Search Results (171)

Background: Electrocardiograms (EKGs) are essential tools in emergency medicine, often used to evaluate chest pain, dyspnea, and other symptoms suggestive of cardiac dysfunction. Yet, EKGs are not universally administered to all emergency department (ED) patients. Understanding and predicting which patients receive an EKG may offer insights into clinical decision making, resource allocation, and potential disparities in care. This study examines whether integrating structured clinical data with free-text patient narratives can improve prediction of EKG utilization in the ED. Methods: We conducted a retrospective observational study to predict electrocardiogram (EKG) utilization using data from 13,115 adult emergency department (ED) visits in the nationally representative 2021 National Hospital Ambulatory Medical Care Survey–Emergency Department (NHAMCS-ED), leveraging both structured features—demographics, vital signs, comorbidities, arrival mode, and triage acuity, with the most influential selected via Lasso regression—and unstructured patient narratives transformed into numerical embeddings using Clinical-BERT. Four supervised learning models—Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF) and Extreme Gradient Boosting (XGB)—were trained on three inputs (structured data only, text embeddings only, and a late-fusion combined model); hyperparameters were optimized by grid search with 5-fold cross-validation; performance was evaluated via AUROC, accuracy, sensitivity, specificity and precision; and interpretability was assessed using SHAP values and Permutation Feature Importance. Results: EKGs were administered in 30.6% of adult ED visits. Patients who received EKGs were more likely to be older, White, Medicare-insured, and to present with abnormal vital signs or higher triage severity. Across all models, the combined data approach yielded superior predictive performance. The SVM and LR achieved the highest area under the ROC curve (AUC = 0.860 and 0.861) when using both structured and unstructured data, compared to 0.772 with structured data alone and 0.823 and 0.822 with unstructured data alone. Similar improvements were observed in accuracy, sensitivity, and specificity. Conclusions: Integrating structured clinical data with patient narratives significantly enhances the ability to predict EKG utilization in the emergency department. These findings support a personalized medicine framework by demonstrating how multimodal data integration can enable individualized, real-time decision support in the ED. Full article

Background: Vascular dysfunction is increasingly recognized as a shared contributor to both cognitive impairment and late-life depression (LLD). However, the combined diagnostic value of cerebral hemodynamics, neuroimaging markers, and neuropsychological outcomes remains underexplored. This study aimed to investigate the associations be-tween transcranial Doppler (TCD) ultrasound parameters, cognitive performance, and depressive symptoms in older adults with mild cognitive impairment (MCI) and LLD. Importantly, we evaluated the integrative value of TCD-derived indices alongside MRI-confirmed white matter lesions (WMLs) and standardized neurocognitive and affective assessments. Methods: In this cross-sectional study, 96 older adults were enrolled including 78 cognitively unimpaired individuals and 18 with MCI. All participants underwent structured clinical, neuropsychological, and imaging evaluations including the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Geriatric Depression Scale (GDS-15), MRI-based Fazekas scoring of WMLs, and TCD ultrasonography of the middle cerebral artery. Hemodynamic variables included mean blood flow velocity (MBFV), end-diastolic velocity (EDV), pulsatility index (PI), and resistive index (RI). Logistic regression and receiver operating characteristic (ROC) analyses were used to identify independent predictors of MCI. Results: Participants with MCI showed significantly lower MBFV and EDV, and higher PI and RI (p < 0.05 for all) compared with cognitively unimpaired participants. In multivariate analysis, lower MBFV (OR = 0.64, p = 0.02) and EDV (OR = 0.70, p = 0.03), and higher PI (OR = 3.2, p < 0.01) and RI (OR = 1.9, p < 0.01) remained independently associated with MCI. ROC analysis revealed excellent discriminative performance for RI (AUC = 0.919) and MBFV (AUC = 0.879). Furthermore, PI correlated positively with depressive symptom severity, while RI was inversely related to the GDS-15 scores. Conclusions: Our findings underscore the diagnostic utility of TCD-derived hemodynamic parameters—particularly RI and MBFV—in identifying early vascular contributions to cognitive and affective dysfunction in older adults. The integration of TCD with MRI-confirmed WML assessment and standardized cognitive/mood measures represents a novel and clinically practical multi-modal approach for neurovascular profiling in aging populations. Full article

