Trajectory data based freeway high-risk events prediction and its influencing factors analyses

2021 ◽  
Vol 154 ◽  
pp. 106085
Author(s):  
Rongjie Yu ◽  
Lei Han ◽  
Hui Zhang
Keyword(s):  
High Risk ◽  
Influencing Factors ◽  
Trajectory Data

scholarly journals Risky Driver Recognition with Class Imbalance Data and Automated Machine Learning Framework

2021 ◽  
Vol 18 (14) ◽  
pp. 7534
Author(s):  
Ke Wang ◽  
Qingwen Xue ◽  
Jian John Lu
Keyword(s):  
Machine Learning ◽  
High Risk ◽  
Loss Function ◽  
Class Imbalance ◽  
Support Vector ◽  
Trajectory Data ◽  
Recognition Model ◽  
Learning Framework ◽  
Sampling Cost ◽  
Automated Machine Learning

Identifying high-risk drivers before an accident happens is necessary for traffic accident control and prevention. Due to the class-imbalance nature of driving data, high-risk samples as the minority class are usually ill-treated by standard classification algorithms. Instead of applying preset sampling or cost-sensitive learning, this paper proposes a novel automated machine learning framework that simultaneously and automatically searches for the optimal sampling, cost-sensitive loss function, and probability calibration to handle class-imbalance problem in recognition of risky drivers. The hyperparameters that control sampling ratio and class weight, along with other hyperparameters, are optimized by Bayesian optimization. To demonstrate the performance of the proposed automated learning framework, we establish a risky driver recognition model as a case study, using video-extracted vehicle trajectory data of 2427 private cars on a German highway. Based on rear-end collision risk evaluation, only 4.29% of all drivers are labeled as risky drivers. The inputs of the recognition model are the discrete Fourier transform coefficients of target vehicle’s longitudinal speed, lateral speed, and the gap between the target vehicle and its preceding vehicle. Among 12 sampling methods, 2 cost-sensitive loss functions, and 2 probability calibration methods, the result of automated machine learning is consistent with manual searching but much more computation-efficient. We find that the combination of Support Vector Machine-based Synthetic Minority Oversampling TEchnique (SVMSMOTE) sampling, cost-sensitive cross-entropy loss function, and isotonic regression can significantly improve the recognition ability and reduce the error of predicted probability.

Health-related quality of life and its influencing factors in individuals with high cardiovascular disease risk: a cross-sectional study based on EQ-5D utility scores in Inner Mongolia, China

2021 ◽  
Author(s):  
Ning Cao ◽  
Yunfeng Xi ◽  
Yumin Gao ◽  
Hailing Li ◽  
Yanchao Liu ◽  
...  
Keyword(s):  
Quality Of Life ◽  
High Risk ◽  
Influencing Factors ◽  
Inner Mongolia ◽  
Health Related Quality ◽  
Tobit Regression ◽  
Cvd Risk ◽  
Related Quality ◽  
Health Related

Abstract Background The high risk of cardiovascular disease (CVD) being associated with impaired Health-related quality of life (HRQoL). However, few studies have assessed the HRQoL of individuals with a high risk of CVD in Inner Mongolia, or even in China. We aimed to assess health-related quality of life (HRQoL) among individuals in Inner Mongolia with a high risk of CVD and its risk factors, to provide a reference to improve HRQoL in individuals with high CVD risk. Methods From 2015 to 2017, residents of six villages or communities in Inner Mongolia, selected using a multi-stage stratified cluster random sampling method, were invited to complete a questionnaire and undergo physical examination and laboratory testing. We selected participants whose predicted 10-year risk for CVD exceeded 10% as those with high CVD risk. HRQoL in individuals with high CVD risk was investigated based on the EuroQol-5 Dimension (EQ-5D) scale. The Chinese utility value integral system was used to calculate EQ-5D utility scores, and the Tobit regression model were used to analyze the influencing factors of HRQoL among individuals with high CVD risk. Results Of 13,359 participants with high CVD risk, 65.63% reported no problems in any of the five dimensions; the most frequently reported difficulty was pain/discomfort. The mean utility score was 1.000 (0.869, 1.000). Tobit regression analysis showed that sex, age, education level, residence area, household income, physical activity, hypertension, and dyslipidemia were influencing factors of HRQoL. Conclusion We found that female sex, older age, living in an urban area, lower education level, lower household income, and lower physical activity levels were associated with reduced HRQoL. People with a high risk of CVD should maintain their blood glucose and lipid levels within the normal range.

