Search

Objective: To summarize the current research progress of machine learning and venous thromboembolism. Methods: The literature on risk factors, diagnosis, prevention and prognosis of machine learning and venous thromboembolism in recent years was reviewed. Results: Machine learning is the future of biomedical research, personalized medicine, and computer-aided diagnosis, and will significantly promote the development of biomedical research and healthcare. However, many medical professionals are not familiar with it. In this review, we will introduce several commonly used machine learning algorithms in medicine, discuss the application of machine learning in venous thromboembolism, and reveal the challenges and opportunities of machine learning in medicine. Conclusion: The incidence of venous thromboembolism is high, the diagnostic measures are diverse, and it is necessary to classify and treat machine learning, and machine learning as a research tool, it is more necessary to strengthen the special research of venous thromboembolism and machine learning. [ABSTRACT FROM AUTHOR]

Abstract To solve the problems of high sampling requirements and low predictive accuracy resulting from the complexity, uniqueness, and randomness of predicting the risk of the CO2 and N2 injection to enhance coal seam gas drainage (CO2/N2–ECGD) technology. The principal component analysis (PCA) method to reduce the dimensionality of the factor data that contribute to the effect risk of the technology was adopted. And the particle swarm optimization (PSO) method was implemented to search for optimal hyperparameters in support vector machine (SVM) by particle search, as a solution to the traditional SVM hyperparameters optimization problem. A novel risk prediction model using machine learning algorithms for gas injection displacement technology was constructed. The prediction results were tested and compared with those of backpropagation (BP), Random Forest (RF), and Decision Tree (DT) models using data from 29 gas injection displacement field projects in China. The results demonstrated that the SVM model had greater accuracy in prediction than the other three models. Additionally, after PSO optimization and dimensionality reduction, the PCA–PSO–SVM model reached 100% prediction accuracy, while requiring less modeling and operation time. The study provided a reliable and reasonable model for predicting technical effects, along with a theoretical basis for risk management and prevention. First, the technology's influencing indicators were analyzed by examining its mechanisms. Second, we utilized the PCA method to reduce the dimensionality of the factor data that contribute to the risk of the technology's effects. Third, we implemented the PSO method to search for optimal hyperparameters in the SVM through particle search, as a solution to the traditional SVM hyperparameters optimization problem. Finally, the prediction results were tested and compared with those of BP, RF, and DT models using data from 29 gas injection displacement field projects in China. The SVM model was found to have greater accuracy in prediction than the other three models. After PSO optimization and dimensionality reduction, the PCA–PSO–SVM model achieved 100% prediction accuracy while requiring less modeling and operation time. The study presents a valid and reasonable model for predicting technical effects and a theoretical basis for risk management and prevention.

BACKGROUND: Until recently, takotsubo syndrome (TS) was considered a fairly benign disease in both early and late periods. However, in recent years, in-hospital complications in the acute period of TS are common and can be life-threatening. AIM: Based on clinical and laboratory-instrumental data, this study aimed to build models for predicting the risk of developing in-hospital complications in patients with TS during the acute period. MATERIALS AND METHODS: This prospective cohort study included 60 patients with TS, with an average age of 65.5±13.4 years. In the acute period (7–14 days), standard clinical and laboratory examination, peripheral arterial tonometry using the «EndoPAT 2000» apparatus, and psychological testing using validated questionnaires (hospital anxiety and depression scale and Beck depression scale) were performed. RESULTS: The integrated model for predicting the risk of acute heart failure (AHF) development (pulmonary edema and cardiogenic shock) identified admission LVEF as the leading prognostic parameter. With LVEF ≤40.5%, the probability of AHF in patients with TS in the acute period was 62.5%, and in patients with TS and QTc interval 487 ms, the risk reached 100%. The sensitivity and specificity of the developed model were 72.7% and 97.4%, respectively. The integrated model for predicting the development of cardiovascular complications (CVCs) revealed that the number of leukocytes in the peripheral blood was the leading risk factor for adverse events in patients with TS in the acute period. With a leukocyte count 11.1×109/L, the risk of CVCs in patients with TS increased to 89.9%, and if the erythrocyte count was 4.69×1012/L or ≤4.29×1012/L, it could reach 100%. The sensitivity and specificity of the resulting model were 92.6 and 97.0%, respectively. CONCLUSION: The models proposed in this study for predicting the likelihood of developing severe AHF and the overall risk of CVCs in the acute period of TS are personalized and easy to use, allowing for the selection of optimized treatment techniques.

Abstract The rise in sea levels, driven by global climate change, poses a significant threat to cultural heritage in coastal regions. Traditional risk assessment methods, focusing on direct inundation, often fail to consider the crucial impact of socio-economic factors, which are significantly vulnerable to sea level rise. To bridge this gap, this study introduces an innovative Sea Level Rise Cultural Heritage Impact Assessment Model (SLR-CHIA Model), a novel approach that integrates both land inundation and socio-economic aspects. This comprehensive model evaluates potential risks to various types of cultural heritage in coastal China, including intangible cultural heritage, relics, and traditional villages. The study’s findings are striking: (1) About 7.79% of coastal villages, 53.94% of relics, and 2.53% of intangible cultural heritage are potentially at high risk in a 100-year sea level rise event; (2) Relics in the Eastern coast and villages in the Southern coast are most vulnerable; (3) Different types of cultural heritage rely on diverse principal factors for protection, resulting in varied risk levels under sea level rise conditions. The SLR-CHIA Model provides a vital methodological framework for evaluating cultural heritage risks in other global regions.

Abstract Electrocardiography (ECG) is an accessible diagnostic tool for screening patients with hypertensive left ventricular hypertrophy (LVH). However, its diagnostic sensitivity is low, with a high probability of false‐negatives. Thus, this study aimed to establish a clinically useful nomogram to supplement the assessment of LVH in patients with hypertension and without ECG‐LVH based on Cornell product criteria (low‐risk hypertensive population). A cross‐sectional dataset was used for model construction and divided into development (n = 2906) and verification (n = 1447) datasets. A multivariable logistic regression risk model and nomogram were developed after screening for risk factors. Of the 4353 low‐risk hypertensive patients, 673 (15.4%) had LVH diagnosed by echocardiography (Echo‐LVH). Eleven risk factors were identified: hypertension awareness, duration of hypertension, age, sex, high waist‐hip ratio, education level, tea consumption, hypochloremia, and other ECG‐LVH diagnostic criteria (including Sokolow–Lyon, Sokolow–Lyon products, and Peguero–Lo Presti). For the development and validation datasets, the areas under the curve were 0.724 (sensitivity = 0.606) and 0.700 (sensitivity = 0.663), respectively. After including blood pressure, the areas under the curve were 0.735 (sensitivity = 0.734) and 0.716 (sensitivity = 0.718), respectively. This novel nomogram had a good predictive ability and may be used to assess the Echo‐LVH risk in patients with hypertension and without ECG‐LVH based on Cornell product criteria.

