Joint Model Research Articles

Pulmonary herpes simplex virus and cytomegalovirus in patients with acute respiratory distress syndrome related to COVID-19

PurposeHuman herpesviruses, particularly cytomegalovirus (CMV) and herpes simplex virus (HSV), frequently reactivate in critically ill patients, including those with acute respiratory distress syndrome (ARDS) related to coronavirus disease 2019 (COVID-19). The clinical interpretation of pulmonary herpesvirus reactivation is challenging and there is ongoing debate about its association with mortality and benefit of antiviral medication. We aimed to quantify the incidence and pathogenicity of pulmonary CMV and HSV reactivations in critically ill COVID-19 patients.MethodsMechanically ventilated COVID-19 patients seropositive for CMV or HSV were included in this observational cohort study. Diagnostic bronchoscopy with bronchoalveolar lavage was performed routinely and analyzed for alveolar viral loads and inflammatory biomarkers. Utilizing joint modeling, we explored the dynamic association between viral load trajectories over time and mortality. We explored alveolar inflammatory biomarker dynamics between reactivated and non-reactivated patients.ResultsPulmonary reactivation (> 104 copies/ml) of CMV occurred in 6% of CMV-seropositive patients (9/156), and pulmonary reactivation of HSV in 37% of HSV-seropositive patients (63/172). HSV viral load dynamics prior to or without antiviral treatment were associated with increased 90-day mortality (hazard ratio [HR] 1.24, 95% confidence interval [CI] 1.04–1.47). The alveolar concentration of several inflammatory biomarkers increased with HSV reactivation, including interleukin (IL)-6, IL-1β, granulocyte colony stimulating factor (G-CSF), and tumor necrosis factor (TNF).ConclusionIn mechanically ventilated COVID-19 patients, HSV reactivations are common, while CMV reactivations were rare. HSV viral load dynamics prior to or without antiviral treatment are associated with mortality. Alveolar inflammation is elevated after HSV reactivation.

Spatial rotational behaviour of intact and damaged column foot joints: numerical modelling

Columns in traditional Chinese timber structures barely rest on stone bases, and the column foot joints are capable of resisting moment around any direction. This study numerically investigated the spatial rotational behaviour of intact and damaged columns based on experiments. Unidirectional cyclic loading tests were carried out on two intact and three damaged column foot joint specimens. A fibre element-based numerical model for the column foot joints was developed and validated by use of the unidirectional loading test results. Numerical analyses were performed based on the fibre element-based model to obtain the rotational behaviour of the intact and damaged column foot joints under spatial loading. The analyses indicated that the moment–rotation curve of an intact column foot joint under spatial loading was on an umbrella-shaped surface. The damage at the column foot not only unidirectionally decreased the moment-resisting capacity of a column foot joint but also resulted in a rotational performance degradation valley on the moment–rotation surface of the joint. The rotational behaviour of both the intact and damaged column foot joints under spatial loading was highly non-linear. This developed fibre element-based model can be further utilised to analyse the structural performance of traditional timber structures under three-dimensional excitations.

The Configuration Path of the New Rural Collective Economy to Promote the Common Prosperity of Farmers: Based on Qualitative Comparative Analysis Method and the Rural Cases in Zhejiang, China

The new rural collective economy is an important external mechanism for promoting the common prosperity of Chinese farmers. At the same time, the livelihood capital of farmers provides an essential internal support. Achieving an effective match between the two elements is a significant research issue. This article, based on the survey data from 1024 rural households in 43 villages in Zhejiang Province, China, defines the economic functions, social functions, management functions, and cultural functions of the new rural collective economy. The study employs the qualitative comparative analysis method to explore the role of the new rural collective economy in the promotion of the common prosperity of rural households. The necessity analysis shows that the single-factor condition is not the necessary condition for the common prosperity of farmers. However, the adequacy analysis reveals that the linkage and match between the new rural collective economy and the farmers’ livelihood capital can create multiple equivalent pathways for the farmers’ common prosperity. These pathways include the economic function-driven model of the new rural collective economy, the driven model of the high-level livelihoods combined economic functions, the joint model of social function and management functions, the natural capital-driven model, and the joint model of human capital and social capital. Based on these findings, this article proposes targeted governance strategies, including creating pillar industries, strengthening public management services, expanding the scope of social services, and building a coordination mechanism between the new type of rural collective economy and farmers’ livelihood capital.

