Analysis of factors influencing the degree of accidental injury of bicycle riders considering data heterogeneity and imbalance.

Bicycle safety has emerged as a pressing concern within the vulnerable transportation community. Numerous studies have been conducted to identify the significant factors that contribute to the severity of cyclist injuries, yet the findings have been subject to uncertainty due to unobserved heterogeneity and class imbalance. This research aims to address these issues by developing a model to examine the impact of key factors on cyclist injury severity, accounting for data heterogeneity and imbalance. To incorporate unobserved heterogeneity, a total of 3,895 bicycle accidents were categorized into three homogeneous sub-accident clusters using Latent Class Cluster Analysis (LCA). Additionally, five over-sampling techniques were employed to mitigate the effects of data imbalance in each accident cluster category. Subsequently, Bayesian Network (BN) structure learning algorithms were utilized to construct 32 BN models after pairing the accident data from the four accident cluster types before and after sampling. The optimal BN models for each accident cluster type provided insights into the key factors associated with cyclist injury severity. The results indicate that the key factors influencing serious cyclist injuries vary heterogeneously across different accident clusters. Female cyclists, adverse weather conditions such as rain and snow, and off-peak periods were identified as key factors in several subclasses of accident clusters. Conversely, factors such as the week of the accident, characteristics of the trafficway, the season, drivers failing to yield to the right-of-way, distracted cyclists, and years of driving experience were found to be key factors in only one subcluster of accident clusters. Additionally, factors such as the time of the crash, gender of the cyclist, and weather conditions exhibit varying levels of heterogeneity across different accident clusters, and in some cases, exhibit opposing effects.

A study on expression recognition based on improved mobilenetV2 network.

This paper proposes an improved strategy for the MobileNetV2 neural network(I-MobileNetV2) in response to problems such as large parameter quantities in existing deep convolutional neural networks and the shortcomings of the lightweight neural network MobileNetV2 such as easy loss of feature information, poor real-time performance, and low accuracy rate in facial emotion recognition tasks. The network inherits the characteristics of MobilenetV2 depthwise separated convolution, signifying a reduction in computational load while maintaining a lightweight profile. It utilizes a reverse fusion mechanism to retain negative features, which makes the information less likely to be lost. The SELU activation function is used to replace the RELU6 activation function to avoid gradient vanishing. Meanwhile, to improve the feature recognition capability, the channel attention mechanism (Squeeze-and-Excitation Networks (SE-Net)) is integrated into the MobilenetV2 network. Experiments conducted on the facial expression datasets FER2013 and CK + showed that the proposed network model achieved facial expression recognition accuracies of 68.62% and 95.96%, improving upon the MobileNetV2 model by 0.72% and 6.14% respectively, and the parameter count decreased by 83.8%. These results empirically verify the effectiveness of the improvements made to the network model.

A novel repair strategy using knotless squeeze anchors for lumbar disc herniation with endplate junction lesions under biportal endoscopic spinal surgery.

BACKGROUND: Annulus fibrosus-endplate (AF-EP) junction lesions are important determinants for lumbar disc herniation (LDH). Utilizing biportal endoscopic spinal surgery (BESS), we introduce a novel repair method using bioabsorbable PushLock anchors with suture fibers to stretch disconnected AF tissues to the vertebral cortex. METHODS: The viewing and working portals are established to excise herniated disc materials causing radiculopathy. Through the working portal, a suture strand is passed through the intact AF tissue near the lesion and retrieved using the Suture Crossing Device. Then, the knotless suture limbs are secured into the cortical bone socket of the vertebral body with a PushLock anchor. CONCLUSION: The procedure is a simple, safe, and feasible knotless suturing technique for the treatment of LDH with AF-EP junction lesions.

Epidemiology of Accidental Injuries at Home and Related Risk Factors for Mortality among Older Adults in South Korea: A Retrospective Cohort Study.