To solve the problems of high sampling requirements and low predictive accuracy resulting from the complexity, uniqueness, and randomness of predicting the risk of the CO2 and N2 injection to enhance coal seam gas drainage (CO2/N2–ECGD) technology. The principal component analysis (PCA) method to reduce the dimensionality of the factor data that contribute to the effect risk of the technology was adopted. And the particle swarm optimization (PSO) method was implemented to search for optimal hyperparameters in support vector machine (SVM) by particle search, as a solution to the traditional SVM hyperparameters optimization problem. A novel risk prediction model using machine learning algorithms for gas injection displacement technology was constructed. The prediction results were tested and compared with those of backpropagation (BP), Random Forest (RF), and Decision Tree (DT) models using data from 29 gas injection displacement field projects in China. The results demonstrated that the SVM model had greater accuracy in prediction than the other three models. Additionally, after PSO optimization and dimensionality reduction, the PCA–PSO–SVM model reached 100% prediction accuracy, while requiring less modeling and operation time. The study provided a reliable and reasonable model for predicting technical effects, along with a theoretical basis for risk management and prevention. First, the technology's influencing indicators were analyzed by examining its mechanisms. Second, we utilized the PCA method to reduce the dimensionality of the factor data that contribute to the risk of the technology's effects. Third, we implemented the PSO method to search for optimal hyperparameters in the SVM through particle search, as a solution to the traditional SVM hyperparameters optimization problem. Finally, the prediction results were tested and compared with those of BP, RF, and DT models using data from 29 gas injection displacement field projects in China. The SVM model was found to have greater accuracy in prediction than the other three models. After PSO optimization and dimensionality reduction, the PCA–PSO–SVM model achieved 100% prediction accuracy while requiring less modeling and operation time. The study presents a valid and reasonable model for predicting technical effects and a theoretical basis for risk management and prevention. [ABSTRACT FROM AUTHOR]

The rise in sea levels, driven by global climate change, poses a significant threat to cultural heritage in coastal regions. Traditional risk assessment methods, focusing on direct inundation, often fail to consider the crucial impact of socio-economic factors, which are significantly vulnerable to sea level rise. To bridge this gap, this study introduces an innovative Sea Level Rise Cultural Heritage Impact Assessment Model (SLR-CHIA Model), a novel approach that integrates both land inundation and socio-economic aspects. This comprehensive model evaluates potential risks to various types of cultural heritage in coastal China, including intangible cultural heritage, relics, and traditional villages. The study's findings are striking: (1) About 7.79% of coastal villages, 53.94% of relics, and 2.53% of intangible cultural heritage are potentially at high risk in a 100-year sea level rise event; (2) Relics in the Eastern coast and villages in the Southern coast are most vulnerable; (3) Different types of cultural heritage rely on diverse principal factors for protection, resulting in varied risk levels under sea level rise conditions. The SLR-CHIA Model provides a vital methodological framework for evaluating cultural heritage risks in other global regions. [ABSTRACT FROM AUTHOR]

Background: Anastomotic leakage (AL) is the most serious complication after rectal cancer surgery. Risk factors associated with AL have been documented in previous studies; however, the consensus is still lacking. In this retrospective study, we aimed to identify risk factors for AL after rectal cancer resection and to create an accurate and effective tool for predicting the risk of this complication. Methods: The study cohort comprised of 276 patients with rectal cancer who had undergone anterior resection between 2015 and 2020. Twenty-four selected variables were assessed by univariate and multivariate logistic regression analyses to identify independent risk factors of AL. A risk assessment model for predicting the risk of AL was established on the basis of the regression coefficients of each identified independent risk factor. Results: Anastomotic leakage occurred in 20 patients (7.2%, 20/276). Multivariate analysis identified the following variables as independent risk or protective factors of AL: perioperative ileus (P < 0.001, odds ratio [OR] = 14.699), tumor size ≥5 cm (P = 0.025, OR = 3.925), distance between tumor and anal verge <7.5 cm (P = 0.045, OR = 3.512), obesity (P = 0.032, OR = 7.256), and diverting stoma (P = 0.008, OR = 0.143). A risk assessment model was constructed and patients were allocated to high-, medium-, and low-risk groups on the basis of risk model scores of 5–7, 2–4, and 0–1, respectively. The incidences of AL in these three groups were 61.5%, 11.9%, and 2.0%, respectively (P < 0.001). Conclusions: Our risk assessment model accurately and effectively identified patients at high risk of AL and could be useful in aiding decision-making aimed at minimizing adverse outcomes associated with leakage. [ABSTRACT FROM AUTHOR]

Background: The goal was to develop a risk assessment model for predicting red blood cell (RBC) transfusion in neonatal patients to assist hospital blood supply departments in providing small portions of RBCs to those requiring RBC transfusion on time. Methods: Clinical information was collected from 1,201 children admitted to the neonatal unit. Clinical factors associated with predicting RBC transfusion were screened, and prediction models were developed using stepwise and multifactorial logistic regression analyses, followed by the evaluation of prediction models using receiver operating characteristic curves, calibration curves, and decision curve analysis (DCA). Results: Overall, 81 neonatal patients were transfused with RBCs, and the variables of gestational age at birth, age < 1 month, receipt of mechanical ventilation, and infant anemia were included in the final prediction model. The area under the curve of the prediction model was 0.936 (0.921 - 0.949), which was significantly higher than that of the individual indicators of gestational age at birth, age at admission < 1 month, receipt of mechanical ventilation, and infant anemia (p < 0.001). DCA showed a standardized net benefit for the possible risk of infant RBC transfusion at 0.1 - 1.0. Conclusions: We developed a risk assessment model to predict the risk of RBC transfusion in neonatal patients that can effectively assess the risk of RBC transfusion in children. [ABSTRACT FROM AUTHOR]

Transportation stands as a fundamental industry that serves as a robust pillar for driving economic development. However, the occurrence of traffic incidents, encompassing traffic accidents and disruptions triggered by meteorological conditions such as rainfall and other atmospheric factors, exerts a substantial impact on both transport safety and the safeguarding of people's property. Consequently, it becomes imperative to undertake a meticulous examination of the risk assessment associated with rainfall-induced meteorological disasters in the realm of road traffic. The focus of this study lies in addressing the urgent requirements associated with the assessment of meteorological disaster risks triggered by rainfall in the field of road traffic. Additionally, it aims to tackle the current issues in domestic research on the assessment of meteorological disaster risks in highways, characterized by a singular research scale and method. Centered around the demand for the assessment of rainfall-induced meteorological disaster risks in road traffic, the study first adopts the "urban agglomeration—city—block" multi-scale integration as a starting point. It establishes a multi-scale integrated risk assessment indicator system based on the dimensions of regional spatial scale and road grade, employing a hierarchical division approach grounded in the pyramid principle. Furthermore, the study develops a novel indicator system weight calculation method that combines both subjective and objective elements, enhancing the comprehensiveness and reliability of the assessment model. Using this assessment model, the paper conducts a study on the risk assessment and zoning of rainfall-induced meteorological disasters on highways in the Wuhan 1 + 8 agglomeration of China, the main roads in Wuhan City, and the main roads in the Hongshan District. The outcome of the study reveals the distribution of risk levels across the examined regions. Notably, within the Wuhan 1 + 8 urban agglomeration, approximately 73% of the area is designated as a high-risk zone, whereas 25% and 3% represent medium- and low-risk areas, respectively. In Wuhan City, 2% of the territory falls under the high-risk category, with 73% classified as medium risk and 25% as low risk. Similarly, within the Hongshan District, 30% is identified as a medium-risk area, while 70% is categorized as low risk. In summary, through practical testing, the research endeavors concerning risk assessment methods conducted in this study serve as an invaluable point of reference for both theoretical and methodological investigations within the realm of meteorological disaster risk assessment. Furthermore, it signifies an efficacious exploration into conducting multi-scale assessments of meteorological disaster risks. Additionally, the study holds the potential to provide methodological and technical support to the pertinent departments involved in the Wuhan 1 + 8 urban agglomeration, thereby contributing to the prevention of highway meteorological disasters. [ABSTRACT FROM AUTHOR]

