysis (PCA) method and stacked auto-encoder (SAE) method were conducted to extract principal spectral features for classification task. Based on these features, multiple classifiers including linear discriminant analysis (LDA), support vector machine (SVM) and soft-max regression (SR) methods were evaluated. Different sizes of datasets were also tested in search for the suitable classification models. Among the results, SAE-based classification models performed better than PCA-based models, achieving classification accuracy of SAE-LDA (93.5%), SAE-SVM (94.9%) and SAE-SR (94.6%), respectively. In contrast, classification results of PCA-based methods such as PCA-LDA, PCA-SVM and PCA-SR were only 75.5%, 85.7% and 77.1%, respectively. The results also suggested the increasing number of training samples have positive effects on classification models. Taking advantage of increasing dataset, the SAE-SR classification model finally performed better than others with average accuracy of 94.9% in classifying STEC serogroups. Specifically, O103 serogroup was classified with the highest accuracy of 97.4%, followed by O111 (96.5%), O26 (95.3%), O121 (95%), O145 (92.9%) and O45 (92.4%), respectively. Thus, the HMI technology coupled with SAE-SR classification model has the potential for "Big-Six" identification.

Application of Hyperspectral Imaging as a Nondestructive Technique for Foodborne Pathogen Detection and Characterization

Microbial food safety is a persistent and exacting global issue due to the multiplicity and complexity of foods and food production systems. Foodborne illnesses caused by foodborne bacterial pathogens frequently occur, thus endangering the safety and health of human beings. Factors such as pretreatments, that is, culturing, enrichment, amplification make the traditional routine identification and enumeration of large numbers of bacteria in a complex microbial consortium complex, expensive, and time-consuming. Therefore, the need for rapid point-of-use detection systems for foodborne bacterial pathogens with high sensitivity and specificity is crucial in food safety control. Hyperspectral imaging (HSI) as a powerful testing technology provides a rapid, nondestructive approach for pathogen detection. This article reviews some fundamental information about HSI, including instrumentation, data acquisition, image processing, and data analysis-the current application of HSI for the detection, classification, and discrimination of various foodborne pathogens. The merits and demerits of HSI for pathogen detection as well as current and future trends are discussed. Therefore, the purpose of this review is to provide a brief overview of HSI, and further lay emphasis on the emerging trend and importance of this technique for foodborne pathogen detection.

Comparison of Agar Media for Detection and Quantification of Shiga Toxin-Producing Escherichia coli in Cattle Feces

The isolation and quantification of non-O157 Shiga toxin-producing Escherichia coli (STEC) from cattle feces are challenging. The primary objective of this study was to evaluate the performance of selected agar media in an attempt to identify an optimal medium for the detection and quantification of non-O157 STEC in cattle feces. Comparison studies were performed using CHROMagar STEC, Possé differential agar (Possé), Possé modified by the reduction or addition of antimicrobials, STEC heart infusion washed blood agar with mitomycin C (SHIBAM), and SHIBAM modified by the addition of antimicrobials. Fourteen STEC strains, two each belonging to serogroups O26, O45, O103, O111, O121, O145, and O157, were used to test detection in inoculated fecal suspensions at concentrations of 10(2) or 10(3) CFU/g. One STEC strain from each of these seven serogroups was used to estimate the concentration of recovered STEC in feces inoculated at 10(3), 10(4), or 10(5) CFU/g. Significantly more suspensions (P < 0.05) were positive for STEC when plated on Possé containing reduced concentrations of novobiocin and potassium tellurite compared with SHIBAM, but not SHIBAM modified by containing these same antimicrobials at the same concentrations. Numerically, more suspensions were positive for STEC by using this same form of modified Possé compared with Possé, but this difference was not statistically significant. More suspensions were positive for STEC cultured on CHROMagar STEC compared with those on Possé (P < 0.05) and on modified Possé (P = 0.05). Most inoculated fecal suspensions below 10(4) CFU/g of feces were underestimated or not quantifiable for the concentration of STEC by using CHROMagar STEC or modified Possé. These results suggest that CHROMagar STEC performs better than Possé or SHIBAM for detection of STEC in bovine feces, but adjustments in the concentrations of novobiocin and potassium tellurite in the latter two media result in significant improvements in their performance.

