Non-destructive quality classification of rice taste properties based on near-infrared spectroscopy and machine learning algorithms.

The taste quality of rice is determined by protein and amylose percentages, with low levels indicating high-quality taste in Japan. However, accurate non-destructive screening remains a challenge for the industry. We explored the use of machine learning models and near-infrared spectra to classify rice taste quality. Three models were optimized using 796 brown rice samples from Hokkaido, Japan, produced between 2008 and 2016, and tested on 278 distinct samples from the same region produced between 2017 and 2019. Logistic regression and support vector machine models outperformed the partial least-squares discriminant analysis model, achieving high accuracy (94%), f1-score (90%), average precision (0.94), and low classification error (4%) and allowing accurate non-destructive classification of rice quality. These results not only improve rice quality, post-harvest technology, and producer output in Japan but also could enhance quality control processes and foster the production of high-quality products for other agricultural goods and food commodities worldwide.

Non-destructive assessment of amylose content in rice using a quality inspection system at grain elevators.

Amylose content is an important determinant of rice quality. Its accurate, non-destructive assessment is a challenge to the industry. We examined whether the automatic quality inspection system accurately measured the amylose content. Brown and milled rice models were calibrated with 902 samples produced between 2008 and 2017 in Hokkaido, Japan, and validated individually by samples collected in 2018 (n = 33) and 2019 (n = 71) from several grain elevators within the region. Models were developed by processing the automatic system data comprising of a near-infrared spectrometer, a visible light grain segregator, combined analysis of chemometrics, and by merging low and ordinary amylose variety validation results. The lower standard error of prediction (<0.52%) and a higher ratio of performance deviation (>4.0) in both models enabled accurate non-destructive assessment of amylose content. Hence, the automatic quality inspection system is a useful tool for meeting the demands of higher quality and palatability of rice.

Why RGB Imaging Should be Used to Analyze Fusarium Graminearum Growth and Estimate Deoxynivalenol Contamination.

Size-based fungal growth studies are limited because they do not provide information about the mold's state of maturity, and measurements such as radius and diameter are not practical if the fungus grows irregularly. Furthermore, the current methods used to detect diseases such as Fusarium head blight (FHB) or mycotoxin contamination are labor-intensive and time consuming. FHB is frequently detected through visual examination and the results can be subjective, depending on the skills and experience of the analyzer. For toxin determination (e.g., deoxynivalenol (DON), the best methods are expensive, not practical for routine. RGB (red, green and blue) imaging analysis is a viable alternative that is inexpensive, easy to use and seemingly better if enhanced with statistical methods. This short communication explains why RGB imaging analysis should be used instead of size-based variables as a tool to measure growth of Fusarium graminearum and DON concentration.

Stochastic modeling of variability in survival behavior of Bacillus simplex spore population during isothermal inactivation at the single cell level using a Monte Carlo simulation.

The control of bacterial reduction is important to maintain food safety during thermal processing. The goal of this study was to illustrate and describe variability in bacterial population behavior during thermal processing as a probability distribution based on individual cell heterogeneity regarding heat resistance. Toward this end, we performed a Monte Carlo simulation via computer, and compared and validated the simulated estimations with observed values. Weibullian fitted parameters were estimated from the kinetic survival data of Bacillus simplex during thermal treatment at 94 °C. The variability in reductions of bacterial sporular populations was illustrated using Monte Carlo simulation based on the Weibull distribution of the parameters. In particular, variabilities in viable spore counts and survival probability of the B. simplex spore population were simulated in various replicates. We also experimentally determined the changes in survival probability and distributions of survival spore counts; notably, these were successfully predicted by the Monte Carlo simulation based on the kinetic parameters. The kinetic parameter-based Monte Carlo simulation could thus successfully illustrate bacterial population behavior variability during thermal processing as a probability distribution. The simulation approach may contribute to improving food quality through risk-based processing designs and enhance risk assessment model accuracy.