Aiming to accurately assess the intensity of rockburst in coal mines, we propose a rockburst risk assessment model using the improved catastrophe progression method based on its sudden, complex, and non-linear characteristics. A risk assessment indicator system for rockburst is established based on the occurrence conditions of rockburst, which comprehensively considers 10 main influencing factors. We introduce a combined weighting method consisting of the variation coefficient method and analytic hierarchy process (AHP) to determine the weight and ranking of evaluation indicators, which improves the catastrophe progression method. Finally, by applying 10 coal mine cases into the established model and comparing them with the unimproved catastrophe progression method and several other risk assessment methods, we conclude that the accuracy of the model in assessing rockburst intensity level has increased by a maximum of 42.85%. Our work proves the effectiveness and practicality of the risk assessment model, which can provide good theoretical guidance for assessing rockburst intensity level in coal mines. Furthermore, based on the model risk assessment results and comparative analysis, this paper provides feasible suggestions for reducing the intensity of rockbursts. [ABSTRACT FROM AUTHOR]

Electrocardiography (ECG) is an accessible diagnostic tool for screening patients with hypertensive left ventricular hypertrophy (LVH). However, its diagnostic sensitivity is low, with a high probability of false‐negatives. Thus, this study aimed to establish a clinically useful nomogram to supplement the assessment of LVH in patients with hypertension and without ECG‐LVH based on Cornell product criteria (low‐risk hypertensive population). A cross‐sectional dataset was used for model construction and divided into development (n = 2906) and verification (n = 1447) datasets. A multivariable logistic regression risk model and nomogram were developed after screening for risk factors. Of the 4353 low‐risk hypertensive patients, 673 (15.4%) had LVH diagnosed by echocardiography (Echo‐LVH). Eleven risk factors were identified: hypertension awareness, duration of hypertension, age, sex, high waist‐hip ratio, education level, tea consumption, hypochloremia, and other ECG‐LVH diagnostic criteria (including Sokolow–Lyon, Sokolow–Lyon products, and Peguero–Lo Presti). For the development and validation datasets, the areas under the curve were 0.724 (sensitivity = 0.606) and 0.700 (sensitivity = 0.663), respectively. After including blood pressure, the areas under the curve were 0.735 (sensitivity = 0.734) and 0.716 (sensitivity = 0.718), respectively. This novel nomogram had a good predictive ability and may be used to assess the Echo‐LVH risk in patients with hypertension and without ECG‐LVH based on Cornell product criteria. [ABSTRACT FROM AUTHOR]

BackgroundMitochondria have always been considered too be closely related to the occurrence and development of malignant tumors. However, the bioinformatic analysis of mitochondria in lung adenocarcinoma (LUAD) has not been reported yet.MethodsIn the present study, we constructed a novel and reliable algorithm, comprising a consensus cluster analysis and risk assessment model, to predict the survival outcomes and tumor immunity for patients with terminal LUAD.ResultsPatients with LUAD were classified into three clusters, and patients in cluster 1 exhibited the best survival outcomes. The patients in cluster 3 had the highest expression of PDL1 (encoding programmed cell death 1 ligand 11) and HAVCR2 (encoding Hepatitis A virus cellular receptor 2), and the highest tumor mutation burden (TMB). In the risk assessment model, patients in the low-risk group tended to have a significantly better survival outcome. Furthermore, the risk score combined with stage could act as a reliable independent prognostic indicator for patients with LUAD. The prognostic signature is a novel and effective biomarker to select anti-tumor drugs. Low-risk patients tended to have a higher expression of CTLA4 (encoding cytotoxic T-lymphocyte associated protein 4) and HAVCR2. Moreover, patients in the high-risk group were more sensitive to Cisplatin, Docetaxel, Erlotinib, Gemcitabine, and Paclitaxel, while low-risk patients would probably benefit more from Gefitinib.ConclusionWe constructed a novel and reliable algorithm comprising a consensus cluster analysis and risk assessment model to predict survival outcomes, which functions as a reliable guideline for anti-tumor drug treatment for patients with terminal LUAD.

Objective To investigate the risk factors of postoperative lower extremity deep venous thrombosis(DVT) in elderly patients with cerebral hemorrhage,and to construct a DVT risk assessment model. Methods A total of ninety five elderly patients with cerebral hemorrhage who underwent surgery in Jiangsu Provincial People's Hospital from July 2020 to January 2023 were selected as the study subjects and divided into DVT group(25 cases) and non-DVT group(70 cases) according to whether DVT occurred or not after surgery.The general data,preoperative and intraoperative related factors of the two groups were compared,and univariate analysis and multivariate logistic regression model analysis were used to screen the risk factors for the occurrence of DVT and construct a DVT risk assessment model. Results There were significant differences in age,smoking history,hypertension history,heart disease history,cerebral hemorrhage volume and operation time between the DVT group and the non-DVT group(P

Abstract Non-small cell lung cancer (NSCLC) is a malignant tumor that threatens human life and health. The development of a new NSCLC risk assessment model based on electronic medical records has great potential for reducing the risk of cancer recurrence. In this process, machine learning is a powerful method for automatically extracting risk factors and indicating impact weights for NSCLC deaths. However, when the number of samples reaches a certain value, it is difficult for machine learning to improve the prediction accuracy, and it is also challenging to use the characteristic data of subsequent patients effectively. Therefore, this study aimed to build a postoperative survival risk assessment model for patients with NSCLC that updates the model parameters and improves model accuracy based on new patient data. The model perspective was a combination of particle filtering and parameter estimation. To demonstrate the feasibility and further evaluate the performance of our approach, we performed an empirical analysis experiment. The study showed that our method achieved an overall accuracy of 92% and a recall of 71% for deceased patients. Compared with traditional machine learning models, the accuracy of the model estimated by particle filter parameters has been improved by 2%, and the recall rate for dead patients has been improved by 11%. Additionally, this study outcome shows that this method can better utilize subsequent patients’ characteristic data, be more relevant to different patients, and help achieve precision medicine.

Abstract Background Deep vein thrombosis (DVT) is a common complication in orthopedic patients. Previous studies have focused on major orthopedic surgery.There are few studies with multiple trauma. We aimed to describe the prevalence of DVT and compare the predictive power of the different risk assessment scales in patients with multiple trauma. Methods This prospective cohort study involved multiple trauma patients admitted to our hospital between October 2021 and December 2022. Data were prospectively collected for thrombotic risk assessments using the Risk Assessment Profile for thromboembolism(RAPT), the DVT risk assessment score (DRAS), and the Trauma Embolic Scoring System (TESS), respectively. The receiver operation characteristic (ROC) curve and the area under the curve (AUC) were evaluated to compare the predictive power. The whole leg duplex ultrasound of both lower extremities Doppler ultrasound was used to determine DVT incidence. Results A total of 210 patients were included, and the incidence of DVT was 26.19%. Distal DVT accounted for 87.27%; postoperative DVT, 72.73%; and bilateral lower extremity thrombosis, 30.91%. There were significant differences in age, education degree, pelvic fracture, surgery, ISS, D-dimer level, length of hospital stay and ICU stay between the thrombosis group and the non-thrombosis group. The AUCs for RAPT, DRAS, and TESS were 0.737, 0.710, and 0.683, respectively. There were no significant differences between the three ROC curves. Conclusions The incidence of DVT was relatively high during hospitalization. We prospectively validated the tests to predict risk of DVT among patients with multiple trauma to help trauma surgeons in the clinical administration of DVT prophylaxis.

