Comparison between random forest and gradient boosting machine methods for predicting Listeria spp. prevalence in the environment of pastured poultry farms

Optimisation led energy-efficient arsenite and arsenate adsorption on various materials with machine learning

The contamination of water by arsenic (As) poses a substantial environmental challenge with far-reaching influence on human health. Accurately predicting adsorption capacities of arsenite (As(III)) and arsenate (As(V)) on different materials is crucial for the remediation and reuse of contaminated water. Nonetheless, predicting the optimal As adsorption on various materials while considering process energy consumption continues to pose a persistent challenge. Literature data regarding the As adsorption on diverse materials were collected and employed to train machine learning models (ML), such as CatBoost, XGBoost, and LGBoost. These models were utilized to predict both As(III) and As(V) adsorption on a variety of materials using their reaction parameters, structural properties, and composition. The CatBoost model exhibited superior accuracy, achieving a coefficient of determination (R²) of 0.99 and a root mean square error (RMSE) of 1.24 for As(III), and an R² of 0.99 and RMSE of 5.50 for As(V). The initial As(III) and As(V) concentrations were proved to be the primary factors influencing adsorption, accounting for 27.9 % and 26.6 % of the variance for As(III) and As(V) individually. The genetic optimization led optimisation process, considering the low energy consumption, determined maximum adsorption capacities of 291.66 mg/g for As(III) and 271.56 mg/g for As(V), using C-Layered Double Hydroxide with reduced graphene oxide and chitosan combined with rice straw biochar, respectively. To further facilitate the process design for different real-life applications, the trained ML models are embedded into a web-app that the user can use to estimate the As(III) and As(V) adsorption under different design conditions. The utilization of ML for the energy-efficient As(III) and As(V) adsorption is deemed essential for advancing the treatment of inorganic As in aquatic settings. This approach facilitates the identification of optimal adsorption conditions for As in various material-amended waters, while also enabling the timely detection of As-contaminated water.

Harnessing Spectral Libraries From AVIRIS-NG Data for Precise PFT Classification: A Deep Learning Approach

The generation of spectral libraries using hyperspectral data allows for the capture of detailed spectral signatures, uncovering subtle variations in plant physiology, biochemistry, and growth stages, marking a significant advancement over traditional land cover classification methods. These spectral libraries enable improved forest classification accuracy and more precise differentiation of plant species and plant functional types (PFTs), thereby establishing hyperspectral sensing as a critical tool for PFT classification. This study aims to advance the classification and monitoring of PFTs in Shoolpaneshwar wildlife sanctuary, Gujarat, India using Airborne Visible/Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) and machine learning techniques. A comprehensive spectral library was developed, encompassing data from 130 plant species, with a focus on their spectral features to support precise PFT classification. The spectral data were collected using AVIRIS-NG hyperspectral imaging and ASD Handheld Spectroradiometer, capturing a wide range of wavelengths (400-1600 nm) to encompass the key physiological and biochemical traits of the plants. Plant species were grouped into five distinct PFTs using Fuzzy C-means clustering. Key spectral features, including band reflectance, vegetation indices, and derivative/continuum properties, were identified through a combination of ISODATA clustering and Jeffries-Matusita (JM) distance analysis, enabling effective feature selection for classification. To assess the utility of the spectral library, three advanced machine learning classifiers-Parzen Window (PW), Gradient Boosted Machine (GBM), and Stochastic Gradient Descent (SGD)-were rigorously evaluated. The GBM classifier achieved the highest accuracy, with an overall accuracy (OAA) of 0.94 and a Kappa coefficient of 0.93 across five PFTs.

Whole-genome sequencing and metagenomics reveal diversity and prevalence of Listeria spp. from soil in the Nantahala National Forest

Listeria spp. are widely distributed environmental bacteria associated with human foodborne illness. The ability to detect and characterize Listeria strains in the natural environment will contribute to improved understanding of transmission routes of contamination. The current standard for surveillance and outbreak source attribution is whole-genome sequencing (WGS) of Listeria monocytogenes clinical isolates. Recently, metagenomic sequencing has also been explored as a tool for the detection of Listeria spp. in environmental samples. This study evaluated soil samples from four locations across altitudes ranging from 1,500 to 4,500 ft in the Nantahala National Forest in North Carolina, USA. Forty-two Listeria isolates were cultured and sequenced, and 12 metagenomes of soil bacterial communities were generated. These isolates comprised 14 distinct strains from five species, including Listeria cossartiae subsp. cayugensis (n = 8; n represents the number of distinct strains), L. monocytogenes (n = 3), "Listeria swaminathanii" (Lsw) (n = 1), Listeria marthii (n = 1), and Listeria booriae (n = 1). Most strains (n = 13) were isolated from lower altitudes (1,500 or 2,500 ft), while the L. swaminathanii strain was isolated from both higher (4,500 ft) and lower (1,500 ft) altitudes. Metagenomic analysis of soil described a reduction in both bacterial community diversity and relative abundance of Listeria spp. as the altitude increased. Soil pH and cation exchange capacity were positively correlated (P < 0.05) with the abundance of Listeria spp. as detected by metagenomics. By integrating culture-independent metagenomics with culture-based WGS, this study advances current knowledge regarding distribution of Listeria spp. in the natural environment and suggests the potential for future use of culture-independent methods in tracking the transmission of foodborne pathogens.

IMPORTANCE

As a foodborne pathogen, Listeria continues to cause numerous illnesses in humans and animals. Studying the diversity and distribution of Listeria in soil is crucial for understanding potential sources of contamination and developing effective strategies to prevent foodborne outbreaks of listeriosis. Additionally, examining the ecological niches and survival mechanisms of Listeria in natural habitats provides insights into its persistence and adaptability, informing risk assessments and public health interventions. This research contributes to a broader understanding of microbial ecology and the factors influencing foodborne pathogen emergence, ultimately enhancing food safety and protecting public health. Moreover, using a metagenomic approach provides a detailed understanding of the soil microbial ecosystems, leading to more effective monitoring and control of foodborne pathogens. This study also highlights the potential for integrating metagenomics into routine surveillance systems for food safety in the near future.

Development of prediction software to describe total mesophilic bacteria in spinach using a machine learning-based regression approach

The purpose of this study was to create a tool for predicting the growth of total mesophilic bacteria in spinach using machine learning-based regression models such as support vector regression, decision tree regression, and Gaussian process regression. The performance of these models was compared to traditionally used models (modified Gompertz, Baranyi, and Huang models) using statistical indices like the coefficient of determination (R2) and root mean square error (RMSE). The results showed that the machine learning-based regression models provided more accurate predictions with an R2 of at least 0.960 and an RMSE of at most 0.154, indicating that they can be used as an alternative to traditional approaches for predictive total mesophilic. Therefore, the developed software in this work has a significant potential to be used as an alternative simulation method to traditionally used approach in the predictive food microbiology field.