Background and Objectives: Accidental home injuries among older adults are increasing globally, but reporting is limited. This study aims to establish foundational data for program development and policies to prevent accidental injuries at home in older adults by using data on the occurrence of accidental injuries at home and analyzing the risk factors of mortality due to accidental injuries among adults aged 65 years and older. Materials and Methods: This retrospective study used data from the community-based Severe Trauma Survey in South Korea. This study identified general, injury-related, and treatment-related characteristics of older adults who were transported to the emergency department with accidental injuries at home. Single-variable and multiple logistic regression analyses were used to identify risk factors for mortality after injury. Results: The majority of older adults in this study who experienced accidental injuries at home were aged 75 to 84 (42.8%) and female (52.8%), with 1465 injured from falls and slips (68.0%). Risk factors for mortality included older age (≥85 years) (ORs 2.25, 95% CI 1.47-3.45), male sex (ORs 1.60, 95% CI 1.15-2.20), mechanism of injury (falls or slips vs. contact injury, ORs 6.76, 95% CI 3.39-13.47; airway obstruction vs. contact injury, ORs 13.96, 95% CI 6.35-30.71), higher severity (moderate vs. mild, ORs 2.56, 95% CI 1.45-4.54; severe vs. mild, ORs 12.24, 95% CI 6.48-23.12; very severe vs. mild, ORs 67.95, 95% CI 38.86-118.81), and receiving a blood transfusion (ORs 2.14, 95% CI 1.24-3.67). Conclusions: Based on these findings, the home and community environments where older adults live should be inspected and monitored, and in-home accidental injury prevention strategies should be developed tailored to the characteristics of older adults' risk factors and their injury-related characteristics.

Voices from the Ground: Community Perspectives on Preventing Unintentional Child Injuries in Low-Income Settings.

Unintentional injuries significantly contribute to mortality and morbidity among children under five, with higher prevalence in low- and middle-income countries (LMICs). Deprived communities in these regions face increased injury risks, yet there is limited research on child safety tailored to their unique challenges. To address this gap, we conducted focus group discussions in rural Uganda, involving parents, village health workers, community leaders, teachers, and maids. The objective was to understand community perceptions around child safety and determine what culturally and age-appropriate solutions may work to prevent child injuries. Analysis of discussions from ten focus groups revealed five main themes: injury causes, child development and behavior, adult behavior, environmental factors, and potential safety kit components. Common injuries included falls, burns, drowning, and poisoning, often linked to environmental hazards such as unsafe bunk beds and wet floors. Financial constraints and limited space emerged as cross-cutting issues. Participants suggested educational resources, first aid knowledge, and practical devices like solar lamps as potential solutions. The study presents invaluable insights into child safety in rural Ugandan homes, emphasizing the role of community awareness and engagement in designing effective, accessible interventions. It underscores the importance of context-specific strategies to prevent childhood injuries in similar resource-constrained environments.

Crossbow Injuries: Predictors of Mortality.

INTRODUCTION: Crossbow injuries are rare but carry significant morbidity and mortality, and there is limited evidence in the medical literature to guide care. This paper reviews the case reports and case series of crossbow injuries and looks for trends regarding morbidity and mortality based on the type of arrow, anatomic location of injury, and intent of injury. METHODS: Multiple databases were searched for cases of crossbow injuries and data were abstracted into a spreadsheet. Statistics were done in SPSS. RESULTS: 358 manuscripts were returned in the search. After deduplication and removal of nonclinical articles, 101 manuscripts remained. Seventy-one articles describing 90 incidents met the inclusion criteria. The mean age was 36.5 years. There were 10 female and 79 male victims. Fatality was 36% for injuries by field tip arrows and 71% for broadhead arrows, p = .024. Assaults were fatal in 84% of cases, suicides in 29%, and accidental injuries in 17%, p < .001. Mortality was similar for wounds to the head and neck (41%), chest (42%), abdomen (33%), extremities (50%), and multiple regions, p = .618. CONCLUSIONS: Crossbows are potentially lethal weapons sold with fewer restrictions than firearms. Injuries caused by broadhead arrows are more likely to be fatal than injuries from field tip arrows. The anatomic location of injury does not correlate with fatality. More than half of crossbow injuries are due to attempted suicide, with a high case-fatality rate.

Reliability in distinguishing murder from accidental event with current forensic criteria in traumatic head injury.

