Suchergebnisse

AbstractInappropriate management of health and safety (H&S) risk in power infrastructure projects can result in occupational accidents and equipment damage. Accidents at work have detrimental effects on workers, company, and the general public. Despite the availability of H&S incident data, utilizing them to mitigate accident occurrence effectively is challenging due to inherent limitations of existing data logging methods. In this study, we used a text‐mining approach for retrieving meaningful terms from data and develop six deep learning (DL) models for H&S risks management in power infrastructure. The DL models include DNNclassify (risk or no risk), DNNreg1 (loss time), DNNreg2 (body injury), DNNreg3 (plant and fleet), DNNreg4 (equipment), and DNNreg5 (environment). An H&S risk database obtained from a leading UK power infrastructure construction company was used in developing the models using the H2O framework of the R language. Performances of DL models were assessed and benchmarked with existing models using test data and appropriate performance metrics. The overall accuracy of the classification model was 0.93. The average R2 value for the five regression models was 0.92, with mean absolute error between 0.91 and 0.94. The presented results, in addition to the developed user‐interface module, will help practitioners obtain a better understanding of H&S challenges, minimize project costs (such as third‐party insurance and equipment repairs), and offer effective strategies to mitigate H&S risk.

Abstract. This study investigates the issues related to underestimation of the earthquake source parameters in the context of tsunami early warning and tsunami risk assessment. The magnitude of a very large event may be underestimated significantly during the early stage of the disaster, resulting in the issuance of incorrect tsunami warnings. Tsunamigenic events in the Tohoku region of Japan, where the 2011 tsunami occurred, are focused on as a case study to illustrate the significance of the problems. The effects of biases in the estimated earthquake magnitude on tsunami loss are investigated using a rigorous probabilistic tsunami loss calculation tool that can be applied to a range of earthquake magnitudes by accounting for uncertainties of earthquake source parameters (e.g., geometry, mean slip, and spatial slip distribution). The quantitative tsunami loss results provide valuable insights regarding the importance of deriving accurate seismic information as well as the potential biases of the anticipated tsunami consequences. Finally, the usefulness of rigorous tsunami risk assessment is discussed in defining critical hazard scenarios based on the potential consequences due to tsunami disasters.

ABSTRACTAutonomous underwater gliders (AUGs) are effective platforms for oceanic research and environmental monitoring. However, complex underwater environments with uncertainties could pose the risk of vehicle loss during their missions. It is therefore essential to conduct risk prediction to assist decision making for safer operations. The main limitation of current studies for AUGs is the lack of a tailored method for risk analysis considering both dynamic environments and potential functional failures of the vehicle. Hence, this study proposed a copula‐based approach for evaluating the risk of AUG loss in dynamic underwater environments. The developed copula Bayesian network (CBN) integrated copula functions into a traditional Bayesian belief network (BBN), aiming to handle nonlinear dependencies among environmental variables and inherent technical failures. Specifically, potential risk factors with causal effects were captured using the BBN. A Gaussian copula was then employed to measure correlated dependencies among identified risk factors. Furthermore, the dependence analysis and CBN inference were performed to assess the risk level of vehicle loss given various environmental observations. The effectiveness of the proposed method was demonstrated in a case study, which considered deploying a Slocum G1 Glider in a real water region. Risk mitigation measures were provided based on key findings. This study potentially contributes a tailored tool of risk prediction for AUGs in dynamic environments, which can enhance the safety performance of AUGs and assist in risk mitigation for decision makers.

AbstractClinical decision support systems (CDSS) that make use of algorithms based on intelligent systems, such as machine learning or deep learning, they suffer from the fact that often the methods used are hard to interpret and difficult to understand on how some decisions are made; the opacity of some methods, sometimes voluntary due to problems such as data privacy or the techniques used to protect intellectual property, makes these systems very complicated. Besides this series of problems, the results obtained also suffer from the poor possibility of being interpreted; in the clinical context therefore it is required that the methods used are as accurate as possible, transparent techniques and explainable results. In this work the problem of the development of cervical cancer is treated, a disease that mainly affects the female population. In order to introduce advanced machine learning techniques in a clinical decision support system that can be transparent and explainable, a robust, accurate ensemble method is presented, in terms of error and sensitivity linked to the classification of possible development of the aforementioned pathology and advanced techniques are also presented of explainability and interpretability (Explanaible Machine Learning) applied to the context of CDSS such as Lime and Shapley. The results obtained, as well as being interesting, are understandable and can be implemented in the treatment of this type of problem.

Spatial risk prediction models describe the features of small spatial units which support or attract child maltreatment behaviors. Due to shared risk factors, community problems such as infant mortality, poor educational readiness, injury-related deaths, and a host of physical and mental health problems associated with toxic stress co-occur in these small areas. Based on administrative data, this spatial intelligence brings cross-sector stakeholders together to collaboratively plan for optimization of critical supports, capacity development for vital services, improvement of professional response, development of supportive infrastructure, and creation of healthier social norms. We present the policy and resource optimization strategies being developed in locations implementing the Predict-Align-Prevent program for child maltreatment prevention.