Using GWAS and Machine Learning to Identify and Predict Genetic Variants Associated with Foodborne Bacteria Phenotypic Traits

One of the main challenges in food microbiology is to prevent the risk of outbreaks by avoiding the distribution of food contaminated by bacteria. This requires constant monitoring of the circulating strains throughout the food production chain. Bacterial genomes contain signatures of natural evolution and adaptive markers that can be exploited to better understand the behavior of pathogen in the food industry. The monitoring of foodborne strains can therefore be facilitated by the use of these genomic markers capable of rapidly providing essential information on isolated strains, such as the source of contamination, risk of illness, potential for biofilm formation, and tolerance or resistance to biocides. The increasing availability of large genome datasets is enhancing the understanding of the genetic basis of complex traits such as host adaptation, virulence, and persistence. Genome-wide association studies have shown very promising results in the discovery of genomic markers that can be integrated into rapid detection tools. In addition, machine learning has successfully predicted phenotypes and classified important traits. Genome-wide association and machine learning tools have therefore the potential to support decision-making circuits intending at reducing the burden of foodborne diseases. The aim of this chapter review is to provide knowledge on the use of these two methods in food microbiology and to recommend their use in the field.

Schistosomiasis transmission: A machine learning analysis reveals the importance of agrochemicals on snail abundance in Rwanda

BACKGROUND

Schistosomiasis is an important snail-borne parasitic disease whose transmission is exacerbated by water resource management activities. In Rwanda, meeting the growing population's demand for food has led to wetlands reclamation for cultivation and increased agrochemical enrichment for crop production. However, the ecological consequences of agrochemical enrichment on schistosomiasis transmission remain unexplored.

METHODS

A malacological survey was conducted in 71 villages selected from 15 schistosomiasis endemic districts. Snail sampling was done in wetlands used for agriculture, along lake Kivu and at constructed multipurpose water dams. Water physico-chemical parameters were collected at all snail sampling sites. Analysis of collected data was performed using Xgboost (gain) and Random Forest (mean decrease in accuracy), machine learning techniques, to construct models that evaluate and categorize the importance of all physico-chemical properties on the presence and abundance of intermediate host snails (IHS).

RESULTS

Different sets of parameters were relevant for the presence and abundance of Biomphalaria spp. and/or Bulinus spp. snails. Electrical conductivity, elevation, magnesium and lead content were deemed to shape the presence and abundance of Bulinus spp. snails. The impact of phosphate ion concentration, ammonia ions, total nitrogen and total organic carbon levels mirrored their importance towards the presence and abundance of Biomphalaria spp. Factors such as pH, electric conductivity, total nitrogen content and total organic carbon influenced the coexistence of both species. Our study highlights the value of integrating a wide range of predictor variables, enabling effective variable selection to uncover important predictors of snail distribution.

CONCLUSION

The results suggest that agrochemical compounds can enhance the abundance of IHS leading to an increased risk of Schistosoma transmission. Snail surveillance could therefore be integrated into agricultural expansion projects in our match towards schistosomiasis elimination. Recognizing the impact of agrochemicals on IHS is crucial for minimizing schistosomiasis transmission among those working in wetlands while meeting the growing need for food.

Machine Learning-Based Software for Predicting Pseudomonas spp. Growth Dynamics in Culture Media

In predictive microbiology, both primary and secondary models are widely used to estimate microbial growth, often applied through two-step or one-step modelling approaches. This study focused on developing a tool to predict the growth of Pseudomonas spp., a prominent bacterial genus in food spoilage, by applying machine learning regression models, including Support Vector Regression (SVR), Random Forest Regression (RFR) and Gaussian Process Regression (GPR). The key environmental factors-temperature, water activity, and pH-served as predictor variables to model the growth of Pseudomonas spp. in culture media. To assess model performance, these machine learning approaches were compared with traditional models, namely the Gompertz, Logistic, Baranyi, and Huang models, using statistical indicators such as the adjusted coefficient of determination (R2adj) and root mean square error (RMSE). Machine learning models provided superior accuracy over traditional approaches, with R2adj values from 0.834 to 0.959 and RMSE values between 0.005 and 0.010, showcasing their ability to handle complex growth patterns more effectively. GPR emerged as the most accurate model for both training and testing datasets. In external validation, additional statistical indices (bias factor, Bf: 0.998 to 1.047; accuracy factor, Af: 1.100 to 1.167) further supported GPR as a reliable alternative for microbial growth prediction. This machine learning-driven approach bypasses the need for the secondary modelling step required in traditional methods, highlighting its potential as a robust tool in predictive microbiology.

Challenges and opportunities for risk- and systems-based control of Listeria monocytogenes transmission through food

Listeria monocytogenes contamination of ready-to-eat (RTE) food products and food-associated built environments (e.g., processing facilities) represents a food safety issue with major public health and business risk implications. A number of factors make L. monocytogenes control a particular challenge, including (i) its frequent presence in different environments, (ii) its propensity for establishing persistence in food-associated environments, (iii) its ability to grow under a variety of stressful conditions, and (iv) its ability to cause severe illness, particularly in immunocompromised individuals and pregnant people. Key sources of L. monocytogenes contamination of RTE foods are food-associated built environments. However, raw materials can also be an important source, particularly for products without a "kill step" (e.g., fresh produce, raw dairy products, cold-smoked seafood). While certain RTE foods (e.g., deli meats, soft cheeses, produce) have commonly been linked to listeriosis outbreaks, cases, and recalls, a number of factors will influence the specific public health risk a given RTE food represents, including the likelihood of contamination, ability to support L. monocytogenes growth, and consumer-related factors (including consumption by pregnant or immunocompromised individuals). Consequently, a risk-based approach presents the most appropriate strategy to minimize the public health and business impact of L. monocytogenes. Key challenges to control L. monocytogenes include (i) development and implementation of food safety systems that prevent L. monocytogenes persistence in food-associated built environments, (ii) minimizing L. monocytogenes contamination of raw material sources, (iii) implementation of effective root cause analysis procedures, (iv) minimizing L. monocytogenes growth in finished product, and (v) consumer education.

Development of a pig wean-quality score using machine-learning algorithms to characterize and classify groups with high mortality risk under field conditions

Mortality during the post-weaning phase is a critical indicator of swine production system performance, influenced by a complex interaction of multiple factors of the epidemiological triad. This study leveraged retrospective data from 1723 groups of pigs marketed within a US swine production system to develop a Wean-Quality Score (WQS) using machine learning techniques. The study evaluated three machine learning models, Random Forest, Support Vector Machine, and Gradient Boosting Machine, to classify groups having high or low 60-day mortality, where high mortality groups represented 25 % of the groups among the study population with the highest mortality values (n=431; 60-day mortality=9.98 %), and the remaining 75 % of the groups were of low mortality (n=1292; 60-day mortality=2.75 %). The best-performing model, Random Forest (RF), outperformed the other ML models in terms of accuracy (0.90), sensitivity (0.84), and specificity (0.92) metrics, and was then selected for further analysis, which consisted of creating the WQS and ranking the most important factors for classifying groups as high or low mortality. The most important factors ranked through the RF model to classify groups with high mortality were pre-weaning mortality, weaning age, average parity of litters in sow farms, and PRRS status. Additionally, stocking conditions such as stocking density and time to fill the barn were important predictors of high mortality. The WQS was developed and correlated (r = 0.74) with the actual 60-day mortality of the groups, offering a valuable tool for assessing post-weaning survivability in swine production systems before weaning. This study highlights the potential of machine learning and comprehensive data utilization to improve the assessment and management of weaned pig quality in commercial swine production, which producers can utilize to identify and intervene in groups, according to the WQS.

