Search Results (45)

Conventional academic warning systems in higher education often rely on end-of-semester grades, which severely limits opportunities for timely intervention. To address this, our interdisciplinary study developed and validated a comprehensive socio-technical framework that integrates social-cognitive theory with learning analytics. The framework combines educational data mining with culturally responsive, personalized interventions tailored to a non-Western context. A two-phase mixed-methods design was employed: first, predictive models were built using Learning Management System (LMS) data from 2,856 students across 64 courses. Second, a quasi-experimental trial (n = 48) was conducted to evaluate intervention efficacy. Historical academic performance, attendance, and assignment submission patterns were the strongest predictors, achieving a Balanced Area Under the Curve (AUC) of 0.85. The intervention, specifically adapted to Confucian educational values, yielded remarkable results: 73% of at-risk students achieved passing grades, with a large effect size for academic improvement (Cohen’s d = 0.91). These findings empirically validate a complete prediction–intervention–evaluation cycle, demonstrating how algorithmic predictions can be effectively integrated with culturally informed human support networks. This study advances socio-technical systems theory in education by bridging computer science, psychology, and educational research. It offers an actionable model for designing ethical and effective early warning systems that balance technological innovation with human-centered pedagogical values. Full article

Coastal communities and marine ecosystems face increasing risks due to changing ocean conditions, yet effective wave monitoring remains limited in many low-resource regions. This study investigates the use of seismic data to predict significant wave height (SWH), offering a low-cost and scalable solution to support coastal conservation and safety. We developed a baseline machine learning (ML) model and improved it using a longest-stretch algorithm for seismic data selection and station-specific hyperparameter tuning. Models were trained and tested on consumer-grade hardware to ensure accessibility and availability. Applied to the Sicily–Malta region, the enhanced models achieved up to a 0.133 increase in R² and a 0.026 m reduction in mean absolute error compared to existing baselines. These results demonstrate that seismic signals, typically collected for geophysical purposes, can be repurposed to support ocean monitoring using accessible artificial intelligence (AI) tools. The approach may be integrated into conservation planning efforts such as early warning systems and ecosystem monitoring frameworks. Future work may focus on improving robustness in data-sparse areas through augmentation techniques and exploring broader applications of this method in marine and coastal sustainability contexts. Full article

Pedestrian intrusion in coal yard work areas is a major cause of accidents, posing challenges for the safe supervision of coal yards. Existing visual detection methods suffer under poor lighting and a lack of 3D data. To overcome these limitations, this study introduces a robust pedestrian intrusion detection method based on 3D LiDAR. Our approach consists of three main components. First, we propose a novel pedestrian detection network called EFT-RCNN. Based on Voxel-RCNN, this network introduces an EnhancedVFE module to improve spatial feature extraction, employs FocalConv to reconstruct the 3D backbone network for enhanced foreground–background distinction, and utilizes TeBEVPooling to optimize bird’s eye view (BEV) generation. Second, a precise 3D hazardous area is defined by combining a polygonal base surface, determined through on-site exploration, with height constraints. Finally, a point–region hierarchical judgment method is designed to calculate the spatial relationship between pedestrians and the hazardous area for graded warning. When evaluated on the public KITTI dataset, the EFT-RCNN network improved the average precision for pedestrian detection by 4.39% in 3D and 4.68% in BEV compared with the baseline, while maintaining a real-time processing speed of 28.56 FPS. In practical tests, the pedestrian detection accuracy reached 92.9%, with an average error in distance measurement of 0.054 m. The experimental results demonstrate that the proposed method effectively mitigates complex environmental interference, enables robust detection, and provides a reliable means for the proactive prevention of pedestrian intrusion accidents. Full article