Discovering a body displaying signs of multiple head trauma requires a thorough examination by the forensic pathologist, and a multidisciplinary approach is recommended. However, determining the manner of death is not always possible. We present a case in which the body of a 60-year-old man was discovered lying face down on the floor of his apartment, partially unclothed, surrounded by blood and vomit, and presenting numerous head injuries. The autopsy concluded that the cause of death was a result of post-traumatic brain injury. Nevertheless, applying current criteria made it challenging to ascertain whether the trauma stemmed from an accidental event or an assault.

Injury Pattern Analysis to Optimize Burn Injury Prevention in the Unhoused Community.

Unhoused patients are an overrepresented group in burn injury, and are a uniquely vulnerable population. Current research focuses on the consequences of homelessness on burn outcomes, with little known about the specific circumstances and behaviors leading to burn injury that may represent specific targets for injury prevention efforts. The burn registry at an urban regional burn center was queried for burn admissions in unhoused adults from 2019 to 2022. Registry data pulled included demographics, urine toxicology, mechanism of injury, and injury subjective history. Subjective injury history was reviewed to determine more specific injury circumstances and activities during which accidental burns occurred. Demographic and mechanistic trends in burn admissions were explored via descriptive statistics. Among 254 admissions for burns from the unhoused community, 58.1% of patients were positive for stimulants on admission. Among accidental injuries (69.7%), common circumstances included preparing food or beverages, cooking or using methamphetamine, smoking cannabis or tobacco, bonfires, and candles. A specific common circumstance was lighting a cigarette while handling accelerants (6.7%). Interventions for stimulant abuse, as well as outreach efforts to educate unhoused patients about situational awareness, safe handling of accelerants, safe smoking practices, and safe cooking practices, may be effective tools in reducing burn admissions in this vulnerable population.

Teen Pregnancy and Risk of Premature Mortality.

Importance: Unintentional injury, suicide, and homicide are leading causes of death among young females. Teen pregnancy may be a marker of adverse life experiences. Objective: To evaluate the risk of premature mortality from 12 years of age onward in association with number of teen pregnancies and age at pregnancy. Design, Setting, and Participants: This population-based cohort study was conducted among all females alive at 12 years of age from April 1, 1991, to March 31, 2021, in Ontario, Canada (the most populous province, which has universal health care and data collection). The study period ended March 31, 2022. Exposures: The main exposure was number of teen pregnancies between 12 and 19 years of age (0, 1, or ≥2). Secondary exposures included how the teen pregnancy ended (birth or miscarriage vs induced abortion) and age at first teen pregnancy. Main Outcomes and Measures: The main outcome was all-cause mortality starting at 12 years of age. Hazard ratios (HRs) were adjusted for year of birth, comorbidities at 9 to 11 years of age, and area-level education, income level, and rurality. Results: Of 2 242 929 teenagers, 163 124 (7.3%) experienced a pregnancy at a median age of 18 years (IQR, 17-19 years). Of those with a teen pregnancy, 60 037 (36.8%) ended in a birth (of which 59 485 [99.1%] were live births), and 106 135 (65.1%) ended in induced abortion. The median age at the end of follow-up was 25 years (IQR, 18-32 years) for those without a teen pregnancy and 31 years (IQR, 25-36 years) for those with a teen pregnancy. There were 6030 deaths (1.9 per 10 000 person-years [95% CI, 1.9-2.0 per 10 000 person-years]) among those without a teen pregnancy, 701 deaths (4.1 per 10 000 person-years [95% CI, 3.8-4.5 per 10 000 person-years]) among those with 1 teen pregnancy, and 345 deaths (6.1 per 10 000 person-years [95% CI, 5.5-6.8 per 10 000 person-years]) among those with 2 or more teen pregnancies; adjusted HRs (AHRs) were 1.51 (95% CI, 1.39-1.63) for those with 1 pregnancy and 2.14 (95% CI, 1.92-2.39) for those with 2 or more pregnancies. Comparing those with vs without a teen pregnancy, the AHR for premature death was 1.25 (95% CI, 1.12-1.40) from noninjury, 2.06 (95% CI, 1.75-2.43) from unintentional injury, and 2.02 (95% CI, 1.54-2.65) from intentional injury. Conclusions and Relevance: In this population-based cohort study of 2.2 million female teenagers, teen pregnancy was associated with future premature mortality. It should be assessed whether supports for female teenagers who experience a pregnancy can enhance the prevention of subsequent premature mortality in young and middle adulthood.