Background: Alcohol-related road-traffic injury is the leading cause of premature death in middle- and lower-income countries, including Thailand. Applying machine-learning algorithms can improve the effectiveness of driver-impairment screening strategies by legal limits. Methods: Using 4794 RTI drivers from secondary cross-sectional data from the Thai Governmental Road Safety Evaluation project in 2002–2004, the machine-learning models (Gradient Boosting Classifier: GBC, Multi-Layers Perceptrons: MLP, Random Forest: RF, K-Nearest Neighbor: KNN) and a parsimonious logistic regression (Logit) were developed for predicting the mortality risk from road-traffic injury in drunk drivers. The predictors included alcohol concentration level in blood or breath, driver characteristics and environmental factors. Results: Of 4974 drivers in the derived dataset, 4365 (92%) were surviving drivers and 429 (8%) were dead drivers. The class imbalance was rebalanced by the Synthetic Minority Oversampling Technique (SMOTE) into a 1:1 ratio. All models obtained good-to-excellent discrimination performance. The AUC of GBC, RF, KNN, MLP, and Logit models were 0.95 (95% CI 0.90 to 1.00), 0.92 (95% CI 0.87 to 0.97), 0.86 (95% CI 0.83 to 0.89), 0.83 (95% CI 0.78 to 0.88), and 0.81 (95% CI 0.75 to 0.87), respectively. MLP and GBC also had a good model calibration, visualized by the calibration plot. Conclusions: Our machine-learning models can predict road-traffic mortality risk with good model discrimination and calibration. External validation using current data is recommended for future implementation.

Background: Alcohol-related road-traffic injury is the leading cause of premature death in middle- and lower-income countries, including Thailand. Applying machine-learning algorithms can improve the effectiveness of driver-impairment screening strategies by legal limits. Methods: Using 4794 RTI drivers from secondary cross-sectional data from the Thai Governmental Road Safety Evaluation project in 2002–2004, the machine-learning models (Gradient Boosting Classifier: GBC, Multi-Layers Perceptrons: MLP, Random Forest: RF, K-Nearest Neighbor: KNN) and a parsimonious logistic regression (Logit) were developed for predicting the mortality risk from road-traffic injury in drunk drivers. The predictors included alcohol concentration level in blood or breath, driver characteristics and environmental factors. Results: Of 4974 drivers in the derived dataset, 4365 (92%) were surviving drivers and 429 (8%) were dead drivers. The class imbalance was rebalanced by the Synthetic Minority Oversampling Technique (SMOTE) into a 1:1 ratio. All models obtained good-to-excellent discrimination performance. The AUC of GBC, RF, KNN, MLP, and Logit models were 0.95 (95% CI 0.90 to 1.00), 0.92 (95% CI 0.87 to 0.97), 0.86 (95% CI 0.83 to 0.89), 0.83 (95% CI 0.78 to 0.88), and 0.81 (95% CI 0.75 to 0.87), respectively. MLP and GBC also had a good model calibration, visualized by the calibration plot. Conclusions: Our machine-learning models can predict road-traffic mortality risk with good model discrimination and calibration. External validation using current data is recommended for future implementation.