Joint modeling of human cortical structure: Genetic correlation network and composite-trait genetic correlation

Heritability and genetic covariance/correlation quantify the marginal and shared genetic effects across traits. They offer insights on the genetic architecture of complex traits and diseases. To explore how genetic variations contribute to brain function variations, we estimated heritability and genetic correlation across cortical thickness, surface area, and volume of 33 anatomically predefined regions in left and right hemispheres, using summary statistics of genome-wide association analyses of 31,968 participants in the UK Biobank. To characterize the relationships between these regions of interest, we constructed a genetic network for these regions using recursive two-way cut-offs in similarity matrices defined by genetic correlations. The inferred genetic network matches the brain lobe mapping more closely than the network inferred from phenotypic similarities. We further studied the associations between the genetic network for brain regions and 30 complex traits through a novel composite-linkage disequilibrium score regression method. We identified seven significant pairs, which offer insights on the genetic basis for regions of interest mediated by cortical measures.

A joint model of longitudinal pharmacokinetic and time-to-event data to study exposure–response relationships: a proof-of-concept study with alectinib

PurposeIn exposure–response analyses of oral targeted anticancer agents, longitudinal plasma trough concentrations are often aggregated into a single value even though plasma trough concentrations can vary over time due to dose adaptations, for example. The aim of this study was to compare joint models to conventional exposure–response analyses methods with the application of alectinib as proof-of-concept.MethodsJoint models combine longitudinal pharmacokinetic data and progression-free survival data to infer the dependency and association between the two datatypes. The results from the best joint model and the standard and time-dependent cox proportional hazards models were compared. To normalize the data, alectinib trough concentrations were normalized using a sigmoidal transformation to transformed trough concentrations (TTC) before entering the models.ResultsNo statistically significant exposure–response relationship was observed in the different Cox models. In contrast, the joint model with the current value of TTC in combination with the average TTC over time did show an exposure–response relationship for alectinib. A one unit increase in the average TTC corresponded to an 11% reduction in progression (HR, 0.891; 95% confidence interval, 0.805–0.988).ConclusionJoint models are able to give insights in the association structure between plasma trough concentrations and survival outcomes that would otherwise not be possible using Cox models. Therefore, joint models should be used more often in exposure–response analyses of oral targeted anticancer agents.

Effect of the Sauvé–Kapandji method on the wrist contact surface for distal radial ulnar joint disorders

BackgroundThe Sauvé–Kapandji (S-K) method is a surgical procedure performed for chronic deformities of the distal radial ulnar joint (DRUJ). Changes to the joint contact surface from pre- to postoperatively under physiological in vivo conditions have not yet been determined for this useful treatment. The aim of the present study was therefore to compare the articular contact area of the wrist joint between before and after the S-K method for DRUJ disorders.MethodsThe SK method was performed for 15 patients with DRUJ osteoarthritis and ulnar impaction syndrome. We calculated the Mayo Wrist Score as the patient’s clinical findings and created 3-dimensional bone models of cases in which the S-K method was performed and calculated the contact area and shift in the center of the contact area using customized software.ResultsThe Mean modified Mayo Wrist Score improved significantly from 60.3 preoperatively to 80.3 postoperatively (P < 0.01). Scaphoid contact area to the radius increased significantly from 112.6 ± 37.0 mm2 preoperatively to 127.5 ± 27.8 mm2 postoperatively (P = 0.03). Lunate contact area to radius-ulna was 121.3 ± 43.3 mm2 preoperatively and 112.5 ± 37.6 mm2 postoperatively, but this decrease was not significant (P = 0.38). Contact area ratio of scaphoid to lunate increased significantly from 1.01 ± 0.4 preoperatively to 1.20 ± 0.3 postoperatively (P = 0.02). Postoperative translations of the center of the scaphoid and lunate contact areas were decomposed into ulnar and proximal directions. Ulnar and proximal translation distances of the scaphoid contact area were 0.8 ± 1.7 mm and 0.4 ± 0.6 mm, respectively, and those of the lunate contact area were 1.1 ± 1.7 mm and 0.4 ± 1.1 mm, respectively. This study revealed changes in wrist contact area and center of the contact area before and after the S-K method.ConclusionThese results may accurately indicate changes in wrist joint contact area from pre- to postoperatively using the S-K method for patients with DRUJ disorder. Evaluation of changes in contact area due to bone surface modeling of the wrist joint using 3DCT images may be useful in considering surgical methods.