Combined analysis of near-infrared spectra, colour, and physicochemical information of brown rice to develop accurate calibration models for determining amylose content.

Amylose content is an important determinant of rice quality. Accurate non-destructive determination of amylose content remains a primary challenge for the rice industry. Here, we analysed the accuracy of three models for the non-destructive determination of amylose content. The models were developed by combining near-infrared spectra, colour, and physicochemical information relative to 832 brown rice samples from ten varieties produced between 2009 and 2017 in various regions of Hokkaido, Japan. Models describing low and ordinary amylose varieties were developed individually, merged, and validated using production year samples (2016-2017) different from the calibration set (2009-2015). The resulting accuracy was suitable for industrial application. With standard error of prediction = 0.70% and ratio of performance deviation = 3.56, the combination of near-infrared spectra and physicochemical information produced the most robust model, enabling more precise rice quality screening at grain elevators.

Calculating stochastic inactivation of individual cells in a bacterial population using variability in individual cell inactivation time and initial cell number.

The traditional log-linear inactivation kinetics model considers microbial inactivation as a process that follows first-order kinetics. A basic concept of log reduction is decimal reduction time (D-value), which means time/dose required to kill 90% of the relevant microorganisms. D-value based on the first-order survival kinetics model is insufficient for reliable estimations of bacterial survivors following inactivation treatment. This is because the model does not consider the inactivation curvature and variability in bacterial inactivation. However, although the D-value has some limitations, it is widely used for risk assessment and sterilization time estimation. In this study, stochastic inactivation models are used in place of the conventional D-value to describe the probability of a population containing survivors. As representative bacterial inactivation normally follows a log-linear or log-Weibull model, we calculate the time required for a specific decrease in the number of cells and the number of survival cells as a probability distribution using the stochastic inactivation of individual cells in a population. We compare the probability of a population containing survivors calculated via the D-value, an inactivation kinetics model, and the stochastic formula. The stochastic calculation can be approximately estimated via a kinetic curvature model with less than 5% difference below the probability of a population containing survivors 0.1. This stochastic formula indicates that the D-value model would over- or under-estimate the probability of a population containing survivors when applied to inactivation kinetics with curvature. The results presented in this study show that stochastic analysis using mathematical models that account for variability in the individual cell inactivation time and initial cell number would lead to a realistic and probabilistic estimation of bacterial inactivation.

Stochastic simulation for death probability of bacterial population considering variability in individual cell inactivation time and initial number of cells.

Decimal reduction time (D-value) based on the first-order survival kinetics model is not sufficient for reliable estimation of the bacterial survivors of inactivation treatment because the model does not consider inactivation curvature. However, even though doubt exists in the calculation of D-value, it is still widely used for risk assessment and sterilisation time estimation. This paper proposes an approach for estimating the time-to-inactivation and death probability of bacterial population that considers individual cell heterogeneity and initial number of cells via computer simulation. In the proposed approach, Weibull and Poisson distributions are respectively used to provide individual cell inactivation time variability and initial number of cells variability. Our simulation results show that the time-to-inactivation significantly depends on kinetics curvature and initial number of cells. For example, with increases in the initial number of cells, the respective variance of the time-to-inactivation of log-linear, concave downward curve, and concave upward curve remains constant, decreases, and increases, respectively. The death probability contour plot was successfully generated via our computer simulation approach without using D-value estimation. Further, the death probability calculated using our stochastic approach was virtually the same as that obtained using inactivation kinetics. We validated the simulation by using literature data for acid inactivation of Salmonella population. The results of this study indicate that inactivation curvature can replace D-value extrapolation to estimate the death probability of bacterial population. Further, our computer simulation facilitates realistic estimation of the time-to-inactivation of bacterial population. The R code used for the above stochastic calculation is outlined.

Fusarium graminearum Colors and Deoxynivalenol Synthesis at Different Water Activity.