Background: Nutrition and lifestyle have been long established as risk factors for colorectal cancer (CRC). Modifiable lifestyle behaviours bear potential to minimize long-term CRC risk; however, translation of lifestyle information into individualized CRC risk assessment has not been implemented. Lifestyle-based risk models may aid the identification of high-risk individuals, guide referral to screening and motivate behaviour change. We therefore developed and validated a lifestyle-based CRC risk prediction algorithm in an asymptomatic European population. Methods: The model was based on data from 255,482 participants in the European Prospective Investigation into Cancer and Nutrition (EPIC) study aged 19 to 70 years who were free of cancer at study baseline (1992–2000) and were followed up to 31 September 2010. The model was validated in a sample comprising 74,403 participants selected among five EPIC centres. Over a median follow-up time of 15 years, there were 3645 and 981 colorectal cancer cases in the derivation and validation samples, respectively. Variable selection algorithms in Cox proportional hazard regression and random survival forest (RSF) were used to identify the best predictors among plausible predictor variables. Measures of discrimination and calibration were calculated in derivation and validation samples. To facilitate model communication, a nomogram and a web-based application were developed. Results: The final selection model included age, waist circumference, height, smoking, alcohol consumption, physical activity, vegetables, dairy products, processed meat, and sugar and confectionary. The risk score demonstrated good discrimination overall and in sex-specific models. Harrell's C-index was 0.710 in the derivation cohort and 0.714 in the validation cohort. The model was well calibrated and showed strong agreement between predicted and observed risk. Random survival forest analysis suggested high model robustness. Beyond age, lifestyle data led to improved model performance overall (continuous net reclassification improvement = 0.307 (95% CI 0.264–0.352)), and especially for young individuals below 45 years (continuous net reclassification improvement = 0.364 (95% CI 0.084–0.575)). Conclusions: LiFeCRC score based on age and lifestyle data accurately identifies individuals at risk for incident colorectal cancer in European populations and could contribute to improved prevention through motivating lifestyle change at an individual level. ; German Research Foundation (DFG) AL 1784/3-1 ; European Commission European Commission Joint Research Centre ; International Agency for Research on Cancer ; Danish Cancer Society ; Ligue Contre le Cancer (France) ; Institut Gustave Roussy (France) ; Mutuelle Generale de l'Education Nationale (France) ; Institut National de la Sante et de la Recherche Medicale (Inserm) ; Deutsche Krebshilfe, Deutsches Krebsforschungszentrum (DKFZ) (Germany) ; Federal Ministry of Education & Research (BMBF) ; Hellenic Health Foundation (Greece) ; Associazione Italiana per la Ricerca sul Cancro (AIRC) ; Consiglio Nazionale delle Ricerche (CNR) ; Netherlands Government ; World Cancer Research Fund International (WCRF) ; Instituto de Salud Carlos III PI13/00061 PI13/01162 ; Junta de Andalucia ; Regional Government of Asturias (Spain) ; Regional Government of Basque Country (Spain) ; Regional Government of Murcia (Spain) 6236 ; Regional Government of Navarra (Spain) ; Regional Government of Catalonia (Catalan Institute of Oncology -ICO-IDIBELL) (Spain) ; Swedish Cancer Society ; Swedish Research Council ; County Council of Skane (Sweden) ; County Council of Vasterbotten (Sweden) ; Cancer Research UK C864/A14136 C8221/A19170 ; UK Research & Innovation (UKRI) Medical Research Council UK (MRC) MR/N003284/1 MC-UU_12015/1 MR/M012190/1 ; Projekt DEAL

In mid April 2020, with more than 2.5 billion people in the world following social distancing measures due to COVID-19, governments are considering relaxing lock-down. We combined individual clinical risk predictions with epidemic modelling to examine simulations of isolation and exit policies.Methods: We developed a method to include personalised risk predictions in epidemic models based on data science principles. We extended a standard susceptible-exposed-infected-removed (SEIR) model to account for predictions of severity, defined by the risk of an individual needing intensive care in case of infection. We studied example isolation policies using simulations with the risk-extended epidemic model, using COVID-19 data and estimates in France as of mid April 2020 (4 000 patients in ICU, around 7 250 total ICU beds occupied at the peak of the outbreak, 0.5 percent of patients requiring ICU upon infection). We considered scenarios varying in the discrimination performance of a risk prediction model, in the degree of social distancing, and in the severity rate upon infection. Confidence intervals were obtained using an Approximate Bayesian Computation approach. The framework may be used with other epidemic models, with other risk predictions, and for other epidemic outbreaks.

In mid April 2020, with more than 2.5 billion people in the world following social distancing measures due to COVID-19, governments are considering relaxing lock-down. We combined individual clinical risk predictions with epidemic modelling to examine simulations of isolation and exit policies.Methods: We developed a method to include personalised risk predictions in epidemic models based on data science principles. We extended a standard susceptible-exposed-infected-removed (SEIR) model to account for predictions of severity, defined by the risk of an individual needing intensive care in case of infection. We studied example isolation policies using simulations with the risk-extended epidemic model, using COVID-19 data and estimates in France as of mid April 2020 (4 000 patients in ICU, around 7 250 total ICU beds occupied at the peak of the outbreak, 0.5 percent of patients requiring ICU upon infection). We considered scenarios varying in the discrimination performance of a risk prediction model, in the degree of social distancing, and in the severity rate upon infection. Confidence intervals were obtained using an Approximate Bayesian Computation approach. The framework may be used with other epidemic models, with other risk predictions, and for other epidemic outbreaks.

In mid April 2020, with more than 2.5 billion people in the world following social distancing measures due to COVID-19, governments are considering relaxing lock-down. We combined individual clinical risk predictions with epidemic modelling to examine simulations of isolation and exit policies.Methods: We developed a method to include personalised risk predictions in epidemic models based on data science principles. We extended a standard susceptible-exposed-infected-removed (SEIR) model to account for predictions of severity, defined by the risk of an individual needing intensive care in case of infection. We studied example isolation policies using simulations with the risk-extended epidemic model, using COVID-19 data and estimates in France as of mid April 2020 (4 000 patients in ICU, around 7 250 total ICU beds occupied at the peak of the outbreak, 0.5 percent of patients requiring ICU upon infection). We considered scenarios varying in the discrimination performance of a risk prediction model, in the degree of social distancing, and in the severity rate upon infection. Confidence intervals were obtained using an Approximate Bayesian Computation approach. The framework may be used with other epidemic models, with other risk predictions, and for other epidemic outbreaks.