An analytical model of longitudinal joints in the segmental lining of shield tunnels with considering geometric nonlinearity

The previous mechanical analysis on the longitudinal joint of segmental linings commonly used the plane section assumption, in which the strains on the normal section are linearly and continuously distributed. This was proven to be overidealized according to a series of full-scale tests in existing literatures. In this paper, the nonlinear characteristics of deformation behavior for longitudinal joint were investigated with emphasis on the geometric nonlinearity of joint. A new analytical model of longitudinal joint was proposed with considering the geometric nonlinearity via employing the local plane section assumption to describe the strain distribution of joint section. The effectiveness of the proposed model was validated and parametric study (e.g. loading cycles and peak bending moment) was conducted. The results indicated the geometric nonlinearity of longitudinal joint refers to the increase area of free surface for the joint section. This nonlinearity will cause the discontinuous change of contact state which can be intuitively charactered by the height of concrete compression zone. Without considering the geometric nonlinearity, the area of concrete compression zone and bending capacity of the joint are overestimated by 45% to 100% based on the calculation result of the verification case. The increasing load cycles and peak load value deteriorate the deformation recoverability and ultimate capacity of joints. The degree of geometric nonlinearity is intensified with the increase of load cycles and peak load value but it can be alleviated if the concrete in external edge contact during overloading process.

Joint risk-based maintenance strategy and integrated monitoring approach for manufacturing system operational process considering workpiece quality

The fluctuation of various parameters in the production process increases the risk of generating unqualified products and suffering from systematic failures and downtimes, and the preventive maintenance and control of potentially faulty machines in manufacturing systems should consider the continuous increase of operational risk. Therefore, to proactive control the manufacturing system operational risk, a joint model of risk-based maintenance (RBM) and integrated monitoring is proposed. The proposed model formulates an appropriate maintenance strategy and appropriate control charts for monitoring the manufacturing process, guaranteeing the stable operation of the manufacturing system. First, considering the co-effect of the production equipment state and workpiece quality, a novel connotation of the manufacturing system mechanism’s operational risk is proposed, and the RBM framework is constructed. Second, considering the manufacturing system operational state, an integrated maintenance decision-making model based on different risk scenarios is established using a dynamic monitoring scheme based on EWMA control charts. Third, a particle swarm optimisation algorithm is adopted to optimise the proposed model to obtain the optimal maintenance strategy for production equipment and monitoring parameter for workpiece quality. Finally, the effectiveness of the proposed approach is illustrated with a case study on a subway current receiver production line.

R2C-GAN: Restore-to-Classify Generative Adversarial Networks for blind X-ray restoration and COVID-19 classification

Restoration of poor-quality medical images with a blended set of artifacts plays a vital role in a reliable diagnosis. As a pioneer study in blind X-ray restoration, we propose a joint model for generic image restoration and classification: Restore-to-Classify Generative Adversarial Networks (R2C-GANs). This is the first generic restoration approach forming an Image-to-Image translation task from poor-quality having noisy, blurry, or over/under-exposed images to high-quality image domain where forward and inverse transformations are learned using unpaired training samples. Simultaneously, the joint classification preserves the diagnostic-related label during restoration. Each R2C-GAN is equipped with operational layers/neurons in a compact architecture. The proposed joint model successfully restores images while achieving state-of-the-art Coronavirus Disease 2019 (COVID-19) classification with above 90% in F1-Score. In qualitative analysis, the restoration performance is confirmed by medical doctors where 68% of the restored images are selected against the original images. We share the software implementation at https://github.com/meteahishali/R2C-GAN.