scholarly journals Influencing Factors and Prevention of Sepsis or Acute Kidney Injury in 85 Patients with Severe Trauma

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Hong Zhang ◽  
Dan Chen ◽  
Lihua Wang ◽  
Bing Li
Keyword(s):  
Risk Factors ◽  
High Risk ◽  
Influencing Factors ◽  
Severe Trauma ◽  
Apache Ii ◽  
Apache Ii Score ◽  
Trauma Patients ◽  
Risk Of Death ◽  
Disturbance Of Consciousness ◽  
Prevention And Treatment

Severe trauma can cause systemic reactions, leading to massive bleeding, shock, asphyxia, and disturbance of consciousness. At the same time, patients with severe trauma are at high risk of sepsis and acute renal injury. The occurrence of complications will increase the difficulty of clinical treatment, improve the mortality rate, and bring heavy physical and mental burdens and economic pressure to patients and their families. It is of great clinical significance to understand the high risk factors of sepsis and AKI and actively formulate prevention and treatment measures. In this study, the clinical data of 85 patients with severe trauma were analyzed by univariate and multivariate logistic regression to identify the risk factors leading to sepsis or AKI and analyze the prevention and treatment strategies. The results showed that multiple injuries, APACHE II score on admission, SOFA score on admission, and mechanical ventilation were independent influencing factors of sepsis in patients with severe trauma, while hemorrhagic shock, APACHE II score on admission, CRRT, and sepsis were independent influencing factors of AKI in patients with severe trauma. Severe trauma patients complicated with sepsis or AKI will increase the risk of death. In the course of treatment, prevention and intervention should be given as far as possible to reduce the incidence of complications.

Application of Fuzzy Assessment Theory in Tower Crane Safety

2011 ◽  
Vol 383-390 ◽  
pp. 7657-7662
Author(s):  
Bin Li ◽  
Li Lei ◽  
Li Wei
Keyword(s):  
High Risk ◽  
Influencing Factors ◽  
Comprehensive Assessment ◽  
Assessment Model ◽  
Construction Site ◽  
Membership Functions ◽  
Fuzzy Comprehensive Assessment ◽  
Tower Crane ◽  
Fuzzy Assessment

Tower crane belongs to high risk equipment in the construction site. Evaluating safety statement of tower crane is a subject of the technical field. The paper setups a new way to evaluate the safety statement of tower crane by means of fuzzy assessment theory. Membership functions have been established for describing the influencing factors of tower crane safety. Fuzzy comprehensive assessment model for evaluating tower crane safety statement has been established. Finally an example for illustrating the method of using the fuzzy comprehensive assessment to analyzing the tower crane safety statement.

scholarly journals Analysis on influencing factors of viral load in patients with high-risk human papillomavirus

2020 ◽  
Author(s):  
Xuerong Lu ◽  
Tiantian Wang ◽  
Youzhong Zhang ◽  
Yuzhen Liu
Keyword(s):  
Multivariate Analysis ◽  
Human Papillomavirus ◽  
Viral Load ◽  
High Risk ◽  
Influencing Factors ◽  
Cervical Screening ◽  
Cervical Lesion ◽  
Cervical Lesions ◽  
Hpv Types ◽  
Hpv Viral Load