Objective To establish a risk assessment model of cardiometabolic multimorbidity (CMM) and validate the model. Methods The influencing factors of CMM were determined by meta-analysis, and the population exposure rate was obtained by consulting a large cohort study. The CMM risk assessment model was constructed based on Harvard cancer risk index method. The model was verified by the data of China Health and Retirement Longitudinal Study (CHARLS). The SPSS 25.0 software was used to calculate the area under the receiver operator characteristic ROC curve (AUC), sensitivity, specificity, and Hosmer-Lemshow goodness-of-fit test to verify the effect of the model. Results A total of 20 articles were included in the meta-analysis. Age, sex, place of residence, education, economic level, BMI, waistline, night sleep, depression, irregular diet, and insufficient intake of vegetables and fruits were the risk factors of CMM. Concerning the model validation results, the area under the ROC curve was 0,712 (95%CZ: 0.696-0.728), the Jordan index was 1.42, the sensitivity was 65%, the specificity was 66.1%, and the result of Hosmer-Lemshow goodness-of-fit test was 於=10,652 (P=0.222). Conclusion The CMM risk assessment model based on meta -analysis has good predictive performance and can be used as a tool for CMM risk assessment to guide clinical identification of CMM high-risk groups and implement intervention measures. [ABSTRACT FROM AUTHOR]

Standardized use of venous thromboembolism (VTE) risk assessment models (RAMs) in surgical patients has been limited, in part due to the cumbersome workflow addition required to use available models. The COBRA score—capturing cancer diagnosis, (old) age, body mass index, race, and American Society of Anesthesiologists Physical Status score—has been reported as a potentially automatable VTE RAM that circumvents the cumbersome workflow addition that most RAMs represent. We aimed to test the ability of the COBRA model to effectively risk-stratify patients across various populations. Patients were included from the 2014-2019 American College of Surgeons National Surgical Quality Improvement Program (NSQIP) Participant Use Data File for two hospitals, representing colorectal, endocrine, breast, transplant, plastic, and general surgery services. COBRA score was calculated for each patient using preoperative characteristics. We calculated negative predictive value (NPV) for VTE outcomes and compared the COBRA score to NSQIP's expected VTE rate for all patients, between the two hospitals, and between subspecialty service lines. Of the 10,711 patients included, those with COBRA <4 (31%) had projected median VTE rate of 0.21% (interquartile range, 0.09-0.68%; mean, 0.54%). Patients with higher scores (69%) had median rate of 0.88% (0.26-2.07%; 1.46%); relative rate 2.7. The median projected VTE rates for patients identified as low risk were 0.21% and 0.16% and as high risk were 0.87% and 0.89% at hospitals one and 2, respectively. The median projected VTE rates for patients identified as low risk were 0.17%, 0.61%, and 0.08% and as high risk were 0.52%, 1.43%, and 0.18% among general, colorectal, and endocrine surgery patients, respectively. COBRA had NPV of 0.995 and sensitivity of 0.871 as compared to NPV 0.997 and sensitivity 0.857 of the NSQIP model. The COBRA score is concordant with the traditional gold standard NSQIP VTE RAM and demonstrates interhospital and service-specific generalizability, although performance was limited in especially low-risk patients. The model adequately risk-stratifies surgical patients preoperatively, potentially providing clinical decision support for perioperative workflows. [ABSTRACT FROM AUTHOR]

With the introduction of the concept of the industry 4.0, automation, robotics, artificial intelligence, communication methods, automotive engineering, mechanics, construction and operation of automotive vehicles, and so on, as well as the methods of corporate management are changing. Following this concept, new risks emerge, when workers have to cooperate with collaborative robots, autonomous systems, artificial intelligence, machine learning and learn new methods different from previous processes and systems. The paper first presents the theoretical background related to the topic addressed. The next sections encompass the literature review, including a list of references relevant to achieving the main objective of the paper, as well as a description of the research methods used in the paper. With regard to the main objective, quantitative research concerning the vehicle construction systems' safety issues in industry 4.0 was conducted; i.e., a questionnaire survey was developed within a sufficiently representative sample of respondents. After conducting the survey, the risk assessment model of vehicle construction systems' safety under the conditions of Industry 4.0 was proposed while applying the principles of system dynamics. An integral part of the paper is represented by the discussion of the obtained results and benefits, as well as the formulation of relevant conclusions. [ABSTRACT FROM AUTHOR]

Non-small cell lung cancer (NSCLC) is a malignant tumor that threatens human life and health. The development of a new NSCLC risk assessment model based on electronic medical records has great potential for reducing the risk of cancer recurrence. In this process, machine learning is a powerful method for automatically extracting risk factors and indicating impact weights for NSCLC deaths. However, when the number of samples reaches a certain value, it is difficult for machine learning to improve the prediction accuracy, and it is also challenging to use the characteristic data of subsequent patients effectively. Therefore, this study aimed to build a postoperative survival risk assessment model for patients with NSCLC that updates the model parameters and improves model accuracy based on new patient data. The model perspective was a combination of particle filtering and parameter estimation. To demonstrate the feasibility and further evaluate the performance of our approach, we performed an empirical analysis experiment. The study showed that our method achieved an overall accuracy of 92% and a recall of 71% for deceased patients. Compared with traditional machine learning models, the accuracy of the model estimated by particle filter parameters has been improved by 2%, and the recall rate for dead patients has been improved by 11%. Additionally, this study outcome shows that this method can better utilize subsequent patients' characteristic data, be more relevant to different patients, and help achieve precision medicine. [ABSTRACT FROM AUTHOR]

Background: Deep vein thrombosis (DVT) is a common complication in orthopedic patients. Previous studies have focused on major orthopedic surgery.There are few studies with multiple trauma. We aimed to describe the prevalence of DVT and compare the predictive power of the different risk assessment scales in patients with multiple trauma. Methods: This prospective cohort study involved multiple trauma patients admitted to our hospital between October 2021 and December 2022. Data were prospectively collected for thrombotic risk assessments using the Risk Assessment Profile for thromboembolism(RAPT), the DVT risk assessment score (DRAS), and the Trauma Embolic Scoring System (TESS), respectively. The receiver operation characteristic (ROC) curve and the area under the curve (AUC) were evaluated to compare the predictive power. The whole leg duplex ultrasound of both lower extremities Doppler ultrasound was used to determine DVT incidence. Results: A total of 210 patients were included, and the incidence of DVT was 26.19%. Distal DVT accounted for 87.27%; postoperative DVT, 72.73%; and bilateral lower extremity thrombosis, 30.91%. There were significant differences in age, education degree, pelvic fracture, surgery, ISS, D-dimer level, length of hospital stay and ICU stay between the thrombosis group and the non-thrombosis group. The AUCs for RAPT, DRAS, and TESS were 0.737, 0.710, and 0.683, respectively. There were no significant differences between the three ROC curves. Conclusions: The incidence of DVT was relatively high during hospitalization. We prospectively validated the tests to predict risk of DVT among patients with multiple trauma to help trauma surgeons in the clinical administration of DVT prophylaxis. [ABSTRACT FROM AUTHOR]

Although several scores stratify venous thromboembolism (VTE) risk in solid tumors, hematologic malignancies (HM) are underrepresented. To develop an internal and external validation of a logistic regression model to predict VTE risk in hospitalized HM patients. Validation of the existing VTE predictive model was performed through a prospective case–control study in 496 hospitalized HM patients between December 2010 and 2020 at the Arnaldo Milián University Hospital, Cuba. The predictive model designed with data from 285 patients includes 5 predictive factors: hypercholesterolemia, tumoral activity, use of thrombogenic drugs, diabetes mellitus, and immobilization. The model was internally validated using bootstrap analysis. External validation was realized in a prospective cohort of 211 HM patients. The predictive model had a 76.4% negative predictive value (NPV) and an 81.7% positive predictive value (PPV) in the bootstrapping validation. The area under curve (AUC) in the bootstrapping set was 0.838. Accuracy was 80.1% and 82.9% in the internal and external validation, respectively. In the external validation, the model produced 89.7% of NPV, 67.7% of PPV, 74.6% of sensitivity, and 86.2% of specificity. The AUC in the external validation was 0.900. VTE predictive model is a reproducible and simple tool with good accuracy and discrimination. [ABSTRACT FROM AUTHOR]