Innovative strategies for protein content determination in dried laver (Porphyra spp.): Evaluation of preprocessing methods and machine learning algorithms through short-wave infrared imaging

In this study, we explored the application of Short-Wave Infrared (SWIR) hyperspectral imaging combined with Competitive Adaptive Reweighted Sampling (CARS) and advanced regression models for the non-destructive assessment of protein content in dried laver. Utilizing a spectral range of 900-1700 nm, we aimed to refine the quality control process by selecting informative wavelengths through CARS and applying various preprocessing techniques (standard normal variate [SNV], Savitzky-Golay filtering [SG], Orthogonal Signal Correction [OSC], and StandardScaler [SS]) to enhance the model's accuracy. The SNV-OSC-StandardScaler- Support vector regression (SVR) model trained on CARS-selected wavelengths significantly outperformed the other configurations, achieving a prediction determination coefficient (Rp2) of 0.9673, root mean square error of prediction of 0.4043, and residual predictive deviation of 5.533. These results highlight SWIR hyperspectral imaging's potential as a rapid and precise tool for assessing dried laver quality, aiding food industry quality control and dried laver market growth.

Application of machine learning approaches to predict ammonium nitrogen transport in different soil types and evaluate the contribution of control factors

The loss of nitrogen in soil damages the environment. Clarifying the mechanism of ammonium nitrogen (NH4+-N) transport in soil and increasing the fixation of NH4+-N after N application are effective methods for improving N use efficiency. However, the main factors are not easily identified because of the complicated transport and retardation factors in different soils. This study employed machine learning (ML) to identify the main influencing factors that contribute to the retardation factor (Rf) of NH4+-N in soil. First, NH4+-N transport in the soil was investigated using column experiments and a transport model. The Rf (1.29 - 17.42) was calculated and used as a proxy for the efficacy of NH4+-N transport. Second, the physicochemical parameters of the soil were determined and screened using lasso and ridge regressions as inputs for the ML model. Third, six machine learning models were evaluated: Adaptive Boosting, Extreme Gradient Boosting (XGB), Random Forest, Gradient Boosting Regression, Multilayer Perceptron, and Support Vector Regression. The optimal ML model of the XGB model with a low mean absolute error (0.81), mean squared error (0.50), and high test r2 (0.97) was obtained by random sampling and five-fold cross-validation. Finally, SHapely Additive exPlanations, entropy-based feature importance, and permutation characteristic importance were used for global interpretation. The cation exchange capacity (CEC), total organic carbon (TOC), and Kaolin had the greatest effects on NH4+-N transport in the soil. The accumulated local effect offered a fundamental insight: When CEC > 6 cmol+ kg-1, and TOC > 40 g kg-1, the maximum resistance to NH4+-N transport within the soil was observed. This study provides a novel approach for predicting the impact of the soil environment on NH4+-N transport and guiding the establishment of an early-warning system of nutrient loss.

Predicting the Bioaccessibility of Soil Cd, Pb, and As with Advanced Machine Learning for Continental-Scale Soil Environmental Criteria Determination in China

Investigating the bioaccessibility of harmful inorganic elements in soil is crucial for understanding their behavior in the environment and accurately assessing the environmental risks associated with soil. Traditional batch experimental methods and linear models, however, are time-consuming and often fall short in precisely quantifying bioaccessibility. In this study, using 937 data points gathered from 56 journal articles, we developed machine learning models for three harmful inorganic elements, namely, Cd, Pb, and As. After thorough analysis, the model optimized through a boosting ensemble strategy demonstrated the best performance, with an average R 2 of 0.95 and an RMSE of 0.25. We further employed SHAP values in conjunction with quantitative analysis to identify the key features that influence bioaccessibility. By utilizing the developed integrated models, we carried out predictions for 3002 data points across China, clarifying the bioaccessibility of cadmium (Cd), lead (Pb), and arsenic (As) in the soils of various sites and constructed a comprehensive spatial distribution map of China using the inverse distance weighting (IDW) interpolation method. Based on these findings, we further derived the soil environmental standards for metallurgical sites in China. Our observations from the collected data indicate a reduction in the number of sites exceeding the standard levels for Cd, Pb, and As in mining/smelting sites from 5, 58, and 14 to 1, 24, and 7, respectively. This research offers a precise and scientific approach for cross-regional risk assessment at the continental scale and lays a solid foundation for soil environmental management.

A comparative study on bath and horn ultrasound-assisted modification of bentonite and their effects on the bleaching efficiency of soybean and sunflower oil: Machine learning as a new approach for mathematical modeling

In this study, the effect of high-power bath and horn ultrasound at different powers on specific surface area (S BET), total pore volume (V total), and average pore volume (D ave) of bleaching clay was examined. After subjecting the bleaching clay to ultrasonication treatment, the SBET values demonstrated an escalation from 31.4 ± 2.7 m2 g-1 to 59.8 ± 3.1 m2 g-1 for HU200BC, 143.8 ± 3.9 m2 g-1 for HU400BC, 54.4 ± 3.6 m2 g-1 for BU400BC, and 137.5 ± 2.8 m2 g-1 for BU800BC. The mean pore diameter (D ave) declined from 29.7 ± 0.14 nm in bleaching clay to 11.3 ± 0.13 nm in HU200BC, 8.3 ± 0.12 nm in HU400BC, 16.7 ± 0.14 nm in BU400BC, and 9.6 ± 0.12 nm in BU800BC. Therefore, horn ultrasound-treated bleaching clay significantly increased S BET and V total, indicating improved adsorption capacity. Moreover, to establish the relationship between bleaching parameters, seven multi-output ML regression models of Feedforward Neural Network (FNN), Random Forest (RF), Support Vector Regression (SVR), Multi-Task Lasso, Ridge regression, Extreme Gradient Boosting (XGBoost), and Gradient Boosting are used, and compared with response surface methodology (RSM). ML has revolutionized the understanding of complex relationships between ultrasonic parameters, oil color, and pigment degradation, providing insights into how various factors such as temperature, ultrasonic power, and time can influence the bleaching process, ultimately enhancing the efficiency and precision of the treatment. The XGBoost model showed outstanding performance in predicting the target variables with a high R 2-train up to 1, R 2-test up to .983, and a minimum mean absolute error (MAE) of 0.498. The lower error between the predicted and experimental values implies the superiority of the XGBoost model to predict outcomes rather than RSM. It represents the suitability of bath ultrasound as a mild condition for low-pigmented oil bleaching. Finally, the Bayesian optimization method in conjunction with XGBoost was used to optimize the amount of bleaching clay and energy consumption, and its performance was compared with RSM. It was observed that the consumption of bleaching clay was reduced by approximately 60% for sunflower oil and 30%-35% for soybean oil.

The suspension stability of nanoplastics in aquatic environments revealed using meta-analysis and machine learning

Nanoplastics (NPs) aggregation determines their bioavailability and risks in natural aquatic environments, which is driven by multiple environmental and polymer factors. The back propagation artificial neural network (BP-ANN) model in machine learning (R2 = 0.814) can fit the complex NPs aggregation, and the feature importance was in the order of surface charge of NPs > dissolved organic matter (DOM) > functional group of NPs > ionic strength and pH > concentration of NPs. Meta-analysis results specified low surface charge (0 ≤ |ζ| < 10 mV) of NPs, low concentration (< 1 mg/L) and low molecular weight (< 10 kg/mol) of DOM, NPs with amino groups, high ionic strength (IS > 700 mM) and acidic solution, and high concentration (≥ 20 mg/L) of NPs with smaller size (< 100 nm) contribute to NPs aggregation, which is consistent with the prediction in machine learning. Feature interaction synergistically (e.g., DOM and pH) or antagonistically (e.g., DOM and cation potential) changed NPs aggregation. Therefore, NPs were predicted to aggregate in the dry period and estuary of Poyang Lake. Research on aggregation of NPs with different particle size,shapes, and functional groups, heteroaggregation of NPs with coexisting particles and aging effects should be strengthened in the future. This study supports better assessments of the NPs fate and risks in environments.