Deoxynivalenol (DON) is a well-known mycotoxin, responsible for outbreaks of gastrointestinal disorders in Japan. Fusarium graminearum, a parasite of cereal crops, produces this toxin and this is one of the reasons why it is important to understand its metabolism. It is possible to predict the mold's color change and the quantity of DON synthesized throughout its lifecycle. Furthermore, aw has been found to affect the amount of DON. This study aimed to analyze the potential of F. graminearum surface color as a predictor of DON concentration at aw = 0.94, 0.97, and 0.99. Thus, 36 specimens were incubated at 25 °C, 12 at each aw. After 4, 8, 12, and 16 days, three specimens from each aw were collected for color analysis and DON quantification. For color analysis, photos were taken and red, green and blue (RGB) channels were measured on ImageJ software. DON was quantified through liquid chromatography (HPLC). Color changes were only observed at aw = 0.99 because at lower aw the molds presented high growth of white mycelium. Yet, DON increased in all cases. It was only possible to relate the colors with DON concentration at aw = 0.99, where they presented inverse proportionality.

A Glance at Aflatoxin Research in Mozambique.

In Mozambique, aflatoxin research started in the 1960's and has been carried through apparently unrelated efforts according to opportunities. However, they can be grouped in two sets: early epidemiological studies and recent agricultural research. Early investigators found a strong correlation between aflatoxin contamination and primary liver cancer. Since then, there have been efforts to examine the extent of contamination, especially in groundnuts and maize. More recent investigations and interventions aimed mostly to reduce the level of contamination, enough to allow such commodities to gain acceptance in the international market. The current status of knowledge is still marginal but the increasing involvement of local authorities, academia, and international organizations seems promising.

Stochastic evaluation of Salmonella enterica lethality during thermal inactivation.

Stochastic models take into account the uncertainty and variability of predictions in quantitative microbial risk assessment. However, a model that considers thermal inactivation conditions can better predict whether or not bacteria in food are alive. To this end, we describe a novel probabilistic modelling procedure for accurately predicting thermal end point, in contrast to conventional kinetic models that are based on extrapolation of the D value. We used this new model to investigate changes in the survival probability of Salmonella enterica serotype Oranienburg during thermal processing. These changes were accurately described by a cumulative gamma distribution. The predicted total bacterial reduction time with a survival probability of 10-6-the commercial standard for sterility-was significantly shorter than that predicted by the conventional deterministic kinetic model. Thus, the survival probability distribution can explain the heterogeneity in total reduction time for a bacterial population. Furthermore, whereas kinetic methodologies may overestimate the time required for inactivation, our method for determining survival probability distribution can provide an accurate estimate of thermal inactivation and is therefore an important tool for quantitative microbial risk assessment of foods.

Growth-Inhibitory Effect of d-Tryptophan on Vibrio spp. in Shucked and Live Oysters.