Studying the association Between longitudinal non-dense breast tissue and breast cancer risk: a joint modelling approach.

Conflicting results have appeared in the literature on whether the amount of non-dense, adipose tissue in the breast is a risk factor or a protective factor for breast cancer, and biological hypotheses supporting both have been proposed. We suggest here that limitations in study design and statistical methodology could potentially explain the inconsistent results. Specifically, we exploit recent advances in methodology and software developed for the joint analysis of multiple longitudinal outcomes and time-to-event data to jointly analyze dense and non-dense tissue trajectories and the risk of breast cancer in a large, Swedish, screening cohort. We also perform extensive sensitivity analyses by mimicking analyses/designs of previously published studies, e.g. ignoring available longitudinal data. Overall, we did not find strong evidence supporting an association between non-dense tissue and the risk of incident breast cancer. We hypothesize that (1) previous studies have not been able to isolate the effect of non-dense tissue from dense tissue or adipose tissue elsewhere in the body, that (2) estimates of the effect of non-dense tissue on risk are strongly sensitive to modeling assumptions, or that (3) the effect size of non-dense tissue on breast cancer risk is likely to be small/not clinically relevant.

A 3D-printed, dynamic, patient-specific knee simulator

Purpose3D-printed devices proved their efficacy across different clinical applications, helping personalize medical treatments. This paper aims to present the procedure for the design and production of patient-specific dynamic simulators of the human knee. The scope of these simulators is to improve surgical outcomes, investigate the motion and load response of the human knee and standardize in-vitro experiments for testing orthopedic devices through a personalized physical representation of the patient’s joint.Design/methodology/approachThis paper tested the approach on three volunteers. For each, a patient-specific mathematical joint model was defined from an magnetic resonance imaging (MRI) of the knee. The model guided the CAD design of the simulators, which was then realized through stereolithography printing. Manufacturing accuracy was tested by quantifying the differences between 3D-printed and CAD geometry. To assess the simulator functionality, its motion was measured through a stereophotogrammetric system and compared with the natural tibio-femoral motion of the volunteers, measured as a sequence of static MRI.FindingsThe 3D-printing accuracy was very high, with average differences between ideal and printed parts below ± 0.1 mm. However, the assembly of different 3D-printed parts resulted in a higher average error of 0.97 mm and peak values of 2.33 mm. Despite that, the rotational and translational accuracy of the simulator was about 5° and 4 mm, respectively.Originality/valueAlthough improvements in the production process are needed, the proposed simulators successfully replicated the individual articular behavior. The proposed approach is general and thus extendible to other articulations.

Joint models reveal genetic architecture of pubertal stage transitions and their association with BMI in admixed Chilean population.

Early or late pubertal onset can lead to disease in adulthood, including cancer, obesity, type 2 diabetes, metabolic disorders, bone fractures, and psychopathologies. Thus, knowing the age at which puberty is attained is crucial as it can serve as a risk factor for future diseases. Pubertal development is divided into five stages of sexual maturation in boys and girls according to the standardized Tanner scale. We performed genome-wide association studies (GWAS) on the "Growth and Obesity Chilean Cohort Study" cohort composed of admixed children with mainly European and Native American ancestry. Using joint models that integrate time-to-event data with longitudinal trajectories of body mass index (BMI), we identified genetic variants associated with phenotypic transitions between pairs of Tanner stages. We identified $42$ novel significant associations, most of them in boys. The GWAS on Tanner $3\rightarrow 4$ transition in boys captured an association peak around the growth-related genes LARS2 and LIMD1 genes, the former of which causes ovarian dysfunction when mutated. The associated variants are expression and splicing Quantitative Trait Loci regulating gene expression and alternative splicing in multiple tissues. Further, higher individual Native American genetic ancestry proportions predicted a significantly earlier puberty onset in boys but not in girls. Finally, the joint models identified a longitudinal BMI parameter significantly associated with several Tanner stages' transitions, confirming the association of BMI with pubertal timing.