The reliability of unmanned aerial vehicle (UAV) energy storage battery systems is critical for ensuring their safe operation and efficient mission execution, and has the potential to significantly advance applications in logistics, monitoring, and emergency response. This paper reviews theoretical and technical advancements in UAV battery reliability, covering definitions and metrics, modeling approaches, state estimation, fault diagnosis, and battery management system (BMS) technologies. Based on international standards, reliability encompasses performance stability, environmental adaptability, and safety redundancy, encompassing metrics such as the capacity retention rate, mean time between failures (MTBF), and thermal runaway warning time. Modeling methods for reliability include mathematical, data-driven, and hybrid models, which are evaluated for accuracy and efficiency under dynamic conditions. State estimation focuses on five key battery parameters and compares neural network, regression, and optimization algorithms in complex flight scenarios. Fault diagnosis involves feature extraction, time-series modeling, and probabilistic inference, with multimodal fusion strategies being proposed for faults like overcharge and thermal runaway. BMS technologies include state monitoring, protection, and optimization, and balancing strategies and the potential of intelligent algorithms are being explored. Challenges in this field include non-unified standards, limited model generalization, and complexity in diagnosing concurrent faults. Future research should prioritize multi-physics-coupled modeling, AI-driven predictive techniques, and cybersecurity to enhance the reliability and intelligence of battery systems in order to support the sustainable development of unmanned systems. Full article

Purpose: Predicting colorectal cancer liver metastasis (CRLM) is essential for prognostic assessment. This study aims to develop and validate an interpretable multimodal machine learning framework based on multiparametric MRI for predicting CRLM, and to enhance the clinical interpretability of the model through SHapley Additive exPlanations (SHAP) analysis and deep learning visualization. Methods: This multicenter retrospective study included 463 patients with pathologically confirmed colorectal cancer from two institutions, divided into training (n = 256), internal testing (n = 111), and external validation (n = 96) sets. Radiomics features were extracted from manually segmented regions on axial T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI). Deep learning features were obtained from a pretrained ResNet101 network using the same MRI inputs. A least absolute shrinkage and selection operator (LASSO) logistic regression classifier was developed for clinical, radiomics, deep learning, and combined models. Model performance was evaluated by AUC, sensitivity, specificity, and F1-score. SHAP was used to assess feature contributions, and Grad-CAM was applied to visualize deep feature attention. Results: The combined model integrating features across the three modalities achieved the highest performance across all datasets, with AUCs of 0.889 (training), 0.838 (internal test), and 0.822 (external validation), outperforming single-modality models. Decision curve analysis (DCA) revealed enhanced clinical net benefit from the integrated model, while calibration curves confirmed its good predictive consistency. SHAP analysis revealed that radiomic features related to T2WI texture (e.g., LargeDependenceLowGrayLevelEmphasis) and clinical biomarkers (e.g., CA19-9) were among the most predictive for CRLM. Grad-CAM visualizations confirmed that the deep learning model focused on tumor regions consistent with radiological interpretation. Conclusions: This study presents a robust and interpretable multiparametric MRI-based model for noninvasively predicting liver metastasis in colorectal cancer patients. By integrating handcrafted radiomics and deep learning features, and enhancing transparency through SHAP and Grad-CAM, the model provides both high predictive performance and clinically meaningful explanations. These findings highlight its potential value as a decision-support tool for individualized risk assessment and treatment planning in the management of colorectal cancer. Full article