Integrated image and location analysis for wound classification: a deep learning approach.

The global burden of acute and chronic wounds presents a compelling case for enhancing wound classification methods, a vital step in diagnosing and determining optimal treatments. Recognizing this need, we introduce an innovative multi-modal network based on a deep convolutional neural network for categorizing wounds into four categories: diabetic, pressure, surgical, and venous ulcers. Our multi-modal network uses wound images and their corresponding body locations for more precise classification. A unique aspect of our methodology is incorporating a body map system that facilitates accurate wound location tagging, improving upon traditional wound image classification techniques. A distinctive feature of our approach is the integration of models such as VGG16, ResNet152, and EfficientNet within a novel architecture. This architecture includes elements like spatial and channel-wise Squeeze-and-Excitation modules, Axial Attention, and an Adaptive Gated Multi-Layer Perceptron, providing a robust foundation for classification. Our multi-modal network was trained and evaluated on two distinct datasets comprising relevant images and corresponding location information. Notably, our proposed network outperformed traditional methods, reaching an accuracy range of 74.79-100% for Region of Interest (ROI) without location classifications, 73.98-100% for ROI with location classifications, and 78.10-100% for whole image classifications. This marks a significant enhancement over previously reported performance metrics in the literature. Our results indicate the potential of our multi-modal network as an effective decision-support tool for wound image classification, paving the way for its application in various clinical contexts.

Optimizing ensemble U-Net architectures for robust coronary vessel segmentation in angiographic images.

Automated coronary angiography assessment requires precise vessel segmentation, a task complicated by uneven contrast filling and background noise. Our research introduces an ensemble U-Net model, SE-RegUNet, designed to accurately segment coronary vessels using 100 labeled angiographies from angiographic images. SE-RegUNet incorporates RegNet encoders and squeeze-and-excitation blocks to enhance feature extraction. A dual-phase image preprocessing strategy further improves the model's performance, employing unsharp masking and contrast-limited adaptive histogram equalization. Following fivefold cross-validation and Ranger21 optimization, the SE-RegUNet 4GF model emerged as the most effective, evidenced by performance metrics such as a Dice score of 0.72 and an accuracy of 0.97. Its potential for real-world application is highlighted by its ability to process images at 41.6 frames per second. External validation on the DCA1 dataset demonstrated the model's consistent robustness, achieving a Dice score of 0.76 and an accuracy of 0.97. The SE-RegUNet 4GF model's precision in segmenting blood vessels in coronary angiographies showcases its remarkable efficiency and accuracy. However, further development and clinical testing are necessary before it can be routinely implemented in medical practice.

A precise model for skin cancer diagnosis using hybrid U-Net and improved MobileNet-V3 with hyperparameters optimization.

Skin cancer is a frequently occurring and possibly deadly disease that necessitates prompt and precise diagnosis in order to ensure efficacious treatment. This paper introduces an innovative approach for accurately identifying skin cancer by utilizing Convolution Neural Network architecture and optimizing hyperparameters. The proposed approach aims to increase the precision and efficacy of skin cancer recognition and consequently enhance patients' experiences. This investigation aims to tackle various significant challenges in skin cancer recognition, encompassing feature extraction, model architecture design, and optimizing hyperparameters. The proposed model utilizes advanced deep-learning methodologies to extract complex features and patterns from skin cancer images. We enhance the learning procedure of deep learning by integrating Standard U-Net and Improved MobileNet-V3 with optimization techniques, allowing the model to differentiate malignant and benign skin cancers. Also substituted the crossed-entropy loss function of the Mobilenet-v3 mathematical framework with a bias loss function to enhance the accuracy. The model's squeeze and excitation component was replaced with the practical channel attention component to achieve parameter reduction. Integrating cross-layer connections among Mobile modules has been proposed to leverage synthetic features effectively. The dilated convolutions were incorporated into the model to enhance the receptive field. The optimization of hyperparameters is of utmost importance in improving the efficiency of deep learning models. To fine-tune the model's hyperparameter, we employ sophisticated optimization methods such as the Bayesian optimization method using pre-trained CNN architecture MobileNet-V3. The proposed model is compared with existing models, i.e., MobileNet, VGG-16, MobileNet-V2, Resnet-152v2 and VGG-19 on the "HAM-10000 Melanoma Skin Cancer dataset". The empirical findings illustrate that the proposed optimized hybrid MobileNet-V3 model outperforms existing skin cancer detection and segmentation techniques based on high precision of 97.84%, sensitivity of 96.35%, accuracy of 98.86% and specificity of 97.32%. The enhanced performance of this research resulted in timelier and more precise diagnoses, potentially contributing to life-saving outcomes and mitigating healthcare expenditures.