STRESS-DEFORMED STATE OF THE ACROMIOCLAVICULAR JOINT IN CASE OF DAMAGE TO THE UPPER ACROMIOCLAVICULAR LIGAMENT SUPERIOR AND VARIOUS METHODS OF FIXATION

During the operative stabilization of the acromial end of the clavicle (AEC) for its dislocation, there are two ways: clavicle – beaklike process, clavicle – acromial process. Fixation of the AEC to the acromial process of the scapula is a priority. Competing metal structures are the hook plate and the Weber method. The significant disadvantages of using the Weber method of fixation are the migration of the tips and the violation of their integrity andthe wire. Objective. To conduct a comparative analysis of the fixation of the acromial end of the clavicle according to Weber, hook plate and the proposed construction, by studying the stressed-deformed state of the clavicular-acromial joint in case of damage to the lig. acromioclavicular superior and various methods of fixation. Methods. A finite-element model of the clavicular-acromial joint was constructed. Damage to the ligaments was modeled. claviculo-acoacromiale superior, as well as fixation of AEC in three ways: according to Weber, hook plate and the proposed construction. Results. The best results, from the point of view of reducing the level of stresses in intact ligaments, are provided by fixation of the AEC according to Weber, but its use leadsto an increase in the level of stresses on the AEC and the acromial process of the scapula, which can cause destruction of the latter. The hinge-type fixator provides the best stress distribution, both in the bony elements of the model and in intact ligaments. The hook plate holder occupies an average position,both in terms of the level of stress and the magnitude of the relative deformations in the ligaments. Conclusions. Weber fixation provides the best results for reducing the level of stresses and relative strains in the intact ligaments, but leads to a several-fold increase in the level of stresses on the AEC and acromial process of the scapula. The hook plate holder occupies an average position, both in terms of the stress level and the magnitude of the relativedeformations in the ligaments. The proposed design provides the best stress distribution, both in the bony elements of the model and in the intact ligaments.

Ergonomic assessment and gait analysis of a knee joint model with an active spring-reinforced centrally-rollable knee bypass support system &amp;nbsp;

Ergonomic assessment and gait analysis of a knee joint model with an active spring-reinforced centrally-rollable knee bypass support system &amp;nbsp;

Position-aware Interactive Attention Network for multi-intent spoken language understanding

Spoken Language Understanding (SLU) is a crucial component of task-oriented dialog systems. In recent years, there has been increasing attention on multi-intent SLU due to its relevance to complex real-world applications. Most existing joint models only utilize multi-intent information to guide the slot filling, with only a small number of models achieving bidirectional interaction. Additionally, unlike traditional single-intent SLU, multi-intent SLU is scope-dependent, where each intent in a sentence has its specific dependency range. However, current bidirectional joint models often employ a pipeline approach to implement interaction between the two sub-tasks, which fails to fully leverage the semantic information between them and can lead to error propagation. Moreover, attention-based multi-intent joint models do not adequately model the positional relationships between words, resulting in suboptimal overall performance. In this paper, we propose a novel multi-intent SLU model called Position-aware Interactive Attention Network (PIAN), which consists of interactive attention and rotary position embedding. The aim of PIAN is to fully exploit the correlation between the two sub-tasks and capturing the positional dependencies between words. This facilitates mutual guidance and enhancement between the two sub-tasks. Additionally, to address the issue of uncoordinated slot problem generated by traditional slot filling decoders, we employ an MCRF slot filling decoder to constrain the slot labels. We evaluate our model on two public multi-intent SLU datasets, and the experimental results demonstrate that our model achieves state-of-the-art performance on key metrics. Code for this paper is publicly available at https://github.com/puhahahahahaha/SLU_with_Co_PRoE.

Diabetic retinopathy incidence, predictors and its association with longitudinal fasting blood sugar level changes among diabetes mellitus patients in Ethiopia: joint model.