Background/Objectives: Genetic testing is important for diagnosing inherited retinal diseases (IRDs), but further evidence is needed on the utility of singleton genetic testing in an Australian cohort. Methods: A consecutive series of individuals with clinically diagnosed IRDs without prior genetic testing underwent commercial panel-based sequencing (Invitae or Blueprint Genetics), clinical assessment, and multimodal imaging. Retinal images were graded using the Human Phenotype Ontology terms. Binary logistic regression was used to evaluate clinical predictors of a positive molecular diagnosis. Results: Among 140 participants (mean age 49 ± 19 years), genetic testing was undertaken, on average, 23 ± 17 years after the initial clinical IRD diagnosis. Of the 60% who received a probable molecular diagnosis, 40% require further phase testing, highlighting the limitations of singleton genetic testing. USH2A, ABCA4, and RPGR were the most common encountered genes; 67% of the probably solved participants had causative genes with targeted experimental treatments in ongoing human clinical trials. Symptom onset before the age of 30 (OR = 3.06 [95% CI: 1.34–7.18]) and a positive IRD family history (OR = 2.87 [95% CI: 1.27–6.78]) were each associated with higher odds of receiving a molecular diagnosis. Diagnostic rates were comparable across retinal imaging phenotypes (atrophy and autofluorescence patterns in widespread IRD, and the extent of dystrophy in macular IRDs). Conclusions: In an Australian IRD population without prior genetic testing, commercial panels yielded higher diagnostic rates in individuals with IRD onset before the age of 30 and those with an IRD family history. Further research is needed to understand the genetic basis of IRDs, especially isolated and late-onset cases, to improve diagnosis and access to emerging therapies. Full article

Accurately estimating the energy consumption of unmanned aerial vehicles (UAVs) in real-world delivery scenarios remains a critical challenge, particularly when UAVs operate in complex urban environments and are coupled with public transportation systems. Most existing models rely on oversimplified assumptions or static mission profiles, limiting their applicability to realistic, scalable drone-based logistics. In this paper, we propose a physically-grounded and scenario-aware energy sizing methodology for UAVs operating as part of a last-mile delivery system integrated with a city’s bus network. The model incorporates detailed physical dynamics—including lift, drag, thrust, and payload variations—and considers real-time mission constraints such as delivery execution windows and infrastructure interactions. To enhance the realism of the energy estimation, we integrate computational fluid dynamics (CFD) simulations that quantify the impact of surrounding structures and moving buses on UAV thrust efficiency. Four mission scenarios of increasing complexity are defined to evaluate the effects of delivery delays, obstacle-induced aerodynamic perturbations, and early return strategies on energy consumption. The methodology is applied to a real-world transport network in Belfort, France, using a graph-based digital twin. Results show that environmental and operational constraints can lead to up to 16% additional energy consumption compared to idealized mission models. The proposed framework provides a robust foundation for UAV battery sizing, mission planning, and sustainable integration of aerial delivery into multimodal urban transport systems. Full article

Small and medium-sized enterprises (SMEs) face growing challenges in optimizing their sustainable supply chains because of fragmented logistics data and changing regulatory requirements. In particular, globally operating manufacturing SMEs often lack suitable tools, resulting in manual data collection and making reliable accounting and benchmarking of transport emissions in lifecycle assessments (LCAs) time-consuming and difficult to scale. This paper introduces a novel hybrid AI-supported knowledge graph (KG) which combines large language models (LLMs) with graph-based optimization to automate industrial supply chain route enrichment, completion, and emissions analysis. The proposed solution automatically resolves transportation gaps through generative AI and programming interfaces to create optimal routes for cost, time, and emission determination. The application merges separate routes into a single multi-modal network which allows users to evaluate sustainability against operational performance. A case study shows the capabilities in simplifying data collection for emissions reporting, therefore reducing manual effort and empowering SMEs to align logistics decisions with Industry 5.0 sustainability goals. Full article