Machine learning application for predicting key properties of activated carbon produced from lignocellulosic biomass waste with chemical activation

The successful application of gradient boosting regression (GBR) in machine learning to forecast surface area, pore volume, and yield in biomass-derived activated carbon (AC) production underscores its potential for enhancing manufacturing processes. The GBR model, collecting 17 independent variables for two-step activation (2-SA) and 14 for one-step activation (1-SA), demonstrates effectiveness across three datasets-1-SA, 2-SA, and a combined dataset. Notably, in 1-SA, the GBR model yields R2 values of 0.76, 0.90, and 0.83 for TPV, yield, and SSA respectively, and records R2 of 0.90 and 0.91 for yield in 2-SA and combined datasets. The model highlights the significance of the soaking procedure alongside activation temperature in shaping AC properties with 1-SA or 2-SA, illustrating machine learning's potential in optimizing AC production processes.

Predicting the Climate Impact of Healthcare Facilities Using Gradient Boosting Machines

Health care accounts for 9-10% of greenhouse gas (GHG) emissions in the United States. Strategies for monitoring these emissions at the hospital level are needed to decarbonize the sector. However, data collection to estimate emissions is challenging, especially for smaller hospitals. We explored the potential of gradient boosting machines (GBM) to impute missing data on resource consumption in the 2020 survey of a consortium of 283 hospitals participating in Practice Greenhealth. GBM imputed missing values for selected variables in order to predict electricity use and beef consumption (R2=0.82) and anesthetic gas desflurane use (R2=0.51), using administrative data readily available for most hospitals. After imputing missing consumption data, estimated GHG emissions associated with these three examples totaled over 3 million metric tons of CO2 equivalent emissions (MTCO2e). Specifically, electricity consumption had the largest total carbon footprint (2.4 MTCO2e), followed by beef (0.6 million MTCO2e) and desflurane consumption (0.03 million MTCO2e) across the 283 hospitals. The approach should be applicable to other sources of hospital GHGs in order to estimate total emissions of individual hospitals and to refine survey questions to help develop better intervention strategies.

An interpretable machine learning model based on contrast-enhanced CT parameters for predicting treatment response to conventional transarterial chemoembolization in patients with hepatocellular carcinoma

OBJECTIVE

To explore the potential of pre-therapy computed tomography (CT) parameters in predicting the treatment response to initial conventional TACE (cTACE) in intermediate-stage hepatocellular carcinoma (HCC) and develop an interpretable machine learning model.

METHODS

This retrospective study included 367 patients with intermediate-stage HCC who received cTACE as first-line therapy from three centers. We measured the mean attenuation values of target lesions on multi-phase contrast-enhanced CT and further calculated three CT parameters, including arterial (AER), portal venous (PER), and arterial portal venous (APR) enhancement ratios. We used logistic regression analysis to select discriminative features and trained three machine learning models via 5-fold cross-validation. The performance in predicting treatment response was evaluated in terms of discrimination, calibration, and clinical utility. Afterward, a Shapley additive explanation (SHAP) algorithm was leveraged to interpret the outputs of the best-performing model.

RESULTS

The mean diameter, ECOG performance status, and cirrhosis were the important clinical predictors of cTACE treatment response, by multiple logistic regression. Adding the CT parameters to clinical variables showed significant improvement in performance (net reclassification index, 0.318, P < 0.001). The Random Forest model (hereafter, RF-combined model) integrating CT parameters and clinical variables demonstrated the highest performance on external validation dataset (AUC of 0.800). The decision curve analysis illustrated the optimal clinical benefits of RF-combined model. This model could successfully stratify patients into responders and non-responders with distinct survival (P = 0.001).

CONCLUSION

The RF-combined model can serve as a robust and interpretable tool to identify the appropriate crowd for cTACE sessions, sparing patients from receiving ineffective and unnecessary treatments.

Machine learning based prediction and experimental validation of arsenite and arsenate sorption on biochars

Arsenic (As) contamination in water is a significant environmental concern with profound implications for human health. Accurate prediction of the adsorption capacity of arsenite [As(III)] and arsenate [As(V)] on biochar is vital for the reclamation and recycling of polluted water resources. However, comprehending the intricate mechanisms that govern arsenic accumulation on biochar remains a formidable challenge. Data from the literature on As adsorption to biochar was compiled and fed into machine learning (ML) based modelling algorithms, including AdaBoost, LGBoost, and XGBoost, in order to build models to predict the adsorption efficiency of As(III) and As(V) to biochar, based on the compositional and structural properties. The XGBoost model showed superior accuracy and performance for prediction of As adsorption efficiency (for As(III): coefficient of determination (R2) = 0.93 and root mean square error (RMSE) = 1.29; for As(V), R2 = 0.99, RMSE = 0.62). The initial concentrations of As(III) and As(V) as well as the dosage of the adsorbent were the most significant factors influencing adsorption, explaining 48 % and 66 % of the variability for As(III) and As(V), respectively. The structural properties and composition of the biochar explained 12 % and 40 %, respectively, of the variability of As(III) adsorption, and 13 % and 21 % of that of As(V). The XGBoost models were validated using experimental data. R2 values were 0.9 and 0.84, and RMSE values 6.5 and 8.90 for As(III) and As(V), respectively. The ML approach can be a valuable tool for improving the treatment of inorganic As in aqueous environments as it can help estimate the optimal adsorption conditions of As in biochar-amended water, and serve as an early warning for As-contaminated water.

Relationship of the Poultry Microbiome to Pathogen Colonization, Farm Management, Poultry Production, and Foodborne Illness Risk Assessment

Despite the continuous progress in food science and technology, the global burden of foodborne illnesses remains substantial, with pathogens in food causing millions of infections each year. Traditional microbiological culture methods are inadequate in detecting the full spectrum of these microorganisms, highlighting the need for more comprehensive detection strategies. This review paper aims to elucidate the relationship between foodborne pathogen colonization and the composition of the poultry microbiome, and how this knowledge can be used for improved food safety. Our review highlights that the relationship between pathogen colonization varies across different sections of the poultry microbiome. Further, our review suggests that the microbiome profile of poultry litter, farm soil, and farm dust may serve as potential indicators of the farm environment's food safety issues. We also agree that the microbiome of processed chicken samples may reveal potential pathogen contamination and food quality issues. In addition, utilizing predictive modeling techniques on the collected microbiome data, we suggest establishing correlations between particular taxonomic groups and the colonization of pathogens, thus providing insights into food safety, and offering a comprehensive overview of the microbial community. In conclusion, this review underscores the potential of microbiome analysis as a powerful tool in food safety, pathogen detection, and risk assessment.