Various health risk assessment models have been developed to evaluate occupational pesticide exposure in China. However, there has been limited investigation into the relationship between health risks and pesticide spraying in orchards. In this study, we analyzed pesticide exposure of applicators while spraying with a stretcher-mounted sprayer in orchards located in four different climatic regions. All garments’ unit exposure (UE) demonstrated a right-skewed distribution, with gloves and shins accounting for the highest proportion of dermal pesticide exposure. We observed little difference in dermal and inhalation UE levels between apple and citrus orchards, except for pesticide exposure levels on wipes and faces. While 57% of the inhalation UE distribution variance was attributed to clustering and location effects, no significant differences were observed in dermal exposure levels. We evaluated the impact of different levels of protective clothing on pesticide exposure levels, according to applicators' working habits in China. Our findings revealed that improved levels of protection significantly reduced dermal exposure to pesticides, particularly when wearing gloves during spraying with a stretcher-mounted sprayer. Based on our empirical data, we utilized a simple random sampling model and an intercept-only lognormal mixed model to estimate dermal and inhalation exposure levels. The estimated dermal UE was accurate to within 3-fold with 95% confidence, and half of the estimated inhalation UE was acceptable according to the fold relative accuracy (fRA). Our established and verified statistics for dermal and inhalation UE can be utilized to evaluate the potential pesticide exposure to applicators during spraying in orchards with a stretcher-mounted sprayer.

Abstract Background Individuals with multiple myeloma (MM) receiving immunomodulatory drugs (IMiDs) are at risk of developing venous thromboembolism (VTE), a serious complication. There is no established clinical model for predicting VTE in the Chinese population. We develop a new risk assessment model (RAM) for IMiD-associated VTE in Chinese MM patients. Methods We retrospectively selected 1334 consecutive MM patients receiving IMiDs from 16 medical centers in China and classified them randomly into the derivation and validation cohorts. A multivariate Cox regression model was used for analysis. Results The overall incidence of IMiD-related VTE in Chinese MM patients was 6.1%. Independent predictive factors of VTE (diabetes, ECOG performance status, erythropoietin-stimulating agent use, dexamethasone use, and VTE history or family history of thrombosis) were identified and merged to develop the RAM. The model identified approximately 30% of the patients in each cohort at high risk for VTE. The hazard ratios (HRs) were 6.08 (P

Abstract Background This study investigated the relationship between fibroblast growth factor 21 (FGF-21) and newly diagnosed type-2 diabetes mellitus (T2DM). Methods In this cross-sectional study, FGF-21 and T2DM risk were analyzed using restricted cubic splines with univariate or multivariate logistic regression analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated via logistic regression analysis. Cluster and subgroup analyses were conducted to evaluate the associations between FGF-21 and diabetes in different subpopulations. Nomograms and ROC curves were used to explore the clinical utility of FGF-21 in the diabetes assessment model. Results High levels of FGF-21 were significantly associated with a high risk of T2DM after adjusting for confounding factors in both the total population and subpopulations (P for trend

Introduction: Viruses have been reported as inducers of tumorigenesis. Little studies have explored the impact of the gut virome on the progression of colorectal cancer. However, there is still a problem with the repeatability of viral signatures across multiple cohorts. Objectives: The present study aimed to reveal the repeatable gut vial signatures of colorectal cancer and adenoma patients and decipher the potential of viral markers in disease risk assessment for diagnosis. Methods: 1,282 available fecal metagenomes from 9 published studies for colorectal cancer and adenoma were collected. A gut viral catalog was constructed via a reference-independent approach. Viral signatures were identified by cross-cohort meta-analysis and used to build predictive models based on machine learning algorithms. New fecal samples were collected to validate the generalization of predictive models. Results: The gut viral composition of colorectal cancer patients was drastically altered compared with healthy, as evidenced by changes in some Siphoviridae and Myoviridae viruses and enrichment of Microviridae, whereas the virome variation in adenoma patients was relatively low. Cross-cohort meta-analysis identified 405 differential viruses for colorectal cancer, including several phages of Porphyromonas, Fusobacterium, and Hungatella that were enriched in patients and some control-enriched Ruminococcaceae phages. In 9 discovery cohorts, the optimal risk assessment model obtained an average cross-cohort area under the curve of 0.830 for discriminating colorectal cancer patients from controls. This model also showed consistently high accuracy in 2 independent validation cohorts (optimal area under the curve, 0.906). Gut virome analysis of adenoma patients identified 88 differential viruses and achieved an optimal area under the curve of 0.772 for discriminating patients from controls. Conclusion: Our findings demonstrate the gut virome characteristics in colorectal cancer and adenoma and highlight gut virus-bacterial synergy in the progression of colorectal cancer. The gut viral signatures may be new targets for colorectal cancer treatment. In addition, high repeatability and predictive power of the prediction models suggest the potential of gut viral biomarkers in non-invasive diagnostic tests of colorectal cancer and adenoma.

Background: The Padua Prediction Score (PPS) recommended by the guidelines lacks effective external validation in a Chinese cohort. This study sought to assess the accuracy of the PPS to predict venous thromboembolism (VTE) risk in medical inpatients with acute respiratory conditions. Methods: This consecutive cohort study included 1,574 inpatients from January to August 2019. The occurrence rate of VTE in patients classified at high-risk and low-risk groups according to PPS and Caprini risk assessment model (RAM) was compared. The discriminatory capability of the RAMs was evaluated in all the patients and the subgroup without pharmacological prophylaxis. Reclassification parameters were also used to assess the clinical utility. Results: 170 (10.8%) patients were objectively confirmed as having VTE during hospitalization. The incidence rate of VTE in low-risk patients was 6.3% by PPS, which was significantly higher than that by Caprini RAM (2.6%, p

Objective To compare the predictive value of four different risk assessment models for postpartum venous thromboembolism (VTE). Methods A retrospective study of the women who gave a birth in Shanghai First Maternal and Infant Health Hospital, Tongji University between Aug 2017 and Dec 2022 was conducted. A total of 255 parturients with VTE were included, and 765 parturients without VTE during the same period were randomly selected as control in a ratio of 1∶3. Four risk assessment models, Caprini Score, RCOG Guidline, Queensland Guideline and Shanghai Expert Consensus, were used respectively to score and classify the two groups of patients. The ROC curves were plotted to evaluate the predictive value of different models for puerperal venous thromboembolism. Classification of VTE patients by risk level and cumulative risk score in four risk assessment models were also made. Results Maternal age, body mass index, and incidence of twin pregnancy, cesarean section, preterm birth, and preeclampsia were significantly higher in the thrombus group than in the control group (P<0.001). The AUCs of Caprini Score, RCOG Guideline, Queensland Guideline and Shanghai Expert Consensus were 0.676, 0.763, 0.782 and 0.767, respectively, and the difference was statistically significant (P<0.05). The Queensland Guideline identified more patients with thrombosis into high risk level. Conclusion RCOG Guideline, Queensland Guideline and Shanghai Expert Consensus all have higher predictive value than Caprini Score, which are suitable for predicting and evaluating VTE in puerperal period. Queensland Guideline seems to have the highest accuracy. [ABSTRACT FROM AUTHOR]

Pregnancy is a major risk factor for venous thromboembolism (VTE) and its associated complications. The hypercoagulable state in both the antenatal and postnatal periods contributes to thromboembolism and continues to be a leading cause of maternal morbidity and mortality worldwide. The non-specific signs and symptoms of VTE in pregnancy and the lack of specific Risk Assessment Models (RAMs) propose a diagnostic challenge in the obstetric population. This review aims to discuss and compare existing RAMs and highlights the important challenges of using established RAMs in obstetric patients. It also emphasizes the importance of enhancing and individualizing RAMs in obstetrics to improve maternal healthcare. Key Points VTE is a major complication of pregnancy, associated with increased maternal morbidity and mortality. VTE RAMs lack sensitivity and specificity in stratifying VTE risk in pregnancy. Validating VTE RAMs in the obstetric population aims to improve maternal outcomes. [ABSTRACT FROM AUTHOR]