It is well known that there is a significant conflict of interest between high-speed rail (HSR) operators and logistics companies. This study proposes an HSR freight pricing strategy based on a multi-objective optimization framework and a freight mode splitting model based on the Logit model. A utility function was constructed to quantify the comprehensive utility of different modes of transportation by integrating five key influencing factors: economy, speed, convenience, stability, and environmental sustainability. A bi-objective optimization model was developed to balance the cost of the logistics with the benefits of high-speed rail operators to achieve a win–win situation. The model is solved by the TOPSIS method, and its effectiveness is verified by the freight case of the Zhengzhou–Chongqing high-speed railway in China. The results of this study showed that (1) HSR has advantages in medium-distance freight transportation; (2) increasing government subsidies can help improve the market competitiveness of high-speed rail in freight transportation. This research provides theoretical foundations and methodological support for optimizing HSR freight pricing mechanisms and improving multimodal transport efficiency. Full article

Background: This paper proposes a conceptual framework for a multifunctional mobility hub (MMH) that co-locates shared e-mobility services, urban logistics, and charging/storage infrastructure within a single site. Aimed at high-density European cities, the MMH model addresses current gaps in both research and practice, where multimodal mobility services, logistics, and energy are rarely planned in an integrated manner. Methods: A mixed-methods approach was applied, including a systematic literature review (PRISMA), expert interviews, case studies, and a stakeholder workshop, to identify synergies across fleet types and operational domains. Results: The analysis reveals key design principles for MMHs, such as interoperable charging, the functional separation of passenger and freight flows, and modular, scalable infrastructure adapted to urban constraints. Conclusions: The MMH serves as a preliminary concept for planning next-generation mobility stations. It offers qualitative insights for urban planners, operators, and policymakers into how multifunctional hubs may support lower emissions, more efficient operations, and shared infrastructure use. Full article

Due to the recent significant increase in the scale of both domestic and international cargo transportation, the transport sector is becoming an important factor in the country’s economic development. This implies the need to improve all links in the cargo transportation chain. A key role in it is played by logistics centers, which in their activities must meet both state (CO₂ emissions, reduction in road load, increase in transportation safety, etc.) and commercial (cargo transportation in the shortest time and at the lowest cost) requirements. The objective of the paper is freight transportation from China to European countries, reflecting issues of CO₂ emissions, reduction in road load, and increase in transportation safety. Transport operations from the manufacturer to the logistics center are especially important in this chain, since the efficiency of transportation largely depends on the decisions made by its employees. They select the appropriate types of transport (air, sea, rail, and road transport) and routes for a specific situation. In methodology, the analyzed problem can be presented as a dynamic multi-criteria decision model. It is assumed that the decision-maker—the manager responsible for planning transportation operations—is interested in achieving three basic goals: financial goal minimizing total delivery costs from factories to the logistics center, environmental goal minimizing the negative impact of supply chain operations on the environment, and high level of customer service goal minimizing delivery times from factories to the logistics center. The proposed methodology allows one to reduce the total carbon dioxide emission by 1.1 percent and the average duration of cargo transportation by 1.47 percent. On the other hand, the total cost of their delivery increases by 1.25 percent. By combining these, it is possible to create optimal transportation options, effectively use vehicles, reduce air pollution, and increase the quality of customer service. All this would significantly contribute to the country’s socio-economic development. It is proposed to solve this complex problem based on a dynamic multi-criteria model. In this paper, the problem of constructing a schedule of transport operations from factories to a logistics center is considered. The analyzed problem can be presented as a dynamic multi-criteria decision model. Linear programming and the AHP method were used to solve it. Full article