Survival strategies of Listeria monocytogenes to environmental hostile stress: biofilm formation and stress responses

Listeria monocytogenes is a critical foodborne pathogen that causes listeriosis and threatens public health. This pathogenic microorganism forms a transmission cycle in nature, food industry, and humans, expanding the areas of contamination among them and influencing food safety. L. monocytogenes forms biofilms to protect itself and promotes survival through stress responses to the various stresses (e.g., temperature, pH, and antimicrobial agents) that may be inflicted during food processing. Biofilms and mechanisms of resistance to hostile external or general stresses allow L. monocytogenes to survive despite a variety of efforts to ensure food safety. The current review article focuses on biofilm formation, resistance mechanisms through biofilms, and external specific or general stress responses of L. monocytogenes to help understand the unexpected survival rates of this bacterium; it also proposes the use of obstacle technology to effectively cope with it in the food industry.

Field Implementation of Forecasting Models for Predicting Nursery Mortality in a Midwestern US Swine Production System

The performance of five forecasting models was investigated for predicting nursery mortality using the master table built for 3242 groups of pigs (~13 million animals) and 42 variables, which concerned the pre-weaning phase of production and conditions at placement in growing sites. After training and testing each model's performance through cross-validation, the model with the best overall prediction results was the Support Vector Machine model in terms of Root Mean Squared Error (RMSE = 0.406), Mean Absolute Error (MAE = 0.284), and Coefficient of Determination (R2 = 0.731). Subsequently, the forecasting performance of the SVM model was tested on a new dataset containing 72 new groups, simulating ongoing and near real-time forecasting analysis. Despite a decrease in R2 values on the new dataset (R2 = 0.554), the model demonstrated high accuracy (77.78%) for predicting groups with high (>5%) or low (<5%) nursery mortality. This study demonstrated the capability of forecasting models to predict the nursery mortality of commercial groups of pigs using pre-weaning information and stocking condition variables collected post-placement in nursery sites.

Prediction of Pseudomonas spp. Population in Food Products and Culture Media Using Machine Learning-Based Regression Methods

Machine learning approaches are alternative modelling techniques to traditional modelling equations used in predictive food microbiology and utilise algorithms to analyse large datasets that contain information about microbial growth or survival in various food matrices. These approaches leverage the power of algorithms to extract insights from the data and make predictions regarding the behaviour of microorganisms in different food environments. The objective of this study was to apply various machine learning-based regression methods, including support vector regression (SVR), Gaussian process regression (GPR), decision tree regression (DTR), and random forest regression (RFR), to estimate bacterial populations. In order to achieve this, a total of 5618 data points for Pseudomonas spp. present in food products (beef, pork, and poultry) and culture media were gathered from the ComBase database. The machine learning algorithms were applied to predict the growth or survival behaviour of Pseudomonas spp. in food products and culture media by considering predictor variables such as temperature, salt concentration, water activity, and acidity. The suitability of the algorithms was assessed using statistical measures such as coefficient of determination (R²), root mean square error (RMSE), bias factor (Bf), and accuracy (A_f). Each of the regression algorithms showed appropriate estimation capabilities with R² ranging from 0.886 to 0.913, RMSE from 0.724 to 0.899, B_f from 1.012 to 1.020, and A_f from 1.086 to 1.101 for each food product and culture medium. Since the predictive capability of RFR was the best among the algorithms, externally collected data from the literature were used for RFR. The external validation process showed statistical indices of B_f ranging from 0.951 to 1.040 and A_f ranging from 1.091 to 1.130, indicating that RFR can be used for predicting the survival and growth of microorganisms in food products. Therefore, machine learning approaches can be considered as an alternative to conventional modelling methods in predictive microbiology. However, it is important to highlight that the prediction power of the machine learning regression method directly depends on the dataset size, and it requires a large dataset to be employed for modelling. Therefore, the modelling work of this study can only be used for the prediction of Pseudomonas spp. in specific food products (beef, pork, and poultry) and culture medium with certain conditions where a large dataset is available.

Characterization of Bacillus Strains from Natural Honeybee Products with High Keratinolytic Activity and Antimicrobial Potential

Two efficient feather-degrading bacteria were isolated from honeybee samples and identified as Bacillus sonorensis and Bacillus licheniformis based on 16S rRNA and genome sequencing. The strains were able to grow on chicken feathers as the sole carbon and nitrogen sources and degraded the feathers in a few days. The highest keratinase activity was detected by the B. licheniformis CG1 strain (3800 U × mL^-1), followed by B. sonorensis AB7 (1450 U × mL^-1). Keratinase from B. licheniformis CG1 was shown to be active across a wide range of pH, potentially making this strain advantageous for further industrial applications. All isolates displayed antimicrobial activity against Micrococcus luteus; however, only B. licheniformis CG1 was able to inhibit the growth of Mycobacterium smegmatis. In silico analysis using BAGEL and antiSMASH identified gene clusters associated with the synthesis of non-ribosomal peptide synthetases (NRPS), polyketide synthases (PKSs) and/or ribosomally synthesized and post-translationally modified peptides (RiPPs) in most of the Bacillus isolates. B. licheniformis CG1, the only strain that inhibited the growth of the mycobacterial strain, contained sequences with 100% similarity to lichenysin (also present in the other isolates) and lichenicidin (only present in the CG1 strain). Both compounds have been described to display antimicrobial activity against distinct bacteria. In summary, in this work, we have isolated a strain (B. licheniformis CG1) with promising potential for use in different industrial applications, including animal nutrition, leather processing, detergent formulation and feather degradation.

Wind Speed and Landscape Context Mediate Campylobacter Risk among Poultry Reared in Open Environments

Foodborne pathogens cause over 9 million illnesses in the United States each year, and Campylobacter from chickens is the largest contributor. Rearing poultry outdoors without the use of antibiotics is becoming an increasingly popular style of farming; however, little is understood about how environmental factors and farm management alter pathogen prevalence. Our survey of 27 farms in California, Oregon, Washington, and Idaho, USA, revealed a diversity of management practices used to rear poultry in the open environment. Here, we assess environmental and management factors that impact Campylobacter spp. prevalence in 962 individual chicken fecal samples from 62 flocks over a three-year period. We detected Campylobacter spp. in 250/962 (26.0%) of fecal samples screened, in 69.4% (43/62) of flocks, and on 85.2% (23/27) of farms. We found that Campylobacter spp. prevalence was predicted to increase in poultry on farms with higher average wind speeds in the seven days preceding sampling; on farms embedded in more agricultural landscapes; and in flocks typified by younger birds, more rotations, higher flock densities, and the production of broilers. Collectively, our results suggest that farms in areas with higher wind speeds and more surrounding agriculture face greater risk of Campylobacter spp. introduction into their flocks.