Background: Individuals with multiple myeloma (MM) receiving immunomodulatory drugs (IMiDs) are at risk of developing venous thromboembolism (VTE), a serious complication. There is no established clinical model for predicting VTE in the Chinese population. We develop a new risk assessment model (RAM) for IMiD-associated VTE in Chinese MM patients. Methods: We retrospectively selected 1334 consecutive MM patients receiving IMiDs from 16 medical centers in China and classified them randomly into the derivation and validation cohorts. A multivariate Cox regression model was used for analysis. Results: The overall incidence of IMiD-related VTE in Chinese MM patients was 6.1%. Independent predictive factors of VTE (diabetes, ECOG performance status, erythropoietin-stimulating agent use, dexamethasone use, and VTE history or family history of thrombosis) were identified and merged to develop the RAM. The model identified approximately 30% of the patients in each cohort at high risk for VTE. The hazard ratios (HRs) were 6.08 (P < 0.001) and 6.23 (P < 0.001) for the high-risk subcohort and the low-risk subcohort, respectively, within both the derivation and validation cohorts. The RAM achieved satisfactory discrimination with a C statistic of 0.64. The stratification approach of the IMWG guidelines yielded respective HRs of 1.77 (P = 0.053) and 1.81 (P = 0.063). The stratification approach of the SAVED score resulted in HRs of 3.23 (P = 0.248) and 1.65 (P = 0.622), respectively. The IMWG guideline and the SAVED score-based method yielded C statistics of 0.58 and 0.51, respectively. Conclusions: The new RAM outperformed the IMWG guidelines and the SAVED score and could potentially guide the VTE prophylaxis strategy for Chinese MM patients. [ABSTRACT FROM AUTHOR]

Simple Summary: The predictive value of existing venous thromboembolism risk assessment models (RAMs) in lung cancer patients is still debated, and the design of new models represents an unmet clinical need. In a prospective cohort of patients with newly diagnosed metastatic lung cancer, clinical characteristics, and hemostatic biomarkers assessed before initiating chemotherapy were used to generate a more accurate RAM. This easy-to-implement RAM was compared to four previously published scores, which were also externally validated in this study. (1) Background: Venous thromboembolism (VTE) is a frequent complication in ambulatory lung cancer patients during chemotherapy and is associated with increased mortality. (2) Methods: We analyzed 568 newly diagnosed metastatic lung cancer patients prospectively enrolled in the HYPERCAN study. Blood samples collected before chemotherapy were tested for thrombin generation (TG) and a panel of hemostatic biomarkers. The Khorana risk score (KRS), new-Vienna CATS, PROTECHT, and CONKO risk assessment models (RAMs) were applied. (3) Results: Within 6 months, the cumulative incidences of VTE and mortality were 12% and 29%, respectively. Patients with VTE showed significantly increased levels of D-dimer, FVIII, prothrombin fragment 1 + 2, and TG. D-dimer and ECOG performance status were identified as independent risk factors for VTE and mortality by multivariable analysis and utilized to generate a risk score that provided a cumulative incidence of VTE of 6% vs. 25%, death of 19% vs. 55%, and in the low- vs. high-risk group, respectively (p < 0.001). While all published RAMs significantly stratified patients for risk of death, only the CATS and CONKO were able to stratify patients for VTE. (4) Conclusions: A new prediction model was generated to stratify lung cancer patients for VTE and mortality risk, where other published RAMs failed. [ABSTRACT FROM AUTHOR]

Background: This study investigated the relationship between fibroblast growth factor 21 (FGF-21) and newly diagnosed type-2 diabetes mellitus (T2DM). Methods: In this cross-sectional study, FGF-21 and T2DM risk were analyzed using restricted cubic splines with univariate or multivariate logistic regression analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated via logistic regression analysis. Cluster and subgroup analyses were conducted to evaluate the associations between FGF-21 and diabetes in different subpopulations. Nomograms and ROC curves were used to explore the clinical utility of FGF-21 in the diabetes assessment model. Results: High levels of FGF-21 were significantly associated with a high risk of T2DM after adjusting for confounding factors in both the total population and subpopulations (P for trend < 0.001). In the total population, the ORs of diabetes with increasing FGF-21 quartiles were 1.00 (reference), 1.24 (95% CI 0.56–2.80; quartile 2), 2.47 (95% CI 1.18–5.33; quartile 3), and 3.24 (95% CI 1.53–7.14; quartile 4) in Model 4 (P < 0.001), and the trend was consistent in different subpopulations. In addition, compared with the model constructed with conventional noninvasive indicators, the AUC of the model constructed by adding FGF-21 was increased from 0.668 (95% CI: 0.602–0.733) to 0.715 (95% CI: 0.654–0.777), indicating that FGF-21 could significantly improve the risk-assessment efficiency of type-2 diabetes. Conclusion: This study demonstrated that a high level of circulating FGF-21 was positively correlated with diabetes, and levels of FGF-21 could be an important biomarker for the assessment of diabetes risk. [ABSTRACT FROM AUTHOR]

The quick access recorder (QAR), as an airborne device used to monitor and record flight parameters, has been widely installed on various types of aircraft. Based on QAR data, research on runway overrun, a typical flight safety incident, has attracted widespread attention in recent years. However, existing runway overrun risk models generally suffer from oversimplified risk metrics or insufficient consideration of risk dynamics. In this paper, we propose a new dynamic runway overrun risk assessment model based on QAR data. We first consider the noise of aircraft trajectory data in the QAR parameters and present a landing trajectory correction method combining ground speed and runway position information. Second, to improve the accuracy of the risk assessment model, we design an algorithm to automatically recognize the aircraft autobrake level during the landing phase, based on which a new dynamic risk assessment model is developed. Finally, feature engineering is performed to extract the relevant contributing factors of runway overrun risk, based on which classification and regression models are applied to identify risky flights and predict the risk values. The proposed risk assessment model was evaluated using QAR data from an airline in China. The results show that the automatic deceleration rate, the way that the aircraft approaches the runway, the touchdown distance, and the kinetic energy at 50 ft are key factors in the risk of runway overrun during the landing phase. [ABSTRACT FROM AUTHOR]

Since the onset of the pandemic, the cruise industry has faced substantial challenges, experiencing an 81% year-on-year decline in 2020. Notably, China's cruise industry has remained shuttered for nearly two years. The conventional epidemiological investigations relying on human memory have proven inadequate in regard to swiftly and reliably identifying high-risk populations, thus leading to excessive pandemic prevention or other inappropriate measures. Furthermore, current research endeavors have not adequately addressed the critical issues of isolation location selection and the estimation of isolation scale within multi-zone indoor environments. Therefore, how to control the epidemic with minimal impact on the public has become an urgent problem. To solve the problems mentioned above, a multi-zone infection risk assessment model of airborne virus transmission was proposed to rapidly qualify the risk of infection, identify the high-risk population, and provide guidance. The model can be divided into two parts, including a multi-zone airflow model and a risk assessment model based on the Wells–Riley model. A large in-service ro-ro passenger ship already in service was used for the application of the proposed method. The results show that the number of isolated rooms shall be at least 5% of the occupancy and the isolation location can be chosen through the multi-zone infection risk assessment model. This paper provides insights into risk assessment to mitigate the spread of epidemics on a large cruise ship, and the method can be easily applied to all kinds of multi-zone indoor environments. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Railway Science & Engineering is the property of Journal of Railway Science & Engineering Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