Vibrio vulnificus and Vibrio parahaemolyticus are important human pathogens that are frequently transmitted via consumption of contaminated raw oysters. A small amount of d-tryptophan (d-Trp) inhibits some foodborne pathogenic bacteria in high-salt environments. In this study, we aimed to evaluate the antibacterial effect of d-Trp on V. vulnificus and V. parahaemolyticus in culture media, artificial seawater, and shucked and live oysters. The effectiveness of d-Trp in growth inhibition of Vibrio spp. was highly dependent on environmental NaCl concentrations. Higher levels of NaCl (>4.0%) with d-Trp (>20 mM) resulted in higher and more consistent growth inhibition of both Vibrio spp. Treatment with 40 mM d-Trp significantly (P < 0.05) reduced viable V. parahaemolyticus cell counts in tryptic soy broth (TSB) with >4.0% NaCl at 25°C. In contrast, V. vulnificus was more sensitive to d-Trp (20 mM) than V. parahaemolyticus d-Trp (40 mM) treatment with NaCl (>4.5%) significantly (P < 0.05) inhibited the growth of V. parahaemolyticus and V. vulnificus in shucked oysters immersed in peptone water at 25°C throughout a 48-h incubation period. In artificial seawater, d-Trp exhibited a stronger growth-inhibitory effect on V. vulnificus and V. parahaemolyticus at 25°C than in TSB at the same level of salinity and inhibited the growth of both V. parahaemolyticus and V. vulnificus in live oysters at 25°C for 48 h. Furthermore, we tested the synergistic effect of d-Trp and salinity on the inhibition of total viable bacterial counts (TVC) at refrigeration temperature. d-Trp (40 mM) inhibited the growth of TVC in shucked oysters immersed in artificial seawater at 4°C. Therefore, these results revealed that d-Trp will serve as a novel and alternative food preservative to control Vibrio spp. in live oysters at ambient temperature and to extend the shelf-life of shucked oysters at refrigeration temperature.IMPORTANCE Oysters are the primary transmission vehicles for human Vibrio infections. Raw oyster consumption is frequently associated with gastroenteritis. The current postharvest methods, such as high-pressure processing, used to control Vibrio spp. in fresh oysters are still insufficient because of limited facilities, high cost, and potential adverse effects on production. We demonstrate that adding a small amount of d-tryptophan (d-Trp) inhibits the growths of Vibrio parahaemolyticus and Vibrio vulnificus in a high-salt environment at even ambient temperature. We further investigated the d-Trp treatment conditions and clarified the relationship between salt and d-Trp concentrations for optimal growth-inhibitory effect of Vibrio spp. The results will be useful for enhancing the effectiveness of d-Trp by increasing salinity levels. Furthermore, in a nutrientfree environment (artificial seawater), a stronger inhibitory effect could be observed at relatively lower salinity levels, indicating that d-Trp may be regarded as effective food preservation in terms of salinity reduction. Therefore, we suggest the use of exogenous d-Trp in a seawater environment as a novel and effective strategy not only for controlling Vibrio in live oysters at even ambient temperature but also for effectively retarding spoilage bacterial growth and extending the shelf life of shucked oysters at refrigeration temperature.

The Use of Colors as an Alternative to Size in Fusarium graminearum Growth Studies.

Size-based fungal growth studies have limitations. For example, the growth in size stops in closed systems once it reaches the borders and poorly describes metabolic status, especially in the stationary phase. This might lead mycotoxin studies to unrealistic results. Color change could be a viable alternative, as pigments result from a mold’s metabolic activity. This study aimed to verify the possibility of using gray values and the RGB system to analyze the growth of Fusarium graminearum. It consisted of color and area measurements using ImageJ software for specimens grown in yeast extract agar (YEA). The results suggest the utility of color and gray values as reliable tools to analyze the growth of F. graminearum.

Characteristics of d-Tryptophan as an Antibacterial Agent: Effect of Sodium Chloride Concentration and Temperature on Escherichia coli Growth Inhibition.

We previously reported that d-tryptophan exhibits adverse effects on bacterial physiology under osmotic stress. However, the mechanism by which d-tryptophan acts as an inhibitor and/or incompatible solute for bacterial growth has not yet been investigated in detail. In this study, we aimed to determine how osmotic pressure and temperature affect the antimicrobial effect of d-tryptophan. Even at the same level of osmotic pressure, d-tryptophan in conjunction with sodium chloride (NaCl) had a stronger inhibitory effect on the growth of Escherichia coli than that obtained by incubation with potassium chloride (KCl) and sucrose. Because d-tryptophan with NaCl showed the strongest inhibitory effect, we determined the optimum concentration combination of d-tryptophan and NaCl. The growth inhibition boundary conditions as a function of d-tryptophan and NaCl concentrations were determined by a logistic regression model. We found that the minimum level of NaCl for E. coli growth inhibition was 2.5% (w/v) together with 40 mM d-tryptophan. Moreover, the higher the NaCl concentration, the lower the concentration of d-tryptophan that was needed to inhibit bacterial growth. The logistic regression model that we developed enabled us to predict the concentrations required to inhibit bacterial growth. Furthermore, we examined the effect of incubation temperatures ranging from 15 to 46°C on the antimicrobial effect of d-tryptophan. The higher the reaction temperature, the more rapid the decrease of viable E. coli that was observed. This trend is likely attributable to activation of physiological metabolism under the optimum growth temperature. Together, our findings should make a significant contribution to the development of a novel bacterial growth control strategy using d-tryptophan.