Background and Objectives: Early ambulation and timely detection of postoperative complications are cornerstones of colorectal Enhanced Recovery After Surgery (ERAS) programmes, yet the traditional bedside checks performed every 4–8 h may miss clinically relevant deterioration. The consumer wearables boom has spawned a new generation of wrist- or waistband-mounted sensors that stream step count, heart-rate and temperature data continuously, creating an opportunity for data-driven early-warning strategies. No previous systematic review has focused exclusively on colorectal surgery. Methods: Three databases (PubMed, Embase, and Scopus) were searched (inception—1 May 2025) for prospective or retrospective studies that used a consumer-grade or medical-grade wearable to collect objective physical-activity or vital-sign data during the peri-operative period of elective colorectal resection. Primary outcomes were postoperative complication rates, length-of-stay (LOS) and 30-day readmission. Two reviewers screened records, extracted data and performed risk-of-bias appraisals with ROBINS-I or RoB 2. Narrative synthesis was adopted because of the heterogeneity in devices, recording windows and outcome definitions. Results: Nine studies (n = 778 patients) met eligibility: one randomised controlled trial (RCT), seven prospective cohort studies and one retrospective analysis. Five studies relied on step-count metrics alone; four combined step-count with heart-rate or skin-temperature streams. Median wear time was 6 d (range 2–30). Higher day-1 step count (≥1000 steps) was associated with shorter LOS (odds ratio 0.63; 95% CI 0.45–0.84). Smart-band–augmented ERAS pathways shortened protocol-defined LOS by 1.1 d. Pre-operative inactivity (<5000 steps·day⁻¹) and low “return-to-baseline” activity on the day before discharge independently predicted any complication (OR 0.39) and 30-day readmission (OR 0.60 per 10% increment). A prospective 101-patient study that paired pedometer-recorded ambulation with daily lung-ultrasound scores found fewer pulmonary complications when patients walked further (Spearman r = –0.36, p < 0.05). Conclusions: Continuous, patient-worn sensors are feasible and yield clinically meaningful data after colorectal surgery. Early postoperative step-count trajectories and activity-derived recovery indices correlate with LOS, complications and readmission, supporting their incorporation into digital ERAS dashboards. Standardised outcome definitions, open algorithms for signal processing and multicentre validation are now required. Full article

Foreign object intrusion in railway perimeter areas poses significant risks to train operation safety. To address the limitation of current visual detection technologies that overly focus on target identification while lacking quantitative risk assessment, this paper proposes a railway intrusion risk quantification method integrating track semantic segmentation and spatiotemporal features. An improved BiSeNetV2 network is employed to accurately extract track regions, while physical-constrained risk zones are constructed based on railway structure gauge standards. The lateral spatial distance of intruding objects is precisely calculated using track gauge prior knowledge. A lightweight detection architecture is designed, adopting ShuffleNetV2 as the backbone to reduce computational complexity, with an incorporated Dilated Transformer module to enhance global context awareness and sparse feature extraction, significantly improving detection accuracy for small-scale objects. The comprehensive risk assessment formula integrates object category weights, lateral risk coefficients in intrusion zones, longitudinal distance decay factors, and dynamic velocity compensation. Experimental results demonstrate that the proposed method achieves 84.9% mean average precision (mAP) on our proprietary dataset, outperforming baseline models by 3.3%. By combining lateral distance detection with multidimensional risk indicators, the method enables quantitative intrusion risk assessment and graded early warning, providing data-driven decision support for active train protection systems and substantially enhancing intelligent safety protection capabilities. Full article

Grading and providing early warning of traffic congestion density is crucial for the timely coordination and optimization of traffic management. However, current traffic density detection methods primarily rely on historical traffic flow data, resulting in ambiguous thresholds for congestion classification. To overcome these challenges, this paper proposes a traffic density grading algorithm for remote sensing images that integrates adaptive clustering and multi-scale fusion. A dynamic neighborhood radius adjustment mechanism guided by spatial distribution characteristics is introduced to ensure consistency between the density clustering parameter space and the decision domain for image cropping, thereby addressing the issues of large errors and low efficiency in existing cropping techniques. Furthermore, a hierarchical detection framework is developed by incorporating a dynamic background suppression strategy to fuse multi-scale spatiotemporal features, thereby enhancing the detection accuracy of small objects in remote sensing imagery. Additionally, we propose a novel method that combines density analysis with pixel-level gradient quantification to construct a traffic state evaluation model featuring a dual optimization strategy. This enables precise detection and grading of traffic congestion areas while maintaining low computational overhead. Experimental results demonstrate that the proposed approach achieves average precision (AP) scores of 32.6% on the VisDrone dataset and 16.2% on the UAVDT dataset. Full article