Background: The rapid advancement of radiomics and artificial intelligence (AI) technology has provided novel tools for the diagnosis of esophageal cancer. This study innovatively combines muscle imaging features with conventional esophageal imaging features to construct deep learning diagnostic models. Methods: This retrospective study included 1066 patients undergoing radical esophagectomy. Preoperative computed tomography (CT) images covering esophageal, stomach, and muscle (bilateral iliopsoas and erector spinae) regions were segmented automatically with manual adjustments. Diagnostic models were developed using deep learning (2D and 3D neural networks) and traditional machine learning (11 algorithms with PyRadiomics-derived features). Multimodal features underwent Principal Component Analysis (PCA) for dimension reduction and were fused for final analysis. Results: Comparative analysis of 1066 patients’ CT imaging revealed the muscle-based model outperformed the esophageal plus stomach model in predicting N2 staging (0.63 ± 0.11 vs. 0.52 ± 0.11, p = 0.03). Subsequently, multimodal fusion models were established for predicting pathological subtypes, T staging, and N staging. The logistic regression (LR) fusion model showed optimal performance in predicting pathological subtypes, achieving accuracy (ACC) of 0.919 in the training set and 0.884 in the validation set. For predicting T staging, the support vector machine (SVM) model demonstrated the highest accuracy, with training and validation accuracies of 0.909 and 0.907, respectively. The multilayer perceptron (MLP) fusion model achieved the best performance among all models tested for N staging prediction, although the accuracy remained moderate (ACC = 0.704 in the training set and 0.685 in the validation set), indicating potential for further optimization. Fusion models significantly outperformed single-modality models. Conclusions: Based on CT imaging data from 1066 patients, this study systematically constructed predictive models for pathological subtypes, T staging, and N staging of esophageal cancer. Comparative analysis of models using esophageal, esophageal plus stomach, and muscle modalities demonstrated that muscle imaging features contribute to diagnostic accuracy. Multimodal fusion models consistently showed superior performance. Full article

Musculoskeletal (MSK) disorders remain a major global cause of disability, with diagnostic complexity arising from their heterogeneous presentation and multifactorial pathophysiology. Recent advances across imaging modalities, molecular biomarkers, artificial intelligence applications, and point-of-care technologies are fundamentally reshaping musculoskeletal diagnostics. This review offers a novel synthesis by unifying recent innovations across multiple diagnostic imaging modalities, such as CT, MRI, and ultrasound, with emerging biochemical, genetic, and digital technologies. While existing reviews typically focus on advances within a single modality or for specific MSK conditions, this paper integrates a broad spectrum of developments to highlight how use of multimodal diagnostic strategies in combination can improve disease detection, stratification, and clinical decision-making in real-world settings. Technological developments in imaging, including photon-counting detector computed tomography, quantitative magnetic resonance imaging, and four-dimensional computed tomography, have enhanced the ability to visualize structural and dynamic musculoskeletal abnormalities with greater precision. Molecular imaging and biochemical markers such as CTX-II (C-terminal cross-linked telopeptides of type II collagen) and PINP (procollagen type I N-propeptide) provide early, objective indicators of tissue degeneration and bone turnover, while genetic and epigenetic profiling can elucidate individual patterns of susceptibility. Point-of-care ultrasound and portable diagnostic devices have expanded real-time imaging and functional assessment capabilities across diverse clinical settings. Artificial intelligence and machine learning algorithms now automate image interpretation, predict clinical outcomes, and enhance clinical decision support, complementing conventional clinical evaluations. Wearable sensors and mobile health technologies extend continuous monitoring beyond traditional healthcare environments, generating real-world data critical for dynamic disease management. However, standardization of diagnostic protocols, rigorous validation of novel methodologies, and thoughtful integration of multimodal data remain essential for translating technological advances into improved patient outcomes. Despite these advances, several key limitations constrain widespread clinical adoption. Imaging modalities lack standardized acquisition protocols and reference values, making cross-site comparison and clinical interpretation difficult. AI-driven diagnostic tools often suffer from limited external validation and transparency (“black-box” models), impacting clinicians’ trust and hindering regulatory approval. Molecular markers like CTX-II and PINP, though promising, show variability due to diurnal fluctuations and comorbid conditions, complicating their use in routine monitoring. Integration of multimodal data, especially across imaging, omics, and wearable devices, remains technically and logistically complex, requiring robust data infrastructure and informatics expertise not yet widely available in MSK clinical practice. Furthermore, reimbursement models have not caught up with many of these innovations, limiting access in resource-constrained healthcare settings. As these fields converge, musculoskeletal diagnostics methods are poised to evolve into a more precise, personalized, and patient-centered discipline, driving meaningful improvements in musculoskeletal health worldwide. Full article