Estimation of the probability of bacterial population survival: Development of a probability model to describe the variability in time to inactivation of Salmonella enterica.

Despite the development of numerous predictive microbial inactivation models, a model focusing on the variability in time to inactivation for a bacterial population has not been developed. Additionally, an appropriate estimation of the risk of there being any remaining bacterial survivors in foods after the application of an inactivation treatment has not yet been established. Here, Gamma distribution, as a representative probability distribution, was used to estimate the variability in time to inactivation for a bacterial population. Salmonella enterica serotype Typhimurium was evaluated for survival in a low relative humidity environment. We prepared bacterial cells with an initial concentration that was adjusted to 2 × 10n colony-forming units/2 µl (n = 1, 2, 3, 4, 5) by performing a serial 10-fold dilution, and then we placed 2 µl of the inocula into each well of 96-well microplates. The microplates were stored in a desiccated environment at 10-20% relative humidity at 5, 15, or 25 °C. The survival or death of bacterial cells for each well in the 96-well microplate was confirmed by adding tryptic soy broth as an enrichment culture. The changes in the death probability of the 96 replicated bacterial populations were described as a cumulative Gamma distribution. The variability in time to inactivation was described by transforming the cumulative Gamma distribution into a Gamma distribution. We further examined the bacterial inactivation on almond kernels and radish sprout seeds. Additionally, we described certainty levels of bacterial inactivation that ensure the death probability of a bacterial population at six decimal reduction levels, ranging from 90 to 99.9999%. Consequently, the probability model developed in the present study enables us to estimate the death probability of bacterial populations in a desiccated environment over time. This probability model may be useful for risk assessment to estimate the amount of remaining bacteria in a given sample.

Inactivation of Nonpathogenic Escherichia coli, Escherichia coli O157:H7, Salmonella enterica Typhimurium, and Listeria monocytogenes in Ice Using a UVC Light-Emitting Diode.

Ice, widely used in the food industry, is a potential cause of food poisoning resulting from microbial contamination. Direct microbial inactivation of ice is necessary because microorganisms may have been present in the source water used to make it and/or may have been introduced due to poor hygiene during production or handling of the ice. Nonthermal and nondestructive microbial inactivation technologies are needed to control microorganisms in ice. We evaluated the applicability of a UVC light-emitting diode (UVC-LED) for microbial inactivation in ice. The effects of UV intensity and UV dose of the UVC-LED on Escherichia coli ATCC 25922 and a comparison of UVC-LED with a conventional UV lamp for effective bacterial inactivation in distilled water and ice cubes were investigated to evaluate the performance of the UVC-LED. Finally, we assessed the effects of the UVC-LED on pathogens such as E. coli O157:H7, Salmonella Typhimurium, and Listeria monocytogenes in ice cubes. The results indicated that UVC-LED effectiveness depended on the UV dose at all UV intensity conditions (0.084, 0.025, 0.013, 0.007, and 0.005 mW/cm2) in ice and that UVC-LED could more efficiently inactivate E. coli ATCC 25922 in distilled water and ice than the UV lamp. At a UV dose of 2.64 mJ/cm2, E. coli in distilled water was decreased by 0.90 log CFU/mL (UV lamp) and by more than 7.0 log CFU/mL (UVC-LED). At 15.2 mJ/cm2, E. coli in ice was decreased by 3.18 log CFU/mL (UV lamp) and by 4.45 CFU/mL (UVC-LED). Furthermore, UVC-LED irradiation reduced the viable number of pathogens by 6 to 7 log cycles at 160 mJ/cm2, although the bactericidal effect was somewhat dependent on the type of bacteria. L. monocytogenes in ice was relatively more sensitive to UVC irradiation than were E. coli O157:H7 and Salmonella Typhimurium. These results demonstrate that UVC-LED irradiation could contribute to the safety of ice in the food industry.