With the in-depth application of digital transformation in the oil industry, data-driven methods provide a new technical path for drilling engineering safety evaluation. In this paper, a data-driven drilling safety level evaluation system is proposed. By integrating the three-dimensional visualization technology of wellbore trajectory and the prediction model of friction torque, a dynamic and intelligent drilling risk evaluation framework is constructed. The Python platform is used to integrate geomechanical parameters, real-time drilling data, and historical working condition records, and the machine learning algorithm is used to train the friction torque prediction model to improve prediction accuracy. Based on the K-means clustering evaluation method, a three-tier drilling safety classification standard is established: Grade I (low risk) for friction (0–100 kN) and torque (0–10 kN·m), Grade II (medium risk) for friction (100–200 kN) and torque (10–20 kN·m), and Grade III (high risk) for friction (>200 kN) and torque (>20 kN·m). This enables intelligent quantitative evaluation of drilling difficulty. The system not only dynamically optimizes bottom-hole assembly (BHA) and drilling parameters but also continuously refines the evaluation model’s accuracy through a data backtracking mechanism. This provides a reliable theoretical foundation and technical support for risk early warning, parameter optimization, and intelligent decision-making in drilling engineering. Full article

Background/Objectives: Hospitalized peripartum patients who later decompensate and require an upgrade to the intensive care unit (ICU) may have an increased risk for poor outcomes. Most of the literature regarding the need for ICU involves Modified Early Warning Scores in already hospitalized patients or the evaluation of specific comorbid conditions or diagnoses. This systematic review and meta-analysis aimed to assess the differences in clinical scores at admission among adult peripartum patients to identify the later need for ICU. Methods: We systematically searched Ovid-Medline, PubMed, EMBASE, Web of Science and Google Scholar for randomized and observational studies of adult patients ≥18 years of age who were ≥20 weeks pregnant or up to 40 days post-partum, were admitted to the wards from the emergency department and later required critical care services. The primary outcome was the Sequential Organ Failure Assessment (SOFA) score. Secondary outcomes included other clinical scores, the hospital length of stay (HLOS) and mortality. The Newcastle–Ottawa Scale was utilized to grade quality. Descriptive analyses were performed to report demographic data, with means (±standard deviation [SD]) for continuous data and percentages for categorical data. Random-effects meta-analyses were performed for all outcomes when at least two studies reported a common outcome. Results: Seven studies met the criteria, with a total of 1813 peripartum patients. The mean age was 27.2 (±2.36). Patients with ICU upgrades were associated with larger differences in mean SOFA scores. The pooled difference in means was 2.76 (95% CI 1.07–4.46, p < 0.001). There were statistically significant increases in Sepsis in Obstetrics Scores, APACHE II scores, and HLOS in ICU upgrade patients. There was a non-significantly increased risk of mortality in ICU upgrade patients. There was high overall heterogeneity between patient characteristics and management in our included studies. Conclusions: This systematic review and meta-analysis demonstrated higher SOFA or other physiologic scores in ICU upgrade patients compared to those who remained on the wards. ICU upgrade patients were also associated with a longer HLOS and higher mortality compared with control patients. Full article

Short-term precipitation forecasting is a core task in meteorological science, aiming to achieve accurate predictions by modeling the spatiotemporal evolution of radar echo sequences, thereby supporting meteorological services and disaster warning systems. However, existing spatiotemporal sequence prediction methods still struggle to disentangle complex spatiotemporal dependencies effectively and fail to capture the nonlinear chaotic characteristics of precipitation systems. This often results in ambiguous predictions, attenuation of echo intensity, and spatial localization errors. To address these challenges, this paper proposes a unified spatiotemporal sequence prediction framework based on spatiotemporal masking, which comprises two stages: self-supervised pre-training and task-oriented fine-tuning. During pre-training, the model learns global structural features of meteorological systems from sparse contexts by randomly masking local spatiotemporal regions of radar images. In the fine-tuning stage, considering the importance of the temporal dimension in short-term precipitation forecasting and the complex long-range dependencies in spatiotemporal evolution of precipitation systems, we design an RNN-based cyclic temporal mask self-encoder model (MAE-RNN) and a transformer-based spatiotemporal attention model (STMT). The former focuses on capturing short-term temporal dynamics, while the latter simultaneously models long-range dependencies across space and time via a self-attention mechanism, thereby avoiding the smoothing effects on high-frequency details that are typical of conventional convolutional or recurrent structures. The experimental results show that STMT improves 3.73% and 2.39% in CSI and HSS key indexes compared with the existing advanced models, and generates radar echo sequences that are closer to the real data in terms of air mass morphology evolution and reflection intensity grading. Full article