Abstract Background: High-risk human papillomavirus (HR-HPV) load is thought to be influenced by many factors, and relationship between viral load and degree of cervical lesion is controversial. This study was to explore possible influencing factors of HR-HPV viral load in uterine cervix.Methods: 605 women who needed colposcopic evaluation for abnormal cervical screening in Affiliated Hospital of Weifang Medical University, China, between November 2017 and September 2018 were enrolled. Cervical specimens were collected from endo- and ectocervix separately using two different cervical brushes. Hybrid Capture II test was used to measure HR-HPV load. Age, histological severity, number of viral types, area and location of cervical lesions were recorded. The correlation between viral load and influencing factors was analysed using univariate and multivariate analysis.Results: HR-HPV load was positively correlated with age, histological severity, multiple HPV types and area of cervical lesions (P<0.05). Viral load with combination of endo- and ectocervical sampling was significantly higher than simple endocervical sampling (P<0.001). Multivariate analysis showed that age, multiple HPV types and area of cervical lesions were independent factors for HR-HPV load with combination of endo- and ectocervical sampling (P<0.05). However, only age and area of cervical lesions were independent factors for viral load with simple endocervical sampling (P<0.05). No significant association was found between viral load and lesion severity in multivariate analysis (P>0.05).Conclusion: HR-HPV load is influenced by age, histological severity, multiple viral types, area of cervical lesion and sampling methods. Age and area of cervical lesions are independent factors for viral load.

Influencing factors on high-risk sexual behaviors in young people: an ecological perspective

2017 ◽  
Vol 31 (2) ◽  
Author(s):  
Hoda Arabi-Mianrood ◽  
Zeinab Hamzehgardeshi ◽  
Elham Khoori ◽  
Mahmood Moosazadeh ◽  
Zohreh Shahhosseini
Keyword(s):  
High Risk ◽  
Young People ◽  
Risk Behaviors ◽  
Influencing Factors ◽  
Sexual Behaviors ◽  
Unprotected Sex ◽  
Cochrane Library ◽  
Theory Approach ◽  
High Risk Behaviors ◽  
The Impact

Abstract Background In recent years, high-risk sexual behaviors due to their negative consequences both for the individual and society have received more attention than other high-risk behaviors. Objective The aim of this study was to review the influencing factors of high-risk sexual behaviors among young people from an ecological point of view. Methods This review was conducted through searching databases including PubMed, Web of Science, Scopus, Google Scholar and the Cochrane Library with keywords such as sexual risk-taking behavior, high-risk sex, unprotected sex and unsafe sex. The relevant papers published between 1995 and 2016 were extracted. After reviewing the abstract and full text of the articles, 45 papers were used to write this article. Results From an ecological theory approach, factors which influence high-risk sexual behaviors are divided into three categories – the microsystem, the mesosystem and the macrosystem. The microsystem includes factors such as age, gender, race, marital status, place of residence, religion, level of education, personality traits, psychological problems, childhood experiences, body image and coincidence of high-risk behaviors; the mesosystem includes factors such as family structure, peers and sex education; in the macrosystem, the impact of culture and traditions of the society, economic status and the media are presented. Conclusion Given that high-risk sexual behaviors often have multiple causes, it seems that health policymakers must consider multi-dimensional interventions to influence high-risk sexual behaviors based on the ecological approach.

scholarly journals Influencing factors of the neurodevelopment of high-risk infants

2018 ◽  
Vol 31 (3) ◽  
pp. e100034 ◽  
Author(s):  
Yuan Tian ◽  
Chuncao Zhang ◽  
Guangjun Yu ◽  
Xiangying Hu ◽  
Zheng Pu ◽  
...  
Keyword(s):  
High Risk ◽  
Influencing Factors ◽  
Preterm Infants ◽  
Infant Health ◽  
Vital Role ◽  
Postnatal Period ◽  
Full Term ◽  
Developmental Abnormalities ◽  
Neurodevelopmental Abnormalities ◽  
High Risk Infants