Effects of Ohmic Heating, Including Electric Field Intensity and Frequency, on Thermal Inactivation of Bacillus subtilis Spores.

Methods for microbial inactivation are important in the food industry; however, conventional external heating (CH) reduces food quality. Accordingly, the nonthermal effects of ohmic heating (OH) on Bacillus subtilis spores in a sodium chloride aqueous solution at 101°C (i.e., the boiling point), as well as the effects of electric field intensity and frequency during OH, were investigated. Survival kinetics were compared between OH and external CH. The inactivation effect on B. subtilis was greater for all electric field conditions (5, 10, and 20 V/cm) than for CH. In particular, 20 V/cm showed a significantly higher inactivation effect (P < 0.05) on B. subtilis than those of CH at 8, 10, 12, 14, and 16 min. The survival data were fitted to various primary kinetic models. In the Weibull model and the log-linear model, there were significant differences (P < 0.05) in the rate parameters δ and kmax between OH at 20 V/cm and CH. However, there were no significant differences (P > 0.05) in survival kinetics between 20, 40, and 60 kHz; B. subtilis spores were inactivated more efficiently as the frequency increased. B. subtilis spores were almost completely inactivated at 14 to 16 min for the 60-kHz treatment, but spores were still alive at 20 and 40 kHz for the same treatment times. These results demonstrated that OH inactivates B. subtilis spores more effectively than CH. OH conditions with high electric field intensities and high frequencies resulted in efficient B. subtilis spore inactivation.

Modeling Stochastic Variability in the Numbers of Surviving Salmonella enterica, Enterohemorrhagic Escherichia coli, and Listeria monocytogenes Cells at the Single-Cell Level in a Desiccated Environment.

Despite effective inactivation procedures, small numbers of bacterial cells may still remain in food samples. The risk that bacteria will survive these procedures has not been estimated precisely because deterministic models cannot be used to describe the uncertain behavior of bacterial populations. We used the Poisson distribution as a representative probability distribution to estimate the variability in bacterial numbers during the inactivation process. Strains of four serotypes of Salmonella enterica, three serotypes of enterohemorrhagic Escherichia coli, and one serotype of Listeria monocytogenes were evaluated for survival. We prepared bacterial cell numbers following a Poisson distribution (indicated by the parameter λ, which was equal to 2) and plated the cells in 96-well microplates, which were stored in a desiccated environment at 10% to 20% relative humidity and at 5, 15, and 25°C. The survival or death of the bacterial cells in each well was confirmed by adding tryptic soy broth as an enrichment culture. Changes in the Poisson distribution parameter during the inactivation process, which represent the variability in the numbers of surviving bacteria, were described by nonlinear regression with an exponential function based on a Weibull distribution. We also examined random changes in the number of surviving bacteria using a random number generator and computer simulations to determine whether the number of surviving bacteria followed a Poisson distribution during the bacterial death process by use of the Poisson process. For small initial cell numbers, more than 80% of the simulated distributions (λ = 2 or 10) followed a Poisson distribution. The results demonstrate that variability in the number of surviving bacteria can be described as a Poisson distribution by use of the model developed by use of the Poisson process. IMPORTANCE: We developed a model to enable the quantitative assessment of bacterial survivors of inactivation procedures because the presence of even one bacterium can cause foodborne disease. The results demonstrate that the variability in the numbers of surviving bacteria was described as a Poisson distribution by use of the model developed by use of the Poisson process. Description of the number of surviving bacteria as a probability distribution rather than as the point estimates used in a deterministic approach can provide a more realistic estimation of risk. The probability model should be useful for estimating the quantitative risk of bacterial survival during inactivation.