A Pilot Study: the Development of a Facility-Associated Microbiome and Its Association with the Presence of Listeria Spp. in One Small Meat Processing Facility

Microbial communities which persist in food processing facilities may have a detrimental impact on food safety and spoilage. In meat processing, Listeria monocytogenes is an organism of concern due to its ability to cause significant human illnesses and persist in refrigerated environments. The microbial ecology of Listeria spp. in small meat processing facilities has not been well characterized. Therefore, we collected samples from a newly constructed meat processing facility as an opportunity to investigate several research objectives: (i) to determine whether a stable, consistent microbiome develops in a small meat processing facility during the first 18 months of operation, (ii) to evaluate the environmental factors that drive microbial community formation, and (iii) to elucidate the relationship between microbial communities and the presence of Listeria species. We evaluated microbiomes using 16S rRNA gene sequencing and Listeria presence using quantitative PCR. We demonstrated that microbial communities differentiate by the functional room type, which is representative of several environmental differences such as temperature, sources of microbes, and activity. Temperature was an especially important factor; in rooms with low temperatures, communities were dominated by psychotrophs, especially Pseudomonas, while warmer rooms supported greater diversity. A stable core community formed in facility drains, indicating that mechanisms which cause persistence are present in the communities. The overall presence of Listeria in the facility was low but could be tied to specific organisms within a room, and the species of Listeria could be stratified by room function. IMPORTANCE This study provides critical knowledge to improve meat safety and quality from small meat processing facilities. Principally, it demonstrates the importance of facility design and room condition to the development of important microbial communities; temperature, sanitation regimen, and physical barriers all influence the ability of microorganisms to join the stable core community. It also demonstrates a relationship between the microbial community and Listeria presence in the facility, showing the importance of managing facility sanitation plans for not only pathogens, but also the general facility microbiome.

Predicting pesticide dissipation half-life intervals in plants with machine learning models

Pesticide dissipation half-life in plants is an important factor to assessing environmental fate of pesticides and establishing pre-harvest intervals critical to good agriculture practices. However, empirically measured pesticide dissipation half-lives are highly variable and the accurate prediction with models is challenging. This study utilized a dataset of pesticide dissipation half-lives containing 1363 datapoints, 311 pesticides, 10 plant types, and 4 plant component classes. Novel dissipation half-life intervals were proposed and predicted to account for high variations in empirical data. Four machine learning models (i.e., gradient boosting regression tree [GBRT], random forest [RF], supporting vector classifier [SVC], and logistic regression [LR]) were developed to predict dissipation half-life intervals using extended connectivity fingerprints (ECFP), temperature, plant type, and plant component class as model inputs. GBRT-ECFP had the best model performance with F1-micro_binary score of 0.698 ± 0.010 for the binary classification compared with other machine learning models (e.g., LR-ECFP, F1-micro_binary= 0.662 ± 0.009). Feature importance analysis of molecular structures in the binary classification identified aromatic rings, carbonyl group, organophosphate, =C-H, and N-containing heterocyclic groups as important substructures related to pesticide dissipation half-lives. This study suggests the utility of machine learning models in assessing the environmental fate of pesticides in agricultural crops.

An Intelligent Deep Learning Model for CO 2 Adsorption Prediction

In this paper, we propose a supervised deep learning neural network (D-CNN) approach to predict CO2 adsorption form the textural and compositional features of biomass porous carbon waste and adsorption features. Both the textural and compositional features of biomass porous carbon waste are utilized as inputs for the D-CNN architecture. A deep learning neural network (D-CNN) is proposed to predict the adsorption rate of ${\text{CO}}_2$ on zeolites. The adsorbed amount will be classified and predicted by the D-CNN. Three tree machine learning models, namely, gradient decision model (GDM), scalable boosting tree model (SBT), and gradient variant decision tree model (GVD), were fused. A feature importance metric was proposed using feature permutation, and the effect of each feature on the target output variable was investigated. The important extracted features from the three employed model were fused and used as the fusion feature set in our proposed model: fusion matrix deep learning model (FMDL). A dataset of 1400 data items, on adsorbent type and various adsorption pressure, is used as inputs for the D-CNN model. Comparison of the proposed model is done against the three tree models, which utilizes a single training layer. The error measure of the D-CNN and the tree model architectures utilize the mean square error confirming the efficiency of 0.00003 for our model, 0.00062 for the SBT, 0.00091 for the GDM, and 0.00098 for the GVD, after 150 epochs. The produced weight matrix was able to predict the ${\text{CO}}_2$ adsorption under diverse process settings with high accuracy of 96.4%.

Predicting crop root concentration factors of organic contaminants with machine learning models

Accurate prediction of uptake and accumulation of organic contaminants by crops from soils is essential to assessing human exposure via the food chain. However, traditional empirical or mechanistic models frequently show variable performance due to complex interactions among contaminants, soils, and plants. Thus, in this study different machine learning algorithms were compared and applied to predict root concentration factors (RCFs) based on a dataset comprising 57 chemicals and 11 crops, followed by comparison with a traditional linear regression model as the benchmark. The RCF patterns and predictions were investigated by unsupervised t-distributed stochastic neighbor embedding and four supervised machine learning models including Random Forest, Gradient Boosting Regression Tree, Fully Connected Neural Network, and Supporting Vector Regression based on 15 property descriptors. The Fully Connected Neural Network demonstrated superior prediction performance for RCFs (R² =0.79, mean absolute error [MAE] = 0.22) over other machine learning models (R² =0.68-0.76, MAE = 0.23-0.26). All four machine learning models performed better than the traditional linear regression model (R² =0.62, MAE = 0.29). Four key property descriptors were identified in predicting RCFs. Specifically, increasing root lipid content and decreasing soil organic matter content increased RCFs, while increasing excess molar refractivity and molecular volume of contaminants decreased RCFs. These results show that machine learning models can improve prediction accuracy by learning nonlinear relationships between RCFs and properties of contaminants, soils, and plants.

Assessing machine learning techniques in forecasting lumpy skin disease occurrence based on meteorological and geospatial features

Lumpy skin disease virus (LSDV) causes an infectious disease in cattle. Due to its direct relationship with the survival of arthropod vectors, geospatial and climatic features play a vital role in the epidemiology of the disease. The objective of this study was to assess the ability of some machine learning algorithms to forecast the occurrence of LSDV infection based on meteorological and geological attributes. Initially, ExtraTreesClassifier algorithm was used to select the important predictive features in forecasting the disease occurrence in unseen (test) data among meteorological, animal population density, dominant land cover, and elevation attributes. Some machine learning techniques revealed high accuracy in predicting the LSDV occurrence in test data (up to 97%). In terms of area under curve (AUC) and F1 performance metric scores, the artificial neural network (ANN) algorithm outperformed other machine learning methods in predicting the occurrence of LSDV infection in unseen data with the corresponding values of 0.97 and 0.94, respectively. Using this algorithm, the model consisted of all predictive features and the one which only included meteorological attributes as important features showed similar predictive performance. According to the findings of this research, ANN can be used to forecast the occurrence of LSDV infection with high precision using geospatial and meteorological parameters. Applying the forecasting power of these methods could be a great help in conducting screening and awareness programs, as well as taking preventive measures like vaccination in areas where the occurrence of LSDV infection is a high risk.

A Machine Learning Based Model for Energy Usage Peak Prediction in Smart Farms

Context: Energy utilization is one of the most closely related factors affecting many areas of the smart farm, plant growth, crop production, device automation, and energy supply to the same degree. Recently, 4th industrial revolution technologies such as IoT, artificial intelligence, and big data have been widely used in smart farm environments to efficiently use energy and control smart farms’ conditions. In particular, machine learning technologies with big data analysis are actively used as one of the most potent prediction methods supporting energy use in the smart farm. Purpose: This study proposes a machine learning-based prediction model for peak energy use by analyzing energy-related data collected from various environmental and growth devices in a smart paprika farm of the Jeonnam Agricultural Research and Extension Service in South Korea between 2019 and 2021. Scientific method: To find out the most optimized prediction model, comparative evaluation tests are performed using representative ML algorithms such as artificial neural network, support vector regression, random forest, K-nearest neighbors, extreme gradient boosting and gradient boosting machine, and time series algorithm ARIMA with binary classification for a different number of input features. Validate: This article can provide an effective and viable way for smart farm managers or greenhouse farmers who can better manage the problem of agricultural energy economically and environmentally. Therefore, we hope that the recommended ML method will help improve the smart farm’s energy use or their energy policies in various fields related to agricultural energy. Conclusion: The seven performance metrics including R-squared, root mean squared error, and mean absolute error, are associated with these two algorithms. It is concluded that the RF-based model is more successful than in the pre-others diction accuracy of 92%. Therefore, the proposed model may be contributed to the development of various applications for environment energy usage in a smart farm, such as a notification service for energy usage peak time or an energy usage control for each device.

