Broadcasters, receivers, functional groups of metabolites, and the link to heart failure by revealing metabolomic network connectivity.

BACKGROUND AND OBJECTIVE: Blood-based small molecule metabolites offer easy accessibility and hold significant potential for insights into health processes, the impact of lifestyle, and genetic variation on disease, enabling precise risk prevention. In a prospective study with records of heart failure (HF) incidence, we present metabolite profiling data from individuals without HF at baseline. METHODS: We uncovered the interconnectivity of metabolites using data-driven and causal networks augmented with polygenic factors. Exploring the networks, we identified metabolite broadcasters, receivers, mediators, and subnetworks corresponding to functional classes of metabolites, and provided insights into the link between metabolomic architecture and regulation in health. We incorporated the network structure into the identification of metabolites associated with HF to control the effect of confounding metabolites. RESULTS: We identified metabolites associated with higher and lower risk of HF incidence, such as glycine, ureidopropionic and glycocholic acids, and LPC 18:2. These associations were not confounded by the other metabolites due to uncovering the connectivity among metabolites and adjusting each association for the confounding metabolites. Examples of our findings include the direct influence of asparagine on glycine, both of which were inversely associated with HF. These two metabolites were influenced by polygenic factors and only essential amino acids, which are not synthesized in the human body and are obtained directly from the diet. CONCLUSION: Metabolites may play a critical role in linking genetic background and lifestyle factors to HF incidence. Revealing the underlying connectivity of metabolites associated with HF strengthens the findings and facilitates studying complex conditions like HF.

The evaluation and optimization of the agricultural sustainable development based on a data-driven approach: A case from Northern Anhui.

The sustainable advancement of agriculture stands as the fundamental cornerstone of sustainable human progress. This study introduces a data-centric methodological framework founded upon the holistic delineation of measurement, feature assessment, and pathway enhancement for agricultural sustainability. Initially, the research articulates a comprehensive evaluative schema incorporating sub-dimensions encompassing agricultural production, agricultural economics, the agricultural resource environment, and rural society, grounded in sustainable development theory. Subsequently, it devises a methodological apparatus for assessing and enhancing sustainable development capabilities, employing entropy evaluation methods and exploratory spatial data analysis techniques. Employing North Anhui as a case study, the viability of this approach is substantiated. The empirical inquiry conducted within this article operationalizes comprehensive evaluation and explores pathways for optimizing agricultural sustainability, focusing on the period spanning 2011 to 2020 in Northern Anhui. The findings affirm the feasibility and efficacy of the data-driven approach. Recommendations derived from the empirical exploration of agricultural sustainability pathways at the local level offer valuable insights for governmental authorities and policymakers. This research endeavor could be extrapolated to other geographical locales worldwide, fostering innovative strides in the sustainable development of regional agriculture.

Enabling Reliable Visual Detection of Chronic Myocardial Infarction with Native T1 Cardiac MRI Using Data-Driven Native Contrast Mapping.

Purpose To investigate whether infarct-to-remote myocardial contrast can be optimized by replacing generic fitting algorithms used to obtain native T1 maps with a data-driven machine learning pixel-wise approach in chronic reperfused infarct in a canine model. Materials and Methods A controlled large animal model (24 canines, equal male and female animals) of chronic myocardial infarction with histologic evidence of heterogeneous infarct tissue composition was studied. Unsupervised clustering techniques using self-organizing maps and t-distributed stochastic neighbor embedding were used to analyze and visualize native T1-weighted pixel-intensity patterns. Deep neural network models were trained to map pixel-intensity patterns from native T1-weighted image series to corresponding pixels on late gadolinium enhancement (LGE) images, yielding visually enhanced noncontrast maps, a process referred to as data-driven native mapping (DNM). Pearson correlation coefficients and Bland-Altman analyses were used to compare findings from the DNM approach against standard T1 maps. Results Native T1-weighted images exhibited distinct pixel-intensity patterns between infarcted and remote territories. Granular pattern visualization revealed higher infarct-to-remote cluster separability with LGE labeling as compared with native T1 maps. Apparent contrast-to-noise ratio from DNM (mean, 15.01 ± 2.88 [SD]) was significantly different from native T1 maps (5.64 ± 1.58; P < .001) but similar to LGE contrast-to-noise ratio (15.51 ± 2.43; P = .40). Infarcted areas based on LGE were more strongly correlated with DNM compared with native T1 maps (R2 = 0.71 for native T1 maps vs LGE; R2 = 0.85 for DNM vs LGE; P < .001). Conclusion Native T1-weighted pixels carry information that can be extracted with the proposed DNM approach to maximize image contrast between infarct and remote territories for enhanced visualization of chronic infarct territories. Keywords: Chronic Myocardial Infarction, Cardiac MRI, Data-Driven Native Contrast Mapping Supplemental material is available for this article. © RSNA, 2024.