From Farm to Table: Follow-Up of Shiga Toxin-Producing Escherichia coli Throughout the Pork Production Chain in Argentina

Pigs are important reservoirs of Shiga toxin-producing Escherichia coli (STEC). The entrance of these strains into the food chain implies a risk to consumers because of the severity of hemolytic uremic syndrome. This study reports the prevalence and characterization of STEC throughout the pork production chain. From 764 samples, 31 (4.05%) were stx positive by PCR screening. At farms, 2.86% of samples were stx positive; at slaughter, 4.08% of carcasses were stx positive and at boning rooms, 6% of samples were stx positive. These percentages decreased in pork meat ready for sale at sales markets (4.59%). From positive samples, 50 isolates could be characterized. At farms 37.5% of the isolates carried stx1/stx2 genes, 37.5% possessed stx2e and 25%, carried only stx2. At slaughter we detected 50% of isolates positive for stx2, 33% for stx2e, and 16% for stx1/stx2. At boning rooms 59% of the isolates carried stx1/stx2, 14% stx2e, and 5% stx1/stx2/stx2e. At retail markets 66% of isolates were positive for stx2, 17% stx2e, and 17% stx1/stx2. For the other virulence factors, ehxA and saa were not detected and eae gene was detected in 12% of the isolates. Concerning putative adhesins, agn43 was detected in 72%, ehaA in 26%, aida in 8%, and iha in 6% of isolates. The strains were typed into 14 E. coli O groups (O1, O2, O8, O15, O20, O35, O69, O78, O91, O121, O138, O142, O157, O180) and 10 H groups (H9, H10, H16, H21, H26, H29, H30, H32, H45, H46). This study reports the prevalence and characterization of STEC strains through the chain pork suggesting the vertical transmission. STEC contamination originates in the farms and is transferred from pigs to carcasses in the slaughter process and increase in meat pork at boning rooms and sales markets. These results highlight the need to implement an integrated STEC control system based on good management practices on the farm and critical control point systems in the food chain.

Hyperspectral image acquisition and analysis of cultured bacteria for the discrimination of urinary tract infections

Because of their widespread diffusion and impact on human health, early identification of pathogens responsible for urinary tract infections (UTI) is one of the main challenges of clinical microbiology. Currently, bacteria culturing on Chromogenic plates is widely adopted for UTI detection for its readily interpretable visual outcomes. However, the search of alternate solutions can be highly attractive, especially in the rapidly developing context of bacteriology laboratory automation and digitization, as long as they can improve cost-effectiveness or allow early discrimination. In this work, we consider and develop hyperspectral image acquisition and analysis solutions to verify the feasibility of a "virtual chromogenic agar" approach, based on the acquisition of spectral signatures from bacterial colonies growing on blood agar plates, and their interpretation by means of machine learning solutions. We implemented and tested two classification approaches (PCA+SVM and RSIMCA) that evidenced good capability to discriminate among five selected UTI bacteria. For its better performance, robustness and attitude to work with an expanding set of pathogens, we conclude that the RSIMCA-based approach is worth to be further investigated in a clinical usage perspective.

Rapid and early detection of Salmonella serotypes with hyperspectral microscopy and multivariate data analysis

This study was designed to evaluate hyperspectral microscope images for early and rapid detection of Salmonella serotypes Enteritidis, Heidelberg, Infantis, Kentucky, and Typhimurium at incubation times of 6, 8, 10, 12, and 24 h. Images were collected by an acousto-optical tunable filter hyperspectral microscope imaging system with a metal halide light source measuring 89 contiguous wavelengths every 4 nm between 450 and 800 nm. Pearson correlation values were calculated for incubation times of 8, 10, and 12 h and compared with data for 24 h to evaluate the change in spectral signatures from bacterial cells over time. Regions of interest were analyzed at 30% of the pixels in an average cell size. Spectral data were preprocessed by applying a global data transformation algorithm and then subjected to principal component analysis (PCA). The Mahalanobis distance was calculated from PCA score plots for analyzing serotype cluster separation. Partial least-squares regression was applied for calibration and validation of the model, and soft independent modeling of class analogy was utilized to classify serotype clusters in the training set. Pearson correlation values indicate very similar spectral patterns for reduced incubation times ranging from 0.9869 to 0.9990. PCA score plots indicated cluster separation at all incubation times, with incubation time Mahalanobis distances of 2.146 to 27.071. Partial least-squares regression had a maximum root mean squared error of calibration of 0.0025 and a root mean squared error of validation of 0.0030. Soft independent modeling of class analogy correctly classified values at 8 h (98.32%), 10 h (96.67%), 12 h (88.33%), and 24 h (98.67%) with the optimal number of principal components (four or five). The results of this study suggest that Salmonella serotypes can be classified by applying a PCA to hyperspectral microscope imaging data from samples after only 8 h of incubation.