Validity and reliability International Classification of Diseases-10 codes for all forms of injury: A systematic review.

BACKGROUND: Intentional and unintentional injuries are a leading cause of death and disability globally. International Classification of Diseases (ICD), Tenth Revision (ICD-10) codes are used to classify injuries in administrative health data and are widely used for health care planning and delivery, research, and policy. However, a systematic review of their overall validity and reliability has not yet been done. OBJECTIVE: To conduct a systematic review of the validity and reliability of external cause injury ICD-10 codes. METHODS: MEDLINE, EMBASE, COCHRANE, and SCOPUS were searched (inception to April 2023) for validity and/or reliability studies of ICD-10 external cause injury codes in all countries for all ages. We examined all available data for external cause injuries and injuries related to specific body regions. Validity was defined by sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Reliability was defined by inter-rater reliability (IRR), measured by Krippendorff's alpha, Cohen's Kappa, and/or Fleiss' kappa. RESULTS: Twenty-seven published studies from 2006 to 2023 were included. Across all injuries, the mean outcome values and ranges were sensitivity: 61.6% (35.5%-96.0%), specificity: 91.6% (85.8%-100%), PPV: 74.9% (58.6%-96.5%), NPV: 80.2% (44.6%-94.4%), Cohen's kappa: 0.672 (0.480-0.928), Krippendorff's alpha: 0.453, and Fleiss' kappa: 0.630. Poisoning and hand and wrist injuries had higher mean sensitivity (84.4% and 96.0%, respectively), while self-harm and spinal cord injuries were lower (35.5% and 36.4%, respectively). Transport and pedestrian injuries and hand and wrist injuries had high PPVs (96.5% and 92.0%, respectively). Specificity and NPV were generally high, except for abuse (NPV 44.6%). CONCLUSIONS AND SIGNIFICANCE: The validity and reliability of ICD-10 external cause injury codes vary based on the injury types coded and the outcomes examined, and overall, they only perform moderately well. Future work, potentially utilizing artificial intelligence, may improve the validity and reliability of ICD codes used to document injuries.

Recognition of mulberry leaf diseases based on multi-scale residual network fusion SENet.

Silkworms are insects with important economic value, and mulberry leaves are the food of silkworms. The quality and quantity of mulberry leaves have a direct impact on cocooning. Mulberry leaves are often infected with various diseases during the growth process. Because of the subjectivity and time-consuming problems in artificial identification of mulberry leaf diseases. In this work, a multi-scale residual network fusion Squeeze-and-Excitation Networks (SENet) is proposed for mulberry leaf disease recognition. The mulberry leaf disease dataset was expanded by performing operations such as brightness enhancement, contrast enhancement, level flipping and adding Gaussian noise. Multi-scale convolution was used instead of the traditional single-scale convolution, allowing the network to be widened to obtain more feature information and avoiding the overfitting phenomenon caused by the network piling up too deep. SENet was introduced into the residual network to enhance the extraction of key feature information of the model, thus improving the recognition accuracy of the model. The experimental results showed that the method proposed in this paper can effectively improve the recognition performance of the model. The recognition accuracy reached 98.72%. The recall and F1 score were 98.73% and 98.72% respectively. Compared with some other models, this model has better recognition effect and can provide technical reference for intelligent mulberry leaf disease detection.

Evaluation of Risk Factors for the Occurrence of Limb Fractures in Children Due to Unintentional Injury in Podgorica, Montenegro, Mediterranean.