Neural network predictive controller based on the improved TPA-LSTM model for ultra-supercritical units.

To mitigate the impact of large-scale renewable energy power on the national grid in China, it is imperative to enhance the flexible peaking capability of coal-fired thermal power units. The coordinated control system, central to the load control of coal-fired units, faces challenges such as multivariable coupling, sluggish response, and uncertain coal quality parameters. This paper introduces a neural network predictive controller based on the improved TPA-LSTM model, aimed at addressing these issues. Initially, a data-driven control model is established to break through the limitations of traditional linear predictive control and effectively handle disturbance uncertainties. Then, a multivariable coordinated control strategy based on the neural network controller is designed, achieving effective decoupling of multiple parameters and ensuring high adaptability across all load conditions. Additionally, by integrating an automatic model updating mechanism, the system can recalibrate in real-time when model mismatches occur due to equipment aging, maintenance, or changes in coal quality, thereby enhancing overall control performance. Simulation results demonstrate that this strategy has excellent control effectiveness, meeting the flexible peaking demands of 1000 MW ultra-supercritical units. The calibration feature of the data-driven model significantly improves control performance following model mismatches.

Learn to optimize-a brief overview.

Most optimization problems of practical significance are typically solved by highly configurable parameterized algorithms. To achieve the best performance on a problem instance, a trial-and-error configuration process is required, which is very costly and even prohibitive for problems that are already computationally intensive, e.g. optimization problems associated with machine learning tasks. In the past decades, many studies have been conducted to accelerate the tedious configuration process by learning from a set of training instances. This article refers to these studies as learn to optimize and reviews the progress achieved.

What the Eyes Don't See: An Objective Assessment of Players' Contribution to Team Success in Men's Rugby League.

Purpose: This study addresses the lack of objective player-based metrics in men's rugby league by introducing a comprehensive set of novel performance metrics designed to quantify a player's overall contribution to team success. Methods: Player match performance data were captured by Stats Perform for every National Rugby League season from 2018 until 2022; a total of five seasons. The dataset was divided into offensive and defensive variables and further split according to player position. Five machine learning algorithms (Principal Component Regression, Lasso Regression, Random Forest, Regression Tree, and Extreme Gradient Boost) were considered in the analysis, which ultimately generated Wins Created and Losses Created for offensive and defensive performance, respectively. These two metrics were combined to create a final metric of Net Wins Added. The validity of these player performance metrics against traditional objective and subjective measures of performance in rugby league were evaluated. Results: The metrics correctly predicted the winner of 80.9% of matches, as well as predicting the number of team wins per season with an RMSE of 1.9. The metrics displayed moderate agreement (Gwet AC1 = 0.505) when predicting team of the year award recipients. When predicting State of Origin selection, the metrics displayed moderate agreement for New South Wales (0.450) and substantial agreement for Queensland (0.652). Conclusion: The development and validation of these objective player performance metrics represent significant potential to enhance talent evaluation and player recruitment.

Anomaly Detection and Remaining Useful Life Estimation for the Health and Usage Monitoring Systems 2023 Data Challenge.

Gear fault detection and remaining useful life estimation are important tasks for monitoring the health of rotating machinery. In this study, a new benchmark for endurance gear vibration signals is presented and made publicly available. The new dataset was used in the HUMS 2023 conference data challenge to test anomaly detection algorithms. A survey of the suggested techniques is provided, demonstrating that traditional signal processing techniques interestingly outperform deep learning algorithms in this case. Of the 11 participating groups, only those that used traditional approaches achieved good results on most of the channels. Additionally, we introduce a signal processing anomaly detection algorithm and meticulously compare it to a standard deep learning anomaly detection algorithm using data from the HUMS 2023 challenge and simulated signals. The signal processing algorithm surpasses the deep learning algorithm on all tested channels and also on simulated data where there is an abundance of training data. Finally, we present a new digital twin that enables the estimation of the remaining useful life of the tested gear from the HUMS 2023 challenge.