Comparison of whole genome sequences from human and non-human Escherichia coli O26 strains

Shiga toxin-producing Escherichia coli (STEC) O26 is the second leading E. coli serogroup responsible for human illness outbreaks behind E. coli O157:H7. Recent outbreaks have been linked to emerging pathogenic O26:H11 strains harboring stx 2 only. Cattle have been recognized as an important reservoir of O26 strains harboring stx 1; however the reservoir of these emerging stx 2 strains is unknown. The objective of this study was to identify nucleotide polymorphisms in human and cattle-derived strains in order to compare differences in polymorphism derived genotypes and virulence gene profiles between the two host species. Whole genome sequencing was performed on 182 epidemiologically unrelated O26 strains, including 109 human-derived strains and 73 non-human-derived strains. A panel of 289 O26 strains (241 STEC and 48 non-STEC) was subsequently genotyped using a set of 283 polymorphisms identified by whole genome sequencing, resulting in 64 unique genotypes. Phylogenetic analyses identified seven clusters within the O26 strains. The seven clusters did not distinguish between isolates originating from humans or cattle; however, clusters did correspond with particular virulence gene profiles. Human and non-human-derived strains harboring stx 1 clustered separately from strains harboring stx 2, strains harboring eae, and non-STEC strains. Strains harboring stx 2 were more closely related to non-STEC strains and strains harboring eae than to strains harboring stx 1. The finding of human and cattle-derived strains with the same polymorphism derived genotypes and similar virulence gene profiles, provides evidence that similar strains are found in cattle and humans and transmission between the two species may occur.

Light scattering sensor for direct identification of colonies of Escherichia coli serogroups O26, O45, O103, O111, O121, O145 and O157

Background

Shiga-toxin producing Escherichia coli (STEC) have emerged as important foodborne pathogens, among which seven serogroups (O26, O45, O103, O111, O121, O145, O157) are most frequently implicated in human infection. The aim was to determine if a light scattering sensor can be used to rapidly identify the colonies of STEC serogroups on selective agar plates.

Methodology/Principal Findings

Initially, a total of 37 STEC strains representing seven serovars were grown on four different selective agar media, including sorbitol MacConkey (SMAC), Rainbow Agar O157, BBL CHROMagarO157, and R&F E. coli O157:H7, as well as nonselective Brain Heart Infusion agar. The colonies were scanned by an automated light scattering sensor, known as BARDOT (BActerial Rapid Detection using Optical scattering Technology), to acquire scatter patterns of STEC serogroups, and the scatter patterns were analyzed using an image classifier. Among all of the selective media tested, both SMAC and Rainbow provided the best differentiation results allowing multi-class classification of all serovars with an average accuracy of more than 90% after 10–12 h of growth, even though the colony appearance was indistinguishable at that early stage of growth. SMAC was chosen for exhaustive scatter image library development, and 36 additional strains of O157:H7 and 11 non-O157 serovars were examined, with each serogroup producing unique differential scatter patterns. Colony scatter images were also tested with samples derived from pure and mixed cultures, as well as experimentally inoculated food samples. BARDOT accurately detected O157 and O26 serovars from a mixed culture and also from inoculated lettuce and ground beef (10-h broth enrichment +12-h on-plate incubation) in the presence of natural background microbiota in less than 24 h.

Conclusions

BARDOT could potentially be used as a screening tool during isolation of the most important STEC serovars on selective agar plates from food samples in less than 24 h.