Machine learning models for prognosis prediction in endodontic microsurgery

OBJECTIVES

This study aimed to establish and validate machine learning models for prognosis prediction in endodontic microsurgery, avoiding treatment failure and supporting clinical decision-making.

METHODS

A total of 234 teeth from 178 patients were included in this study. We developed gradient boosting machine (GBM) and random forest (RF) models. For each model, 80% of the data were randomly selected for the training set and the remaining 20% were used as the test set. A stratified 5-fold cross-validation approach was used in model training and testing. Correlation analysis and importance ranking were conducted for feature selection. The predictive accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), F1 score, and the area under the curve (AUC) of receiver operating characteristic (ROC) curves were calculated to evaluate the predictive performance.

RESULTS

There were eight important predictors, including tooth type, lesion size, type of bone defect, root filling density, root filling length, apical extension of post, age, and sex. For the GBM model, the predictive accuracy was 0.80, with a sensitivity of 0.92, specificity of 0.71, PPV of 0.71, NPV of 0.92, F1 of 0.80/0.80, and AUC of 0.88. For the RF model, the accuracy was 0.80, with a sensitivity of 0.85, specificity of 0.76, PPV of 0.73, NPV of 0.87, F1 of 0.79/0.81, and AUC of 0.83.

CONCLUSIONS

The trained models were developed by eight common variables, showing the potential ability to predict the prognosis of endodontic microsurgery. The GBM model outperformed the RF model slightly on our dataset.

CLINICAL SIGNIFICANCE

Clinicians can use machine learning models for preoperative analysis in endodontic microsurgery. The models are expected to improve the efficiency of clinical decision-making and assist in clinician-patient communication.

Machine learning approach for predicting the antifungal effect of gilaburu (Viburnum opulus) fruit extracts on Fusarium spp. isolated from diseased potato tubers

This work addresses the mathematical model building to detect the diameter of the inhibition zone of gilaburu (Viburnum opulus L.) extract against eight different Fusarium strains isolated from diseased potato tubers. Gilaburu extracts were obtained with acetone, ethanol or methanol. The isolated Fusarium strains were: F. solani, F. oxysporum, F. sambucinum, F. graminearum, F. coeruleum, F. sulphureum, F. auneaceum and F. culmorum. In general, it was observed that ethanolic extracts showed highest antifungal activity. The antifungal activity of extracts was evaluated with machine learning (ML) methods. Several ML methods (classification and regression trees (CART), support vector machines (SVM), k-Nearest Neighbors (k-NN), artificial neural network (ANN), ensemble algorithms (EA), AdaBoost (AB) algorithm, gradient boosting (GBM) algorithm, random forests (RF) bagging algorithm and extra trees (ET)) were applied and compared for modeling fungal growth. From this research, it is clear that ML methods have the lowest error level. As a result, ML methods are reliable, fast, and cheap tools for predicting the antifungal activity of gilaburu extracts. These encouraging results will attract more research efforts to implement ML into the field of food microbiology instead of traditional methods.

A Comparative Analysis of Machine Learning Models for the Prediction of Insurance Uptake in Kenya

The role of insurance in financial inclusion and economic growth, in general, is immense and is increasingly being recognized. However, low uptake impedes the growth of the sector, hence the need for a model that robustly predicts insurance uptake among potential clients. This study undertook a two phase comparison of machine learning classifiers. Phase I had eight machine learning models compared for their performance in predicting the insurance uptake using 2016 Kenya FinAccessHousehold Survey data. Taking Phase I as a base in Phase II, random forest and XGBoost were compared with four deep learning classifiers using 2019 Kenya FinAccess Household Survey data. The random forest model trained on oversampled data showed the highest F1-score, accuracy, and precision. The area under the receiver operating characteristic curve was furthermore highest for random forest; hence, it could be construed as the most robust model for predicting the insurance uptake. Finally, the most important features in predicting insurance uptake as extracted from the random forest model were income, bank usage, and ability and willingness to support others. Hence, there is a need for a design and distribution of low income based products, and bancassurance could be said to be a plausible channel for the distribution of insurance products.

Applied Machine Learning for Prediction of CO2 Adsorption on Biomass Waste-Derived Porous Carbons

Biomass waste-derived porous carbons (BWDPCs) are a class of complex materials that are widely used in sustainable waste management and carbon capture. However, their diverse textural properties, the presence of various functional groups, and the varied temperatures and pressures to which they are subjected during CO₂ adsorption make it challenging to understand the underlying mechanism of CO₂ adsorption. Here, we compiled a data set including 527 data points collected from peer-reviewed publications and applied machine learning to systematically map CO₂ adsorption as a function of the textural and compositional properties of BWDPCs and adsorption parameters. Various tree-based models were devised, where the gradient boosting decision trees (GBDTs) had the best predictive performance with R² of 0.98 and 0.84 on the training and test data, respectively. Further, the BWDPCs in the compiled data set were classified into regular porous carbons (RPCs) and heteroatom-doped porous carbons (HDPCs), where again the GBDT model had R² of 0.99 and 0.98 on the training and 0.86 and 0.79 on the test data for the RPCs and HDPCs, respectively. Feature importance revealed the significance of adsorption parameters, textural properties, and compositional properties in the order of precedence for BWDPC-based CO₂ adsorption, effectively guiding the synthesis of porous carbons for CO₂ adsorption applications.

Using Machine Learning Algorithms to Predict Hospital Acquired Thrombocytopenia after Operation in the Intensive Care Unit: A Retrospective Cohort Study

Hospital acquired thrombocytopenia (HAT) is a common hematological complication after surgery. This research aimed to develop and compare the performance of seven machine learning (ML) algorithms for predicting patients that are at risk of HAT after surgery. We conducted a retrospective cohort study which enrolled adult patients transferred to the intensive care unit (ICU) after surgery in West China Hospital of Sichuan University from January 2016 to December 2018. All subjects were randomly divided into a derivation set (70%) and test set (30%). ten-fold cross-validation was used to estimate the hyperparameters of ML algorithms during the training process in the derivation set. After ML models were developed, the sensitivity, specificity, area under the curve (AUC), and net benefit (decision analysis curve, DCA) were calculated to evaluate the performances of ML models in the test set. A total of 10,369 patients were included and in 1354 (13.1%) HAT occurred. The AUC of all seven ML models exceeded 0.7, the two highest were Gradient Boosting (GB) (0.834, 0.814-0.853, p < 0.001) and Random Forest (RF) (0.828, 0.807-0.848, p < 0.001). There was no difference between GB and RF (0.834 vs. 0.828, p = 0.293); however, these two were better than the remaining five models (p < 0.001). The DCA revealed that all ML models had high net benefits with a threshold probability approximately less than 0.6. In conclusion, we found that ML models constructed by multiple preoperative variables can predict HAT in patients transferred to ICU after surgery, which can improve risk stratification and guide management in clinical practice.