Diabetes mellitus (DM) is a global public health problem characterized by an elevated blood glucose level. Monitoring blood sugar levels is vital for effective diabetes management and preventing complications. However, the association between longitudinal biomarkers and the incidence of diabetic complications is often overlooked. Therefore, this study aimed to assess the incidence of diabetic retinopathy, predictors, and association with longitudinal fasting blood sugar level changes among diabetes mellitus patients in Ethiopia. A multicenter retrospective follow-up study was carried out in referral hospitals in Amhara region, Ethiopia. A random sample of 462 newly diagnosed DM patients was selected. The proportional hazard assumption was checked for the survival sub-model, and for the longitudinal sub-model, the normality assumption was checked. Then the joint modeling with time-dependent lagged parameterizations was fitted. Model assumptions and comparisons were checked. Finally, the hazard ratio with a 95% confidence interval (CI) with a corresponding P-value<0.05 was used to identify predictors. In this study, Overall, 54 patients developed DR, and the incidence rate was 2.33 per 1000 person-months over the follow-up period, with a 95% CI of [1.78, 3.05]. Rural residence (AHR = 2.21, 95% CI: [1.21, 4.05]), hypertension co-morbidity (AHR = 3.01, 95% CI: [1.85, 6.53]), and longer duration of DM (>5 years) (AHR = 2.28, 95% CI: [1.91, 5.15]) were important predictors for the incidence of DR. In addition, the incidence of DR was substantially correlated with the time-dependent lagged value of FBS change (AHR = 4.20, 95% CI [1.62, 10.85]). In this study, the incidence of diabetic retinopathy was somewhat high when compared to prior similar studies in Ethiopia. A joint model of longitudinal fasting blood sugar level changes was significantly associated with an increased risk of DR. Besides, being rural residence, hypertension co-morbidity, and a longer duration of DM were significant predictors for the incidence of DR. Therefore, public awareness, an integrated care approach, and prioritizing glycemic control are highly recommended.

A quantitative modeling framework to understand the physiology of the hypothalamic-pituitary-adrenal axis and interaction with cortisol replacement therapy.

Congenital adrenal hyperplasia (CAH) is characterized by impaired adrenal cortisol production. Hydrocortisone (synthetic cortisol) is the drug-of-choice for cortisol replacement therapy, aiming to mimic physiological cortisol circadian rhythm. The hypothalamic-pituitary-adrenal (HPA) axis controls cortisol production through the pituitary adrenocorticotropic hormone (ACTH) and feedback mechanisms. The aim of this study was to quantify key mechanisms involved in the HPA axis activity regulation and their interaction with hydrocortisone therapy. Data from 30 healthy volunteers was leveraged: Endogenous ACTH and cortisol concentrations without any intervention as well as cortisol concentrations measured after dexamethasone suppression and single dose administration of (i) 0.5-10mg hydrocortisone as granules, (ii) 20mg hydrocortisone as granules and intravenous bolus. A stepwise model development workflow was used: A newly developed model for endogenous ACTH and cortisol was merged with a refined hydrocortisone pharmacokinetic model. The joint model was used to simulate ACTH and cortisol trajectories in CAH patients with varying degrees of enzyme deficiency, with or without hydrocortisone administration, and healthy individuals. Time-dependent ACTH-driven endogenous cortisol production and cortisol-mediated feedback inhibition of ACTH secretion processes were quantified and implemented in the model. Comparison of simulated ACTH and cortisol trajectories between CAH patients and healthy individuals showed the importance of administering hydrocortisone before morning ACTH secretion peak time to suppress ACTH overproduction observed in untreated CAH patients. The developed framework allowed to gain insights on the physiological mechanisms of the HPA axis regulation, its perturbations in CAH and interaction with hydrocortisone administration, paving the way towards cortisol replacement therapy optimization.

Classifying Neuronal Cell Types Based on Shared Electrophysiological Information from Humans and Mice.