As aging urban expressways become more pronounced, maintenance and construction work on these roadways is increasingly necessary. Some lanes may need to be closed during maintenance and construction, decreasing driving safety and comfort in the work zone. This situation often leads to traffic congestion and a higher risk of traffic accidents. Notably, 80% of work zone traffic accidents occur in the warning and upstream transition areas (or simply warning and transition areas). Therefore, it is crucial to appropriately determine the lengths of these areas to enhance both safety and comfort for drivers. In this study, we examined three different warning lengths (1800 m, 2000 m, and 2200 m) and three transition lengths (120 m, 140 m, and 160 m) using the entropy weighting method to create nine simulation scenarios on a two-way, six-lane urban expressway. We selected various metrics for driving safety and comfort, including drivers’ eye movement, electroencephalogram, and driving behavior indicators. A total of 45 participants (mean age = 23.9 years, standard deviation = 1.8) were recruited for the driving simulation experiment, and each participant completed all 9 simulation scenarios. After eliminating 5 invalid datasets, we obtained valid data from 40 participants. We employed a combination of the analytic network process and entropy weighting method to calculate the comprehensive weights of the eight evaluation indicators. Additionally, we introduced the fuzzy theory, utilizing a trapezoidal membership function to evaluate the membership matrix values of the indicators and the comprehensive evaluation grade eigenvalues. The ranking of the experimental scenarios was determined using these eigenvalues. The results indicated that more extended warning lengths correlated with increased safety and comfort. Specifically, the best driver safety and comfort levels were observed in Scenario I, which featured a 2200 m warning length × 160 m transition length. However, the difference in safety and comfort across different transition lengths diminished as the warning length increased. Therefore, when road space is limited, a thoughtful combination of reasonable lengths can still provide high driving safety and comfort. Full article

At present, there is no research that classifies the epidemic intensity of acute gastroenteritis (AGE) caused by Norovirus (NoV) GII combined cases and environmental surveillance data at the same time. With reference to the experience of the epidemiological-level classification of infectious disease and the actual epidemiological status of NoV AGE in Chaoyang District, Beijing, China, the epidemic intensity of NoV GII was divided into five grades with increasing intensity from grade 1 to grade 5, which corresponds to non-epidemic risk, general risk, moderate risk, high risk, and ultra-high risk, respectively. If the synthetic index of two consecutive monitoring weeks in the epidemic season of 2023–2024 exceeds a certain threshold, an early warning for the corresponding epidemic intensity will be issued and recommendations on the corresponding control measures will be given. This study established and quantified the criteria for the epidemic intensity of AGE caused by NoV GII based on case surveillance data and environmental surveillance data. It provides a reference for other methods to carry out relevant studies in the future. However, mathematical models cannot completely replace skilled experience. Therefore, when making decisions with early warning models in practice, it is necessary to refer to the opinions of professional and experienced experts to avoid the bias of the early warning model from affecting strategy judgment. Full article

Water resources security is crucial to the survival and development of human society. A water resources security assessment and dynamic early warning system was constructed. The weights of water resources evaluation indexes were calculated by the entropy weight method, and the water resources security was evaluated with the comprehensive index method. The obstacle degree model was used to identify and analyze the main obstacle factors. The grey model was adopted to predict the future water resources security situation. The empirical study was carried out in Jinan. The results showed that the grade of water resources security in Jinan from 2008 to 2021 showed a gradually increasing trend. The obstacle factors were mainly concentrated in the pressure subsystem, indicating that the contradiction between supply and demand of water resources was the main problem affecting water resources security, which was accorded with the actual situation. The comprehensive index of water resources security from 2022 to 2026 shows a gradually increasing trend on the whole, and the warning situation develops towards a good trend, indicating that remarkable results in comprehensively building a water-saving society and vigorously promoting water pollution control have been achieved. The measures such as optimizing economic structure, improving water use structure, and improving water use efficiency will promote the further development of water resources security in Jinan. Full article