This paper examines the feasibility of transitioning road cargo to waterborne transport in the UK, aiming to reduce emissions and alleviate road congestion. Key objectives include (1) developing a modal shift technology to establish freight highways across the UK, (2) designing a small, decarbonized barge vessel concept that complements the logistics framework, and (3) assessing the economic and environmental viability of a multimodal logistics network. Using discrete event simulation (DES), four transportation scenarios were analyzed to evaluate the efficiency and sustainability of integrating coastal and inland waterways into the logistics framework. Results indicate that waterborne transport is more cost-effective and environmentally sustainable than road transport. A sweeping design study was conducted to optimize time, cost, and emissions. This model was applied to a case study, providing insights into optimal pathways for transitioning to waterborne freight by finding the optimized number of TEUs. Consequently, our study identified 96 TEUs as the optimal capacity to initiate barge design, balancing cost, time, and emissions, while 126 TEUs emerged as the best option for scalability. Findings offer critical guidance for supporting the UK’s climate goals and governmental policies by advancing sustainable transportation solutions. Full article

Indonesia’s high logistics costs—approximately 14.6% of its GDP—pose a significant challenge to national economic competitiveness. Key contributing factors include complex geography, fragmented multimodal transport systems and inefficient container terminal operations, particularly concerning the handling of empty containers. This study investigates operational optimization in a container terminal using Agile and Lean principles, without additional investment or infrastructure expansion. It compares throughput before and after optimization, focusing on equipment productivity and reduction in idle time, especially related to equipment and human resources. Field implementation began in 2015, followed by simulation-based validation using system dynamics modeling. The terminal demonstrated a sustained increase in capacity beginning in 2016, eventually exceeding its original design capacity while maintaining acceptable berth and Yard Occupancy Ratios (BOR and YOR). Agile practices improved empty container handling, while Lean methods enhanced berthing process efficiency. The findings confirm that significant reductions in port operational costs, shipping operational costs, voyage turnover time, and logistics costs can be achieved through strategic operational reforms and better resource utilization, rather than through capital-intensive expansion. The study provides a replicable model for improving terminal efficiency in ports facing similar constraints. Full article

Background: The transportation of petroleum products via multimodal logistics systems is a complex process subject to operational inefficiencies and elevated risk exposure. The efficient and resilient transportation of petroleum products increasingly depends on multimodal logistics systems, where operational risks and process inefficiencies can significantly impact safety and performance. This study addresses the research question of how an integrated risk-based and workflow-driven approach can enhance the management of oil products logistics in complex port environments. Methods: A dual methodological framework was applied at the Port of Midia, Romania, combining a probabilistic risk assessment model, quantifying incident probability, infrastructure vulnerability, and exposure, with dynamic business process modeling (BPM) using specialized software. The workflow simulation replicated real-world multimodal oil operations across maritime, rail, road, and inland waterway segments. Results: The analysis identified human error, technical malfunctions, and environmental hazards as key risk factors, with an aggregated major incident probability of 2.39%. BPM simulation highlighted critical bottlenecks in customs processing, inland waterway lock transit, and road tanker dispatch. Process optimizations based on simulation insights achieved a 25% reduction in operational delays. Conclusions: Integrating risk assessment with dynamic workflow modeling provides an effective methodology for improving the resilience, efficiency, and regulatory compliance of multimodal oil logistics operations. This approach offers practical guidance for port operators and contributes to advancing risk-informed logistics management in the petroleum supply chain. Full article