The brain is an intricate system that controls a variety of functions. It consists of a vast number of cells that exhibit diverse characteristics. To understand brain function in health and disease, it is crucial to classify neurons accurately. Recent advancements in machine learning have provided a way to classify neurons based on their electrophysiological activity. This paper presents a deep-learning framework that classifies neurons solely on this basis. The framework uses data from the Allen Cell Types database, which contains a survey of biological features derived from single-cell recordings from mice and humans. The shared information from both sources is used to classify neurons into their broad types with the help of a joint model. An accurate domain-adaptive model, integrating electrophysiological data from both mice and humans, is implemented. Furthermore, data from mouse neurons, which also includes labels of transgenic mouse lines, is further classified into subtypes using an interpretable neural network model. The framework provides state-of-the-art results in terms of accuracy and precision while also providing explanations for the predictions.

The value of D-dimer and platelet-lymphocyte ratio combined with CT signs for predicting intestinal ischemia in patients with bowel obstruction.

To investigate the diagnostic value of D-dimer, platelet-lymphocyte rate (PLR) and CT signs for intestinal ischemia in patients with bowel obstruction. We retrospectively analyzed the clinical and imaging data of 105 patients diagnosed with bowel obstruction, and performed univariate and multivariate analyses to determine the independent risk factors for intestinal ischemia in patients with bowel obstruction. Moreover, the receiver operating characteristic curve (ROC) was plotted to examine the diagnostic value of D-dimer, PLR and CT signs in patients with bowel obstruction. Besides, Kappa tests were used to assess inter-observer agreement. We included 56 men (53%) and 49 women (47%) with mean age of 66.05 ± 16 years. Univariate and multivariate analyses showed that D-dimer, PLR and two significant CT signs (i.e., increased unenhanced bowel-wall attenuation and mesenteric haziness) were independent risk factors for intestinal ischemia in patients with bowel obstruction. ROC analysis showed that the combined use of D-dimer, PLR and the said two CT signs had better performance than single indicators in predicting intestinal ischemia in patients with bowel obstruction. The area under the curve (AUC) of the joint model III was 0.925 [95%CI: 0.876-0.975], with a sensitivity of 79.2% [95CI%: 67.2-91.1] and a specificity of 91.2% [95%CI: 83.7-98.9]. The combined use of D-dimer, PLR and CT signs has high diagnostic value for intestinal ischemia in patients with bowel obstruction and will prompt surgical exploration to evaluate intestinal blood flow.

Analyzing speed-difference impact on freeway joint injury severities of Leading-Following vehicles using statistical and data-driven models

Rear-end (RE) crashes are notably prevalent and pose a substantial risk on freeways. This paper explores the correlation between speed difference among the following and leading vehicles (Δν) and RE crash risk. Three joint models, comprising both uncorrelated and correlated joint random-parameters bivariate probit (RPBP) approaches (statistical methods) and a cross-stitch multilayer perceptron (CS-MLP) network (a data-driven method), were estimated and compared against three separate models: Support Vector Machines (SVM), eXtreme Gradient Boosting (XGBoost), and MLP networks (all data-driven methods). Data on 15,980 two-vehicle RE crashes were collected over a two-year period, from January 1, 2021, to December 31, 2022, considering two possible levels of injury severity: no injury and injury/fatality for both drivers of following and leading vehicles. The comparative performance analysis demonstrates the superior predictive capability of the CS-MLP network over the uncorrelated/correlated joint RPBP model, SVM, XGBoost, and MLP networks in terms of recall, F-1 Score, and AUC. Significantly, numerous shared variables influence the injury severity outcomes for the following and leading vehicles across both statistical and data-driven approaches. Among these factors, the following vehicle (a truck) and the leading vehicle (a passenger car) demonstrate contrasting effects on the injury severity outcomes for both vehicles. Furthermore, the SHapley Additive exPlanations (SHAP) values from the CS-MLP network visually show the relationship between Δν and injury severity, revealing non-linear trends unlike the average effects shown by statistical methods. They indicate that the least injury outcomes for both following and leading vehicles occurs at a Δν of 0 to 10 mph, matching observed patterns in RE crash data. Additionally, a marked variation in the trend of SHAP values for the two vehicles is noted as the speed difference increases. Therefore, the findings affirm the superior performance of joint model development and substantiate the non-linear impacts of speed difference on injury outcomes. The adoption of dynamic speed control measures is recommended to mitigate the injury outcomes involved in two-vehicle RE crashes.