Using Support Vector Machines to Classify Road Surface Conditions to Promote Safe Driving.

Accurate detection of road surface conditions in adverse winter weather is essential for traffic safety. To promote safe driving and efficient road management, this study presents an accurate and generalizable data-driven learning model for the estimation of road surface conditions. The machine model was a support vector machine (SVM), which has been successfully applied in diverse fields, and kernel functions (linear, Gaussian, second-order polynomial) with a soft margin classification technique were also adopted. Two learner designs (one-vs-one, one-vs-all) extended their application to multi-class classification. In addition to this non-probabilistic classifier, this study calculated the posterior probability of belonging to each group by applying the sigmoid function to the classification scores obtained by the trained SVM. The results indicate that the classification errors of all the classifiers, excluding the one-vs-all linear learners, were below 3%, thereby accurately classifying road surface conditions, and that the generalization performance of all the one-vs-one learners was within an error rate of 4%. The results also showed that the posterior probabilities can analyze certain atmospheric and road surface conditions that correspond to a high probability of hazardous road surface conditions. Therefore, this study demonstrates the potential of data-driven learning models in classifying road surface conditions accurately.

Validation and Evaluation of a Vendor-Provided Head Motion Correction Algorithm on the uMI Panorama PET/CT System.

Brain PET imaging often faces challenges from head motion (HM), which can introduce artifacts and reduce image resolution, crucial in clinical settings for accurate treatment planning, diagnosis, and monitoring. United Imaging Healthcare has developed NeuroFocus, an HM correction (HMC) algorithm for the uMI Panorama PET/CT system, using a data-driven, statistics-based approach. The HMC algorithm automatically detects HM using a centroid-of-distribution technique, requiring no parameter adjustments. This study aimed to validate NeuroFocus and assess the prevalence of HM in clinical short-duration 18F-FDG scans. Methods: The study involved 317 patients undergoing brain PET scans, divided into 2 groups: 15 for HMC validation and 302 for evaluation. Validation involved patients undergoing 2 consecutive 3-min single-bed-position brain 18F-FDG scans-one with instructions to remain still and another with instructions to move substantially. The evaluation examined 302 clinical single-bed-position brain scans for patients with various neurologic diagnoses. Motion was categorized as small or large on the basis of a 5% SUV change in the frontal lobe after HMC. Percentage differences in SUVmean were reported across 11 brain regions. Results: The validation group displayed a large negative difference (-10.1%), with variation of 5.2% between no-HM and HM scans. After HMC, this difference decreased dramatically (-0.8%), with less variation (3.2%), indicating effective HMC application. In the evaluation group, 38 of 302 patients experienced large HM, showing a 10.9% ± 8.9% SUV increase after HMC, whereas most exhibited minimal uptake changes (0.1% ± 1.3%). The HMC algorithm not only enhanced the image resolution and contrast but also aided in disease identification and reduced the need for repeat scans, potentially optimizing clinical workflows. Conclusion: The study confirmed the effectiveness of NeuroFocus in managing HM in short clinical 18F-FDG studies on the uMI Panorama PET/CT system. It found that approximately 12% of scans required HMC, establishing HMC as a reliable tool for clinical brain 18F-FDG studies.

A perspective-driven and technical evaluation of machine learning in bioreactor scale-up: A case-study for potential model developments.

Bioreactor scale-up and scale-down have always been a topical issue for the biopharmaceutical industry and despite considerable effort, the identification of a fail-safe strategy for bioprocess development across scales remains a challenge. With the ubiquitous growth of digital transformation technologies, new scaling methods based on computer models may enable more effective scaling. This study aimed to evaluate the potential application of machine learning (ML) algorithms for bioreactor scale-up, with a specific focus on the prediction of scaling parameters. Factors critical to the development of such models were identified and data for bioreactor scale-up studies involving CHO cell-generated mAb products collated from the literature and public sources for the development of unsupervised and supervised ML models. Comparison of bioreactor performance across scales identified similarities between the different processes and primary differences between small- and large-scale bioreactors. A series of three case studies were developed to assess the relationship between cell growth and scale-sensitive bioreactor features. An embedding layer improved the capability of artificial neural network models to predict cell growth at a large-scale, as this approach captured similarities between the processes. Further models constructed to predict scaling parameters demonstrated how ML models may be applied to assist the scaling process. The development of data sets that include more characterization data with greater variability under different gassing and agitation regimes will also assist the future development of ML tools for bioreactor scaling.