Background and Objectives: Limb injuries in childhood are very common, with most of them being unintentional and often accompanied by soft tissue injuries. The aim of our study was to determine the risk factors that contribute to the occurrence of limb fractures as the most common type of accidental injury to children in our conditions. Materials and Methods: This study was designed as a prospective clinical analysis of predictive factors with a "nested" case-control study. It included all patients under the age of 18 who were diagnosed with unintentional limb injury and limb fracture due to accidental injury, at the Clinical Center of Montenegro, Podgorica, in the period of 7 January 2020-30 June 2021. Results: The gender of the child and the occurrence of the fracture are not related, and a statistically significant relationship was found between the occurrence of the fracture and the place of residence, the child's age, body mass index (BMI), the affected limb, the method of injury, and the mental state of the parents of the injured child, as well as their economic status. It was proved that the older the child was, the lower the chance of injury, while multivariate analysis proved that BMI could be a predictor of accidental fracture. The most common method of accidental limb fractures in children was a fall from a height. Conclusions: The analysis of factors that influence the occurrence of children's injuries is of great importance for public health. Such and similar research can enable a better understanding of the factors that influence accidental injuries, and therefore influence the prevention of these injuries by organizing various educational materials at the primary healthcare level or at the school level, for both children and parents.

A dynamical behavior of the coupled Broer-Kaup-Kupershmidt equation using two efficient analytical techniques.

The aim of the present study is to identify multiple soliton solutions to the nonlinear coupled Broer-Kaup-Kupershmidt (BKK) system, including beta, conformable, local-fractional, and M-truncated derivatives. The coupled Broer-Kaup-Kupershmidt system is employed for modelling nonlinear wave evolution in mathematical models of fluid dynamics, plasmic, optical, dispersive, and nonlinear long-gravity waves. The travelling wave solutions to the above model are found using the Unified and generalised Bernoulli sub-ODE techniques. By modifying certain parameter values, we may create bright soliton, squeezed bell-shaped wave, expanded v-shaped soliton, W-shaped wave, singular soliton, and periodic solutions. The four distinct kinds of derivatives are compared quite effectively using 2D line graphs. Also, contour plots and 3D graphics are given by using Mathematica 10. Lastly, any pair of propagating wave solutions has symmetrical geometrical forms.

Applying the antecedent, behaviour, and consequence taxonomy to unintentional firearm injury as determined by a collaborative firearm injury database.

BACKGROUND: Unintentional firearm injury (UFI) remains a significant problem in the USA with respect to preventable injury and death. The antecedent, behaviour and consequence (ABC) taxonomy has been used by law enforcement agencies to evaluate unintentional firearm discharge. Using an adapted ABC taxonomy, we sought to categorise civilian UFI in our community to identify modifiable behaviours. METHODS: Using a collaborative firearm injury database (containing both a university-based level 1 trauma registry and a metropolitan law enforcement database), all UFIs from August 2008 through December 2021 were identified. Perceived threat (antecedent), behaviour and injured party (consequence) were identified for each incident. RESULTS: During the study period, 937 incidents of UFI were identified with 64.2% of incidents occurring during routine firearm tasks. 30.4% of UFI occurred during neglectful firearm behaviour such as inappropriate storage. Most injuries occurred under situations of low perceived threat. UFI involving children was most often due to inappropriate storage of weapons, while cleaning a firearm was the most common behaviour in adults. Overall, 16.5% of UFI involved injury to persons other than the one handling the weapon and approximately 1.3% of UFI resulted in mortality. CONCLUSIONS: The majority of UFI occurred during routine and expected firearm tasks such as firearm cleaning. Prevention programmes should not overlook these modifiable behaviours in an effort to reduce UFIs, complications and deaths.

Predicting and factor analysis of rider injury severity in two-wheeled motorcycle and vehicle crash accidents based on an interpretable machine learning framework.