[Display omitted] • The gut virome research in colorectal cancer and adenoma includes >1,200 samples. • Siphoviridae and Microviridae viruses were notably different between controls and CRC patients. • The viral markers of colorectal cancer and adenoma were identified. • For CRC patients, our model had better predictive ability than other bacteria-based models. • The virome analysis achieved an optimal AUC of 0.772 to distinct adenoma patients and controls. Viruses have been reported as inducers of tumorigenesis. Little studies have explored the impact of the gut virome on the progression of colorectal cancer. However, there is still a problem with the repeatability of viral signatures across multiple cohorts. The present study aimed to reveal the repeatable gut vial signatures of colorectal cancer and adenoma patients and decipher the potential of viral markers in disease risk assessment for diagnosis. 1,282 available fecal metagenomes from 9 published studies for colorectal cancer and adenoma were collected. A gut viral catalog was constructed via a reference-independent approach. Viral signatures were identified by cross-cohort meta -analysis and used to build predictive models based on machine learning algorithms. New fecal samples were collected to validate the generalization of predictive models. The gut viral composition of colorectal cancer patients was drastically altered compared with healthy, as evidenced by changes in some Siphoviridae and Myoviridae viruses and enrichment of Microviridae, whereas the virome variation in adenoma patients was relatively low. Cross-cohort meta -analysis identified 405 differential viruses for colorectal cancer, including several phages of Porphyromonas, Fusobacterium, and Hungatella that were enriched in patients and some control-enriched Ruminococcaceae phages. In 9 discovery cohorts, the optimal risk assessment model obtained an average cross-cohort area under the curve of 0.830 for discriminating colorectal cancer patients from controls. This model also showed consistently high accuracy in 2 independent validation cohorts (optimal area under the curve, 0.906). Gut virome analysis of adenoma patients identified 88 differential viruses and achieved an optimal area under the curve of 0.772 for discriminating patients from controls. Our findings demonstrate the gut virome characteristics in colorectal cancer and adenoma and highlight gut virus-bacterial synergy in the progression of colorectal cancer. The gut viral signatures may be new targets for colorectal cancer treatment. In addition, high repeatability and predictive power of the prediction models suggest the potential of gut viral biomarkers in non-invasive diagnostic tests of colorectal cancer and adenoma. [ABSTRACT FROM AUTHOR]

Objective: We aimed to explore immune-related prognosis genes of lung adenocarcinoma (LUAD). Materials and methods: TCGA-LUAD and GSE31210 data sets were accessed from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) respectively. By using "WGCNA" R package, we established a gene co-expression network and clustered genes into various modules. The correlation between immune scores and module eigengenes by using Pearson analysis. Screened hub genes and constructed prognostic model by using LASSO and Cox regression analysis. Evaluated model by survival analysis and receiver operating characteristic (ROC) curves. Hub genes expression in clinical tissues of LUAD patients by qRT-PCR analysis. ssGSEA and TIMER (a website tool for examination of different immune cells in different cancers) analyzed immune correlation of hub genes. Gene set variation analysis (GSVA) uncovered difference of signal pathway between high- and low-risk score group. Results: We found that brown module significantly correlated with the immune scores of immune cells. Therefore, we constructed a 7-gene prognostic model based on brown module genes, and indicated that this model possessed good predictive performance. Patients in training and validation sets were stratified into the high- and low-risk group using this model. Also, hub genes CDCP1, PLSCR1 and CD79A were highly expressed in clinical tissues of LUAD patients, while ID1, CLEC7A, KIAA1324 and CMTM7 were lowly expressed. Both ssGSEA and TIMER revealed a significant negative correlation between risk score and B cell infiltration. Additionally, some signal pathways were suppressed in the high-risk group. Conclusion: We identified 7 immune-associated prognostic markers, which may play vital roles in LUAD and could be used as hopeful targets for immunotherapy of LUAD. [ABSTRACT FROM AUTHOR]

As a disaster-bearing body, coal mines are vulnerable to the threats of flood disasters and are prone to serious impacts of floods. Floods will not only directly lead to flooding accidents and industrial square destruction, but also indirectly induce abnormal water gushing underground, water permeability, groundwater pollution and other chain disasters. In order to quantitatively assess the risk of coal mine flood accidents caused by flood disasters and strengthen the prevention and control of coal mine flood disasters. Based on the classic theory of three elements of disaster risk science: disaster-causing factors, disaster-pregnant environment, and disaster-bearing body, and the current situation of coal mine flood disaster prevention and control, the investigation index system of flood disaster prevention in coal mine is constructed, and the risk assessment model for flood disaster-induced coal mine water disaster accidents was established, and the application was verified in a coal mine in Tongzi County of Zunyi City. The research shows that the risk assessment of flood-induced coal mine water disasters is a comprehensive assessment of the system risk based on the assessment of the risk of disaster-causing factors, the stability or sensitivity of disaster-pregnant environment and the vulnerability of disaster-bearing body. The time distribution, spatial distribution and intensity of disaster-causing factors are considered in the hazard evaluation of disaster-causing factors. The stability evaluation of disaster-pregnant environment is to evaluate the sensitivity of disaster-pregnant environment to disaster-inducing factors by considering terrain, water system and other factors. The vulnerability of hazard-affected body is determined by two factors and one is the level of flood disaster prevention. The higher the level of prevention is, the lower the vulnerability of hazard-affected body is. The other is the number of people and assets exposed to the disaster-causing factors in the disaster-pregnant environment. And the greater the number is, the higher the vulnerability of the disaster-affected body is. Through the application of the proposed risk assessment model in this Coal Mine, it can be concluded that the risk level of flood-induced coal mine water disasters and accidents in this Coal Mine is low risk. In this paper, the establishment of risk assessment model provides a method for quantitatively evaluating the risk of coal mine water disaster caused by floods. It can provide a new reference for the prevention and control of flood disasters in coal mines.

In the present study, an appropriate risk assessment and allocation model was developed between the various elements of PPP based infrastructure projects. For this purpose, after a review of the research literature and identification of the most important risks, with a case study on PPP-based freeway projects in Iran, a systematic model for risk assessment using fuzzy MCDM methods including FAHP has developed in order to critical risk assessment (Objective 1), FTOPSIS Identification of Common Risks in PPP Transportation Projects (Objective 2) and FDEMATEL for Allocation and Impact of Risk Between Different Project Elements (Objective 3). The results of risk analysis by FAHP method showed that economic and financing, construction and operational risks have the greatest impact, respectively. In addition, the results of FTOPSIS method showed that projects Isfahan-Shiraz Freeway, Salafchegan-Arak Freeway and Khorramabad-Arak Freeway, respectively, were more affected by risks. Finally, the results of FDEMATEL analysis show the degree of interaction and impact strength of risks for each risk and from the perspective of each project manager, which indicates a significant relationship between different risks. The results of this research can be used as a management model in the process of evaluation and risk management of PPP projects. [ABSTRACT FROM AUTHOR]

The aim of this research was to conduct a risk assessment and management of psychological, structural, social and economic determinants (PSSED) in a suicide attempt. The sample consisted of 353 individuals who had a recorded history of suicidal attempt; and 20 professional individuals by purposive sampling method within a descriptive cross-sectional design. Worksheets for RAM and AHP were used for data collection in this study. The rate of suicide attempt was 7.21 per 100,000 population in this study. Analysis showed that depression and mental disorders; personality disorders; family problems; socio-cultural and economic problems; lack of awareness; and low level of education have a high level of risk for suicide attempts. Psychiatric and psychological services; awareness and knowledge of life skills; medical services to dysfunctional families; development of community-based planning for PSSED of suicide; and employment and entrepreneurship services may lower suicide attempt risk. [ABSTRACT FROM AUTHOR]