Application of data-driven blended online-offline teaching in medicinal chemistry for pharmacy students: a randomized comparison.

BACKGROUND: The purpose of this study was to evaluate the effectiveness and efficiency of implementing a data-driven blended online-offline (DDBOO) teaching approach in the medicinal chemistry course. METHODS: A total of 118 third-year students majoring in pharmacy were enrolled from September 2021 to January 2022. The participants were randomly assigned to either the DDBOO teaching group or the traditional lecture-based learning (LBL) group for medicinal chemistry. Pre- and post-class quizzes were administered, along with an anonymous questionnaire distributed to both groups to assess students' perceptions and experiences. RESULTS: There was no significant difference in the pre-class quiz scores between the DDBOO and LBL groups (T=-0.637, P = 0.822). However, after class, the mean quiz score of the DDBOO group was significantly higher than that of the LBL group (T = 3.742, P < 0.001). Furthermore, the scores for learning interest, learning motivation, self-learning skill, mastery of basic knowledge, teamwork skills, problem-solving ability, innovation ability, and satisfaction, as measured by the questionnaire, were significantly higher in the DDBOO group than in the traditional group (all P < 0.05). CONCLUSION: The DDBOO teaching method effectively enhances students' academic performance and satisfaction. Further research and promotion of this approach are warranted.

Practical one-shot data-driven design of fractional-order PID controller: Fictitious reference signal approach.

This study proposes a one-shot data-driven tuning method for a fractional-order proportional-integral-derivative (FOPID) controller. The proposed method tunes the FOPID controller in the model-reference control formulation. A loss function is defined to evaluate the match between a given reference model and the closed-loop response while explicitly considering the closed-loop stability. A loss function value is based on the fictitious reference signal computed using the input/output data. Model matching is achieved via loss function minimization. The proposed method is simple and practical: it needs only one-shot input/output data of a plant (no plant model required), considers the bounded-input bounded-output stability of the closed-loop system from a bounded reference input to a bounded output, and automatically determines the appropriate parameter value via optimization. Numerical simulations show that the proposed approach facilitates good control performance, and destabilization can be avoided even if perfect model matching is unachievable.

Data-driven gated positron emission tomography/computed tomography for radiotherapy.

Purpose: Software-based data-driven gated (DDG) positron emission tomography/computed tomography (PET/CT) has replaced hardware-based 4D PET/CT. The purpose of this article was to review DDG PET/CT, which could improve the accuracy of treatment response assessment, tumor motion evaluation, and target tumor contouring with whole-body (WB) PET/CT for radiotherapy (RT). Material and methods: This review covered the topics of 4D PET/CT with hardware gating, advancements in PET instrumentation, DDG PET, DDG CT, and DDG PET/CT based on a systematic literature review. It included a discussion of the large axial field-of-view (AFOV) PET detector and a review of the clinical results of DDG PET and DDG PET/CT. Results: DDG PET matched or outperformed 4D PET with hardware gating. DDG CT was more compatible with DDG PET than 4D CT, which required hardware gating. DDG CT could replace 4D CT for RT. DDG PET and DDG CT for DDG PET/CT can be incorporated in a WB PET/CT of less than 15 min scan time on a PET/CT scanner of at least 25 cm AFOV PET detector. Conclusions: DDG PET/CT could correct the misregistration and tumor motion artifacts in a WB PET/CT and provide the quantitative PET and tumor motion information of a registered PET/CT for RT.

Decoding children dental health risks: a machine learning approach to identifying key influencing factors.

Introduction and objectives: This study investigates key factors influencing dental caries risk in children aged 7 and under using machine learning techniques. By addressing dental caries' prevalence, it aims to enhance early identification and preventative strategies for high-risk individuals. Methods: Data from clinical examinations of 356 children were analyzed using Logistic Regression, Decision Trees, and Random Forests models. These models assessed the influence of dietary habits, fluoride exposure, and socio-economic status on caries risk, emphasizing accuracy, precision, recall, F1 score, and AUC metrics. Results: Poor oral hygiene, high sugary diet, and low fluoride exposure were identified as significant caries risk factors. The Random Forest model demonstrated superior performance, illustrating the potential of machine learning in complex health data analysis. Our SHAP analysis identified poor oral hygiene, high sugary diet, and low fluoride exposure as significant caries risk factors. Conclusion: Machine learning effectively identifies and quantifies dental caries risk factors in children. This approach supports targeted interventions and preventive measures, improving pediatric dental health outcomes. Clinical significance: By leveraging machine learning to pinpoint crucial caries risk factors, this research lays the groundwork for data-driven preventive strategies, potentially reducing caries prevalence and promoting better dental health in children.