OBJECTIVE: As one of the vulnerable road users in accidents, how to improve the two-wheeled motorcyclist's driving safety and reduce accident injury is a public health issue. Accurate identification of the factors influencing the severity of accidents is an important prerequisite for mitigating injury from crashes. METHODS: Based on a vehicle and a two-wheeled motorcycle crash accident data from the China in-depth accident study database (CIDAS), this study uses the performance evaluation indicators of accuracy, precision, recall, F1-score, AUC, and the ROC curve. The classification and prediction performances of the six machine learning methods on the dataset are compared, and the LightGBM algorithm with the best performance is selected to model the accident injury severity of the motorcyclists. The SHAP method is used to extend the interpretability of the LightGBM model results. Based on the SHAP method, the importance, main effect, and the interaction effect of factors under each accident injury severity are quantitatively analyzed. RESULTS: The model prediction accuracy is 92.6%, the F1-Score is 92.8%, and the AUC value is 0.986. The importance of factors varies with the accident injury severity of motorcyclists. The kilometers traveled per year by the driver, the throwing distance of the motorcyclist, and the road speed limit are the three most important factors. The motorcyclist is more likely to suffer fatal injuries when the throwing distance is >1,000 cm. CONCLUSIONS: The prediction model of driver injury severity based on LightGBM algorithm has a good prediction performance. It can be used to analyze the influence factors of injury severity in two-wheeled motorcyclist accident by combining the model with SHAP method. These results could help the traffic management department to take measures to reduce accident injury of motorcyclists.

CAT-Seg: cascaded medical assistive tool integrating residual attention mechanisms and Squeeze-Net for 3D MRI biventricular segmentation.

Cardiac image segmentation is a critical step in the early detection of cardiovascular disease. The segmentation of the biventricular is a prerequisite for evaluating cardiac function in cardiac magnetic resonance imaging (CMRI). In this paper, a cascaded model CAT-Seg is proposed for segmentation of 3D-CMRI volumes. CAT-Seg addresses the problem of biventricular confusion with other regions and localized the region of interest (ROI) to reduce the scope of processing. A modified DeepLabv3+ variant integrating SqueezeNet (SqueezeDeepLabv3+) is proposed as a part of CAT-Seg. SqueezeDeepLabv3+ handles the different shapes of the biventricular through the different cardiac phases, as the biventricular only accounts for small portion of the volume slices. Also, CAT-Seg presents a segmentation approach that integrates attention mechanisms into 3D Residual UNet architecture (3D-ResUNet) called 3D-ARU to improve the segmentation results of the three major structures (left ventricle (LV), Myocardium (Myo), and right ventricle (RV)). The integration of the spatial attention mechanism into ResUNet handles the fuzzy edges of the three structures. The proposed model achieves promising results in training and testing with the Automatic Cardiac Diagnosis Challenge (ACDC 2017) dataset and the external validation using MyoPs. CAT-Seg demonstrates competitive performance with state-of-the-art models. On ACDC 2017, CAT-Seg is able to segment LV, Myo, and RV with an average minimum dice symmetry coefficient (DSC) performance gap of 1.165%, 4.36%, and 3.115% respectively. The average maximum improvement in terms of DSC in segmenting LV, Myo and RV is 4.395%, 6.84% and 7.315% respectively. On MyoPs external validation, CAT-Seg outperformed the state-of-the-art in segmenting LV, Myo, and RV with an average minimum performance gap of 6.13%, 5.44%, and 2.912% respectively.

Prevalence and correlates of unintentional nonfatal injuries among school-going adolescents in Central America.

OBJECTIVES: The aim of this study is to report on the frequency of serious physical injuries (SPI) among adolescents in Central America during the previous decade, 2009-2018. METHODS: In total, 15,807 school adolescents (14.4 years mean age; SD=1.4) from six Central American countries participated in cross-sectional Global School-based Student Health Surveys in 2009-2018 (ranging from 1,779 students in Honduras in 2012 to 4,374 students in Guatemala in 2015). RESULTS: The prevalence of SPI was 33.8 % (22.9 % once, 7.4 % 2 or 3 times and 3.6 % 4 or more times), ranging from 31.8 % in Guatemala to 45.0 % in Belize and 45.6 % in Panama. The most frequent causes of SPI included fall (11.4 %, ranging from 6.9 % in Costa Rica to 15.6 % in Panama), and the type of SPI was fracture/dislocation (5.7 %, ranging from 4.3 % in Costa Rica to 6.7 % in Panama). In adjusted Poisson regression, male sex, food insecurity, a history of alcohol intoxication, soft drink consumption, fast food intake, truancy, multiple sexual partners, psychological distress, physical fight, physically attacked, bullied, and suicide attempt were significantly associated with a higher number of injury event counts. CONCLUSIONS: Overall, about one in three adolescents in Central America had sustained unintentional injuries in the past 12 months and several contributing factors were identified which if addressed could aid injury prevention among adolescents.