Do bacterial cell numbers follow a theoretical Poisson distribution? Comparison of experimentally obtained numbers of single cells with random number generation via computer simulation.

We investigated a bacterial sample preparation procedure for single-cell studies. In the present study, we examined whether single bacterial cells obtained via 10-fold dilution followed a theoretical Poisson distribution. Four serotypes of Salmonella enterica, three serotypes of enterohaemorrhagic Escherichia coli and one serotype of Listeria monocytogenes were used as sample bacteria. An inoculum of each serotype was prepared via a 10-fold dilution series to obtain bacterial cell counts with mean values of one or two. To determine whether the experimentally obtained bacterial cell counts follow a theoretical Poisson distribution, a likelihood ratio test between the experimentally obtained cell counts and Poisson distribution which parameter estimated by maximum likelihood estimation (MLE) was conducted. The bacterial cell counts of each serotype sufficiently followed a Poisson distribution. Furthermore, to examine the validity of the parameters of Poisson distribution from experimentally obtained bacterial cell counts, we compared these with the parameters of a Poisson distribution that were estimated using random number generation via computer simulation. The Poisson distribution parameters experimentally obtained from bacterial cell counts were within the range of the parameters estimated using a computer simulation. These results demonstrate that the bacterial cell counts of each serotype obtained via 10-fold dilution followed a Poisson distribution. The fact that the frequency of bacterial cell counts follows a Poisson distribution at low number would be applied to some single-cell studies with a few bacterial cells. In particular, the procedure presented in this study enables us to develop an inactivation model at the single-cell level that can estimate the variability of survival bacterial numbers during the bacterial death process.

Survival Kinetics of Salmonella enterica and Enterohemorrhagic Escherichia coli on a Plastic Surface at Low Relative Humidity and on Low-Water Activity Foods.

We investigated the survival kinetics of Salmonella enterica and enterohemorrhagic Escherichia coli under various water activity (aw) conditions to elucidate the net effect of aw on pathogen survival kinetics and to pursue the development of a predictive model of pathogen survival as a function of aw. Four serotypes of S. enterica (Stanley, Typhimurium, Chester, and Oranienburg) and three serotypes of enterohemorrhagic E. coli ( E. coli O26, E. coli O111, and E. coli O157:H7) were examined. These bacterial strains were inoculated on a plastic plate surface at a constant relative humidity (RH) (22, 43, 58, 68, or 93% RH, corresponding to the aw) or on a surface of almond kernels (aw 0.58), chocolate (aw 0.43), radish sprout seeds (aw 0.58), or Cheddar cheese (aw 0.93) at 5, 15, or 25°C for up to 11 months. Under most conditions, the survival kinetics were nonlinear with tailing regardless of the storage aw, temperature, and bacterial strain. For all bacterial serotypes, there were no apparent differences in pathogen survival kinetics on the plastic surface at a given storage temperature among the tested RH conditions, except for the 93% RH condition. Most bacterial serotypes were rapidly inactivated on Cheddar cheese when stored at 5°C compared with their inactivation on chocolate, almonds, and radish sprout seeds. Distinct trends in bacterial survival kinetics were also observed between almond kernels and radish sprout seeds, even though the aws of these two foods were not significantly different. The survival kinetics of bacteria inoculated on the plastic plate surface showed little correspondence to those of bacteria inoculated on food matrices at an identical aw. Thus, these results demonstrated that, for low-aw foods and/or environments, aw alone is insufficient to account for the survival kinetics of S. enterica and enterohemorrhagic E. coli .