Data driven surrogate signal extraction for dynamic PET using selective PCA: time windows versus the combination of components.

Respiratory motion correction is beneficial in PET, as it can reduce artefacts caused by motion and improve quantitative accuracy. Methods of motion correction are commonly based on a respiratory trace obtained through an external device (like the Real Time Position Management System) or a data driven method, such as those based on dimensionality reduction techniques (for instance PCA). PCA itself being a linear transformation to the axis of greatest variation. Data driven methods have the advantage of being non-invasive, and can be performed post-acquisition. However, their main downside being that they are adversely affected by the tracer kinetics of the dynamic PET acquisition. Therefore, they are mostly limited to static PET acquisitions. This work seeks to extend on existing PCA-based data-driven motion correction methods, to allow for their applicability to dynamic PET imaging. The methods explored in this work include; a moving window approach (similar to the Kinetic Respiratory Gating method from Schleyer et al.), extrapolation of the principal component from later time points to earlier time points, and a method to score, select, and combine multiple respiratory components. The resulting respiratory traces were evaluated on 22 data sets from a dynamic 18FFDG study on patients with Idiopathic Pulmonary Fibrosis. This was achieved by calculating their correlation with a surrogate signal acquired using a Real Time Position Management System. The results indicate that all methods produce better surrogate signals than when applying conventional PCA to dynamic data (for instance, a higher correlation with a gold standard respiratory trace). Extrapolating a late time point principal component produced more promising results than using a moving window. Scoring, selecting, and combining components held benefits over all other methods. This work allows for the extraction of a surrogate signal from dynamic PET data earlier in the acquisition and with a greater accuracy than previous work. This potentially allows for numerous other methods (for instance, respiratory motion correction) to be applied to this data (when they otherwise could not be previously used).

Community-Engaged Data Science (CEDS): A Case Study of Working with Communities to Use Data to Inform Change.

Data-informed decision making is a critical goal for many community-based public health research initiatives. However, community partners often encounter challenges when interacting with data. The Community-Engaged Data Science (CEDS) model offers a goal-oriented, iterative guide for communities to collaborate with research data scientists through data ambassadors. This study presents a case study of CEDS applied to research on the opioid epidemic in 18 counties in Ohio as part of the HEALing Communities Study (HCS). Data ambassadors provided a pivotal role in empowering community coalitions to translate data into action using key steps of CEDS which included: data landscapes identifying available data in the community; data action plans from logic models based on community data needs and gaps of data; data collection/sharing agreements; and data systems including portals and dashboards. Throughout the CEDS process, data ambassadors emphasized sustainable data workflows, supporting continued data engagement beyond the HCS. The implementation of CEDS in Ohio underscored the importance of relationship building, timing of implementation, understanding communities' data preferences, and flexibility when working with communities. Researchers should consider implementing CEDS and integrating a data ambassador in community-based research to enhance community data engagement and drive data-informed interventions to improve public health outcomes.

Leveraging advances in data-driven deep learning methods for hybrid epidemic modeling.

Mathematical modeling of epidemic dynamics is crucial to understand its underlying mechanisms, quantify important parameters, and make predictions that facilitate more informed decision-making. There are three major types of models: mechanistic models including the SEIR-type paradigm, alternative data-driven (DD) approaches, and hybrid models that combine mechanistic models with DD approaches. In this paper, we summarize our work in the COVID-19 Scenario Modeling Hub (SMH) for more than 12 rounds since early 2021 for informed decision support. We emphasize the importance of deep learning techniques for epidemic modeling via a flexible DD framework that substantially complements the mechanistic paradigm to evaluate various future epidemic scenarios. We start with a traditional curve-fitting approach to model cumulative COVID-19 based on the underlying SEIR-type mechanisms. Hospitalizations and deaths are modeled as binomial processes of cases and hospitalization, respectively. We further formulate two types of deep learning models based on multivariate long short term memory (LSTM) to address the challenges of more traditional DD models. The first LSTM is structurally similar to the curve fitting approach and assumes that hospitalizations and deaths are binomial processes of cases. Instead of using a predefined exponential curve, LSTM relies on the underlying data to identify the most appropriate functions, and is capable of capturing both long-term and short-term epidemic behaviors. We then relax the assumption of dependent inputs among cases, hospitalizations, and death. Another type of LSTM that handles all input time series as parallel signals, the independent multivariate LSTM, is developed. Independent multivariate LSTM can incorporate a wide range of data sources beyond traditional case-based epidemiological surveillance. The DD framework unleashes its potential in big data era with previously neglected heterogeneous surveillance data sources, such as syndromic, environment, genomic, serologic, infoveillance, and mobility data. DD approaches, especially LSTM, complement and integrate with the mechanistic modeling paradigm, provide a feasible alternative approach to model today's complex socio-epidemiological systems, and further leverage our ability to explore different scenarios for more informed decision-making during health emergencies.