IntroductionVenous thromboembolism (VTE) risk assessment at admission is of great importance for early screening and timely prophylaxis and management during hospitalization. The purpose of this study is to develop and validate novel risk assessment models at admission based on machine learning (ML) methods.MethodsIn this retrospective study, a total of 3078 individuals were included with their Caprini variables within 24 hours at admission. Then several ML models were built, including logistic regression (LR), random forest (RF), and extreme gradient boosting (XGB). The prediction performance of ML models and the Caprini risk score (CRS) was then validated and compared through a series of evaluation metrics.ResultsThe values of AUROC and AUPRC were 0.798 and 0.303 for LR, 0.804 and 0.360 for RF, and 0.796 and 0.352 for XGB, respectively, which outperformed CRS significantly (0.714 and 0.180, P

Machine learning methods can establish complex nonlinear relationships between input and response variables for stadium fire risk assessment. However, the output of machine learning models is considered very difficult due to their complex "black box" structure, which hinders their application in stadium fire risk assessment. The SHapley Additive exPlanations (SHAP) method makes a local approximation to the predictions of any regression or classification model so as to be faithful and interpretable, and assigns significant values (SHAP value) to each input variable for a given prediction. In this study, we designed an indicator attribute threshold interval to classify and quantify different fire risk category data, and then used a random forest model combined with SHAP strategy in order to establish a stadium fire risk assessment model. The main objective is to analyze the impact analysis of each risk characteristic on four different risk assessment models, so as to find the complex nonlinear relationship between risk characteristics and stadium fire risk. This helps managers to be able to make appropriate fire safety management and smart decisions before an incident occurs and in a targeted manner to reduce the incidence of fires. The experimental results show that the established interpretable random forest model provides 83% accuracy, 86% precision, and 85% recall for the stadium fire risk test dataset. The study also shows that the low level of data makes it difficult to identify the range of decision boundaries for Critical mode and Hazardous mode. [ABSTRACT FROM AUTHOR]

Several DVT risk assessment tools have been developed and validated to stratify the patients. The 2005 and 2010 Caprini risk assessment models (RAMs) proved reliable in detecting deep vein thrombosis (DVT) and are widely used worldwide. Although the 2010 version was recently modified, the 2005 version is still more exact and popular in the surgical population. This study aimed to compare the predictive power of the 2005 and 2010 Caprini RAMs in Chinese orthopedic patients at admission and describe the incidence of DVT during hospitalization. This prospective cohort study involved 2733 patients from the Arthritis Clinic and Research Center, Orthopedics and Traumatology Department, Orthopedic Oncology Department, and Spinal Surgery Department of Peking University People's Hospital from September 2018 to January 2019. The DVT risk scores and levels were determined for all patients using the 2005 and 2010 Caprini RAMs at admission. The sensitivity and specificity of the two tools were calculated. The patients were divided into the screened and unscreened groups according to whether or not a Doppler ultrasound (DUS) was performed. Depending upon the DUS results, the patients were divided into the DVT and non-DVT groups. The receiver operation characteristic (ROC) curve and the area under the curve (AUC) were evaluated to compare the predictive power. Eighty-four patients had DVT in the four orthopedic departments, and the incidence of DVT was 3.1 %–17.2 %. The scores between the 2005 and 2010 Caprini RAMs differed (P < 0.001) among orthopedic patients at admission. Among DVT patients, the highest-risk patients accounted for 63.5 % and 52.9 % based on the 2005 and 2010 Caprini RAMs, respectively. The AUCs of the 2005 and 1020 Caprini RAMs were 0.669 ± 0.030 and 0.648 ± 0.032, respectively. There were no significant differences between the two ROC curves (P = 0.164). The 2005 and 2010 Caprini RAMs showed a similar predictive power in detecting DVT, but neither met the high sensitivity criteria for screening DVT in orthopedic patients. The incidence of DVT in orthopedic patients was relatively high during hospitalization. [Display omitted] [ABSTRACT FROM AUTHOR]

Copyright of Coal Science & Technology (0253-2336) is the property of Coal Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Objective: To analyze the characteristics of peripheral blood lymphocyte subsets in systemic lupus erythematosus (SLE) patients with infection and non-infection group. Explore the risk factors of infection in SLE patients and establish a risk matrix model to predict the occurrence of co-infection. Methods: total of 333 SLE patients without infection, 163 patients suffering from infection, and 132 healthy controls (HCs) were recruited. General clinical data and disease activity indicators were collected. The levels of total T, B, CD4+T, CD8+T, NK, Th1, Th2, Th17, and Treg cells in peripheral blood of HCs, SLE patients (including infected and non-infected group) were analyzed by flow cytometry. The risk assessment model was constructed, and the receiver operating characteristic curve was drawn. 39 SLE patients with infection and 20 patients without infection were randomly selected to evaluate the predictive power of the regression model. Results: The levels of T, B, CD4+T, CD8+T, and NK cells in the infected patients were significantly decreased when compared with that of both non-infected patients and HCs (p <.05). The non-infected patients had a higher level of Th17 than that of HCs (p <. 05), but the absolute numbers of Th17 in infected patients was the lowest among the three groups (p <.001). The number of Treg cells in SLE patients was significantly lower than that of HCs (p <.01), and the infected patients had the fewest Treg cells among all these groups (p <. 05). A risk assessment model for SLE with infection was established, p = 1/(1−e−y), Y = 1.763-0.004 × Absolute number of CD4 + T cells-0.005 × Absolute number of NK cells −0.005 × Platelet count(×1012/L) + 1.033 × Absolute number of lymphocytes (×109/L) + 0.023 × C-reactive protein (mg/dL), whose predictive sensitivity is 77.5%, and specificity is 78.3%. Conclusion: The new risk assessment model exhibits good predictive ability to assess co-infection risk in SLE patients. T cells, NK cells, and CD4 + T cells along with other parameters help in differentiating Lupus with infection from Lupus alone. [ABSTRACT FROM AUTHOR]

Purpose: To study the incidence, risk factors for developing asymptomatic venous thromboembolism and the compliance of patients on anticoagulants for asymptomatic venous thromboembolism (VTE) in nononcological urological medium-high risk inpatients, and build a risk assessment model (RAM) for early screening for asymptomatic VTE. Materials and Methods: We conducted a retrospective analysis of 573 inpatients admitted to a nononcological urological ward of a tertiary hospital in China from January 1, 2017, to June 30, 2019. Data were collected using the electronic medical record system, and patients underwent a follow-up by phone 6 months after discharge. Results: Among the 573 medium-high risk inpatients, 73 (15.4%) were diagnosed with VTE, including 20 (4.2%) symptomatic and 53 (11.2%) asymptomatic. Prior history of VTE, a history of anticoagulants or antiplatelet agents before admission, and D-dimer = 1 were the potential risk factors identified for asymptomatic VTE. Patients with poor awareness of VTE and its dangers, and patients who lived more than 1 hour away from the hospital had a high probability of poor compliance with anticoagulation therapy after discharge. Using D-dimer (1.785 µg/ml), we built a RAM for the early diagnosis of asymptomatic VTE. Conclusion: We found that patients with urinary nontumor VTE had low compliance with anticoagulation therapy after discharge. The key factors for determining asymptomatic VTE in nononcological urological inpatients included prior history of VTE, a history of taking anticoagulants or anti-platelet agents before admission, and D-dimer = 1. Furthermore, we found that the threshold of D-dimer should be elevated to 1.785 µg/ml to predict asymptomatic VTE. [ABSTRACT FROM AUTHOR]