Coastal areas are biodiversity hotspots, providing essential ecosystem services, yet they are among the most threatened systems, particularly by alien species invasion. The European regulation on invasive alien species (IAS) highlights early detection as a key prerequisite for effective containment or eradication strategies. Traditional monitoring methods are costly and time-consuming, and Citizen Science (CS) may be a promising alternative. We assessed the contribution of the generalist CS project “Wild Coast Adriatic” (WCA) developed on the iNaturalist platform to the detection of alien species (AS) along the Central Adriatic coast. Using WCA, we extracted alien occurrences and explored AS seasonal patterns, geographic origins, dangers (EU regulation), and distributions inside protected areas (Natura 2000 and LTER sites). Between 2020 and 2023, WCA gathered 2194 research-grade observations of 687 species, including 139 records of 50 AS, five of which are of European concern. Asteraceae and Fabaceae (plants) as well as insects and mollusks (fauna) were the most abundant aliens. The observations increased over time, with more records concentrated in autumn and summer. Most AS come from the Americas and occurred outside the protected areas. Our results underline the contribution of CS data for detecting AS in coastal ecosystems, offering a valid support for early warning, monitoring, and management strategies. Full article

The Yangtze River Economic Belt (YREB), spanning 11 provinces and municipalities across China, is of paramount importance due to its high economic development and strategic role in national distribution. However, the YREB, which has experienced rapid economic growth, faces challenges resulting from its previously expansive development model, including regional resource and environmental issues. Consequently, a systematic analysis encompassing socio-economic, ecological, and resource-environmental aspects is vital for a comprehensive and quantitative understanding of the YREB’s overall condition. This study explores resource and environmental carrying capacity (RECC) by constructing an integrated framework that includes remote sensing data, geographic information data and social statistical data, which allows for a precise analysis of RECC dynamics from 2010 to 2020. The findings demonstrate an upward trend in the overall quality of RECC from 2010 to 2020, achieving higher grades over time. However, there is significant spatial heterogeneity, with a notable decrease in RECC levels moving from the eastern to the western regions within the YREB. Moreover, low-level RECC areas situated in the northwest of the YREB, show a trend of moving toward regions of higher altitude from 2010 to 2020 based on analysis using the standard deviation ellipse (SDE) method. When considering to the three major urban agglomerations within the YREB, overall RECC in middle and lower agglomerations is generally stable and on an upward trend while cities in upper reaches exhibit significant variation and fluctuations, highlighting them as areas requiring future focus. Therefore, specific indicators were applied to monitor RECC risk for each of these three agglomerations, respectively, after which optimized strategies could be proposed based on different early warning levels. Ultimately this study allows local authorities to implement timely and effective interventions to mitigate risks and promote sustainable development. Full article

Abstract: The Association of Mutual Funds of India (AMFI), under the direction of the Securities and Exchange Board of India (SEBI), provided open access to various risk parameters with respect to MidCap and SmallCap funds for the first time from February 2024. Our study utilizes AMFI datasets from February 2024 to September 2024 which consisted of 14 variables. Among these, the primary variable identified in grading mutual funds is the stress test parameter, expressed as number of days required to liquidate between 50% and 25% of the portfolio, respectively, on a pro-rata basis under stress conditions as a response variable. The objective of our paper is to build and test various neural network models which can help in predicting stress levels with the highest accuracy and specificity in MidCap and SmallCap mutual funds based on AMFI’s 14 parameters as predictors. The results suggest that the simpler neural network model architectures show higher accuracy. We used Artificial Neural Networks (ANN) over other machine learning methods due to its ability to analyze the impact of dynamic interrelationships among 14 variables on the dependent variable, independent of the statistical distribution of parameters considered. Predicting stress levels with the highest accuracy in MidCap and SmallCap mutual funds will benefit investors by reducing information asymmetry while allocating investments based on their risk tolerance. It will help policy makers in designing controls to protect smaller investors and provide warnings for funds with unusually high risk. Full article