Data-science-guided calibration curve prediction of an MLCT-based ee determination assay for chiral amines.

Circular dichroism (CD) based enantiomeric excess (ee) determination assays are optical alternatives to chromatographic ee determination in high-throughput screening (HTS) applications. However, the implementation of these assays requires calibration experiments using enantioenriched materials. We present a data-driven approach that circumvents the need for chiral resolution and calibration experiments for an octahedral Fe(II) complex (1) used for the ee determination of α-chiral primary amines. By computationally parameterizing the imine ligands formed in the assay conditions, a model of the circular dichroism (CD) response of the Fe(II) assembly was developed. Using this model, calibration curves were generated for four analytes and compared to experimentally generated curves. In a single-blind ee determination study, the ee values of unknown samples were determined within 9% mean absolute error, which rivals the error using experimentally generated calibration curves.

Artificial intelligence classifies primary progressive aphasia from connected speech.

Neurodegenerative dementia syndromes, such as Primary Progressive Aphasias (PPA), have traditionally been diagnosed based in part on verbal and nonverbal cognitive profiles. Debate continues about whether PPA is best divided into three variants and also regarding the most distinctive linguistic features for classifying PPA variants. In this cross-sectional study, we first harnessed the capabilities of artificial intelligence (AI) and Natural Language Processing (NLP) to perform unsupervised classification of short, connected speech samples from 78 PPA patients. We then used NLP to identify linguistic features that best dissociate the three PPA variants. Large Language Models (LLMs) discerned three distinct PPA clusters, with 88.5% agreement with independent clinical diagnoses. Patterns of cortical atrophy of three data-driven clusters corresponded to the localization in the clinical diagnostic criteria. In the subsequent supervised classification, seventeen distinctive features emerged, including the observation that separating verbs into high and low-frequency types significantly improves classification accuracy. Using these linguistic features derived from the analysis of short, connected speech samples, we developed a classifier that achieved 97.9% accuracy in classifying the four groups (three PPA variants and healthy controls). The data-driven section of this study showcases the ability of LLMs to find natural partitioning in the speech of patients with PPA consistent with conventional variants. In addition, the work identifies a robust set of language features indicative of each PPA variant, emphasizing the significance of dividing verbs into high and low-frequency categories. Beyond improving diagnostic accuracy, these findings enhance our understanding of the neurobiology of language processing.

A soft-sensor approach for predicting an indicator virus removal efficiency of a pilot-scale anaerobic membrane bioreactor (AnMBR).

The anaerobic membrane bioreactor (AnMBR) is a promising technology for not only water reclamation but also virus removal; however, the virus removal efficiency of AnMBR has not been fully investigated. Additionally, the removal efficiency estimation requires datasets of virus concentration in influent and effluent, but its monitoring is not easy to perform for practical operation because the virus quantification process is generally time-consuming and requires specialized equipment and trained personnel. Therefore, in this study, we aimed to identify the key, monitorable variables in AnMBR and establish the data-driven models using the selected variables to predict virus removal efficiency. We monitored operational and environmental conditions of AnMBR in Sendai, Japan and measured virus concentration once a week for six months. Spearman's rank correlation analysis revealed that the pH values of influent and mixed liquor suspended solids (MLSS) were strongly correlated with the log reduction value of pepper mild mottle virus, indicating that electrostatic interactions played a dominant role in AnMBR virus removal. Among the candidate models, the random forest model using selected variables including influent and MLSS pH outperformed the others. This study has demonstrated the potential of AnMBR as a viable option for municipal wastewater reclamation with high microbial safety.