Mapping foodborne pathogen contamination throughout the conventional and alternative poultry supply chains

Recently, there has been a consumer push for natural and organic food products. This has caused alternative poultry production, such as organic, pasture, and free-range systems, to grow in popularity. Due to the stricter rearing practices of alternative poultry production systems, different types of levels of microbiological risks might be present for these systems when compared to conventional production systems. Both conventional and alternative production systems have complex supply chains that present many different opportunities for flocks of birds or poultry meat to be contaminated with foodborne pathogens. As such, it is important to understand the risks involved during each step of production. The purpose of this review is to detail the potential routes of foodborne pathogen transmission throughout the conventional and alternative supply chains, with a special emphasis on the differences in risk between the two management systems, and to identify gaps in knowledge that could assist, if addressed, in poultry risk-based decision making.

Interpretability Versus Accuracy: A Comparison of Machine Learning Models Built Using Different Algorithms, Performance Measures, and Features to Predict E. coli Levels in Agricultural Water

Since E. coli is considered a fecal indicator in surface water, government water quality standards and industry guidance often rely on E. coli monitoring to identify when there is an increased risk of pathogen contamination of water used for produce production (e.g., for irrigation). However, studies have indicated that E. coli testing can present an economic burden to growers and that time lags between sampling and obtaining results may reduce the utility of these data. Models that predict E. coli levels in agricultural water may provide a mechanism for overcoming these obstacles. Thus, this proof-of-concept study uses previously published datasets to train, test, and compare E. coli predictive models using multiple algorithms and performance measures. Since the collection of different feature data carries specific costs for growers, predictive performance was compared for models built using different feature types [geospatial, water quality, stream traits, and/or weather features]. Model performance was assessed against baseline regression models. Model performance varied considerably with root-mean-squared errors and Kendall's Tau ranging between 0.37 and 1.03, and 0.07 and 0.55, respectively. Overall, models that included turbidity, rain, and temperature outperformed all other models regardless of the algorithm used. Turbidity and weather factors were also found to drive model accuracy even when other feature types were included in the model. These findings confirm previous conclusions that machine learning models may be useful for predicting when, where, and at what level E. coli (and associated hazards) are likely to be present in preharvest agricultural water sources. This study also identifies specific algorithm-predictor combinations that should be the foci of future efforts to develop deployable models (i.e., models that can be used to guide on-farm decision-making and risk mitigation). When deploying E. coli predictive models in the field, it is important to note that past research indicates an inconsistent relationship between E. coli levels and foodborne pathogen presence. Thus, models that predict E. coli levels in agricultural water may be useful for assessing fecal contamination status and ensuring compliance with regulations but should not be used to assess the risk that specific pathogens of concern (e.g., Salmonella, Listeria) are present.

Using farm management practices to predict Campylobacter prevalence in pastured poultry farms

Contamination of poultry products by Campylobacter is often associated with farm management practices and processing plant practices. A longitudinal study was conducted on 11 pastured poultry farms in southeastern United States from 2014 to 2017. In this study, farm practices and processing variables were used as predictors for a random forest (RF) model to predict Campylobacter prevalence in pastured poultry farms and processing environments. Individual RF models were constructed for fecal, soil and whole carcass rinse after processing (WCR-P) samples. The performance of models was evaluated by the area under curve (AUC) from the receiver operating characteristics curve. The AUC values were 0.902, 0.894, and 0.864 for fecal, soil, and WCR-P models, respectively. Relative importance plots were generated to predict the most important variable in each RF model. Animal source of feces was identified as the most important variable in fecal model and the soy content of the brood feed was the most important variable for soil model. For WCR-P model, the average flock age showed the strongest impact on RF model. These RF models can help pastured poultry growers with food safety control strategies to reduce Campylobacter prevalence in pastured poultry farms.

Machine learning approach for predicting Fusarium culmorum and F. proliferatum growth and mycotoxin production in treatments with ethylene-vinyl alcohol copolymer films containing pure components of essential oils

Fusarium culmorum and F. proliferatum can grow and produce, respectively, zearalenone (ZEA) and fumonisins (FUM) in different points of the food chain. Application of antifungal chemicals to control these fungi and mycotoxins increases the risk of toxic residues in foods and feeds, and induces fungal resistances. In this study, a new and multidisciplinary approach based on the use of bioactive ethylene-vinyl alcohol copolymer (EVOH) films containing pure components of essential oils (EOCs) and machine learning (ML) methods is evaluated. Bioactive EVOH-EOC films were made incorporating cinnamaldehyde (CINHO), citral (CIT), isoeugenol (IEG) or linalool (LIN). Several ML methods (neural networks, random forests and extreme gradient boosted trees) and multiple linear regression (MLR) were applied and compared for modeling fungal growth and toxin production under different water activity (a_w) (0.96 and 0.99) and temperature (20 and 28 °C) regimes. The effective doses to reduce fungal growth rate (GR) by 50, 90 and 100% (ED₅₀, ED₉₀, and ED₁₀₀) of EOCs in EVOH films were in the ranges 200 to >3330, 450 to >3330, and 660 to >3330 μg/fungal culture (25 g of partly milled maize kernels in Petri dish), respectively, depending on the EOC, a_w and temperature. The type of EVOH-EOC film and EOC doses significantly affected GR in both species and ZEA and FUM production. Temperature also affected GR and a_w only affected GR and FUM production of F. proliferatum. EVOH-CIT was the most effective film against both species and ZEA and FUM production. Usually, when the EOC levels increased, GR and mycotoxin levels in the medium decreased although some treatments in combination with certain a_w and temperature values induced ZEA production. Random forest models predicted the GR of F. culmorum and F. proliferatum and ZEA and FUM production better than neural networks or extreme gradient boosted trees. The MLR mode provided the worst performance. This is the first approach on the ML potential in the study of the impact that bioactive EVOH films containing EOCs and environmental conditions have on F. culmorum and F. proliferatum growth and on ZEA and FUM production. The results suggest that these innovative packaging systems in combination with ML methods can be promising tools in the prediction and control of the risks associated with these toxigenic fungi and mycotoxins in food.

Predicting Salmonella prevalence associated with meteorological factors in pastured poultry farms in southeastern United States

Consumer demand has increased for pastured poultry products as the drive for sustainable farming practices and ethical treatments of livestock have become popular in the press. It is necessary to identify the important meteorological factors associated with the prevalence of Salmonella in the pastured poultry settings since the presence of Salmonella in the environment could lead to contamination of the final product. The objective of this study was to develop a model to describe the relationship between meteorological factors and the presence of Salmonella on the pastured poultry farms. The random forest method was used to develop a model where 83 meteorological factors were included as the predicting variables. The soil model identified humidity as the most important variable associated with Salmonella prevalence, while high wind gust speed and average temperature were identified as important meteorological variables in the feces model. The developed models were robust in predicting the prevalence of Salmonella in pastured poultry farms with the area under receiver operating characteristic (ROC) curve values of 0.884 and 0.872 for the soil model and feces model, respectively. The predictive models developed in this study can provide users with practical and effective tools to make informed decisions with scientific evidence regarding the meteorological parameters that are important to monitor for increased on-farm Salmonella prevalence.