BackgroundHigh-risk infants refer to newborns exposed to high-risk factors in the prenatal, natal or postnatal period. High-risk infants are at high risk of developmental retardation, and early identification of developmental abnormalities plays a vital role in improving high-risk infants’ quality of life.AimsTo describe the neurodevelopment of high-risk infants aged less than 1 year old, and to analyse the incidences and influencing factors of neurodevelopmental abnormalities in order to provide a basis for neurodevelopment monitoring and management of high-risk infants.MethodsHigh-risk infants born between January 2016 and December 2016 in the maternity and infant health hospitals of three districts in Shanghai were followed up. The Gesell Developmental Scale was used to assess the neurodevelopmental level at the time of recruitment (0–2 months) and at 9 months. Univariate and multivariate analyses of the influencing factors were conducted.Results484 high-risk infants (male 51%, female 49%) with an average gestation age of 36.5±2.2 weeks were recruited. At the time of recruitment, the average age was 2.1 (0.8) months, and the developmental quotient (DQ) scores of full-term high-risk infants in motor (t=3.542,p=0.001), cognitive (t=3.125,p=0.002), language (t=3.189,p=0.002) and social (t=3.316,p=0.001) areas were higher than those of preterm infants. The incidences of developmental abnormalities of full-term high-risk infants in motor (χ2=9.452,p=0.002), cognitive (χ2=6.258,p=0.012), language (χ2=12.319,p=0.001) and social (χ2=6.811,p=0.009) areas were lower than the preterm infants. At 9 months, there was no difference in the DQ scores and incidences of developmental abnormalities in four areas between full-term and preterm high-risk infants, and the incidence of developmental abnormalities was around 10%.ConclusionThe incidence of neurodevelopmental abnormalities in high-risk infants aged less than 1 year old is high. Preterm birth and parental bad habits are significant factors affecting the neurodevelopment. Monitoring and early interventions help to improve high-risk infants’ neurodevelopment.

scholarly journals Analysis of influencing factors of viral load in patients with high-risk human papillomavirus

2021 ◽  
Vol 18 (1) ◽  
Author(s):  
Xuerong Lu ◽  
Tiantian Wang ◽  
Youzhong Zhang ◽  
Yuzhen Liu
Keyword(s):  
Multivariate Analysis ◽  
Human Papillomavirus ◽  
Viral Load ◽  
High Risk ◽  
Influencing Factors ◽  
Cervical Screening ◽  
Cervical Lesion ◽  
Cervical Lesions ◽  
Hpv Types ◽  
Hpv Viral Load

Abstract Background High-risk human papillomavirus (HR-HPV) load is thought to be influenced by many factors, and the relationship between viral load and the degree of cervical lesion is controversial. This study explored the possible influencing factors of HR-HPV viral load in the uterine cervix. Methods A total of 605 women who needed colposcopic evaluation for abnormal cervical screening at the Affiliated Hospital of Weifang Medical University, China, between November 2017 and September 2018 were enrolled. Cervical specimens were collected from the endo- and ectocervix separately using two different cervical brushes. The hybrid capture II test was used to measure HR-HPV load. Age, histological severity, number of viral types, and area and location of cervical lesions were recorded. The correlations between viral load and influencing factors were analysed using univariate and multivariate analyses. Results HR-HPV load was positively correlated with age, histological severity, multiple HPV types and area of cervical lesions (P < 0.05). Viral load with the combination of endo- and ectocervical sampling was significantly higher than simple endocervical sampling (P < 0.001). Multivariate analysis showed that age, multiple HPV types and area of cervical lesions were independent factors for HR-HPV load with a combination of endo- and ectocervical sampling (P < 0.05). However, only age and area of cervical lesions were independent factors for viral load with simple endocervical sampling (P < 0.05). No significant association was found between viral load and lesion severity in multivariate analysis (P > 0.05). Conclusion HR-HPV load is influenced by age, histological severity, multiple viral types, area of cervical lesion and sampling methods. Age and area of cervical lesions are independent factors for viral load.

Sign in / Sign up

Export Citation Format

Share Document