Search Results (1,794)

Eye movements are regulated by the ocular motor nerves (cranial nerves [CNs] III, IV, and VI), which control the six extraocular muscles of each eye. Palsies of CNs III, IV, and VI can restrict eye movements, resulting in strabismus and diplopia, and so clinical evaluations of eye movements are crucial for diagnosing CN palsies. This study aimed to develop an accurate artificial intelligence (AI) system for classifying CN III, IV, and VI palsies using nine-gaze ocular photographs. We analyzed 478 nine-gaze photographs comprising 70, 29, and 58 cases of CN III, IV, and VI palsies, respectively. The images were processed using MATLAB. For model training, each photograph of eye movements in the nine directions was numerically coded. A multinetwork model was employed to ensure precise analyses of paralytic strabismus. The AI system operates by referring data on minor abnormalities in the nine-gaze image to a network designed to detect CN IV abnormalities, which re-examines downward and lateral gazes to detect distinctions. Data on major abnormalities are directed to a different network trained to differentiate between CN III and VI abnormalities. EfficientNet-B0 was applied to reduce overfitting and improve learning efficiency in training with limited medical imaging data as the neural network architecture. The diagnostic accuracies of the proposed network for CN III, IV, and VI palsies were 99.31%, 97.7%, and 98.22%, respectively. This study has demonstrated the design of an AI model using a relatively small dataset and a multinetwork training system for analyzing nine-gaze photographs in strabismus patients with CN III, IV, and VI palsies, achieving an overall accuracy of 98.77%. Full article

Physics-Informed Neural Networks (PINNs) have emerged as a promising paradigm for solving partial differential equations (PDEs) by embedding physical laws into the learning process. However, standard PINNs often suffer from training instabilities and unbalanced optimization when handling multi-term loss functions, especially in problems involving singular perturbations, fractional operators, or multi-scale behaviors. To address these limitations, we propose a novel gradient variance weighting physics-informed neural network (GVW-PINN), which adaptively adjusts the loss weights based on the variance of gradient magnitudes during training. This mechanism balances the optimization dynamics across different loss terms, thereby enhancing both convergence stability and solution accuracy. We evaluate GVW-PINN on three representative PDE models and numerical experiments demonstrate that GVW-PINN consistently outperforms the conventional PINN in terms of training efficiency, loss convergence, and predictive accuracy. In particular, GVW-PINN achieves smoother and faster loss reduction, reduces relative errors by one to two orders of magnitude, and exhibits superior generalization to unseen domains. The proposed framework provides a robust and flexible strategy for applying PINNs to a wide range of integer- and fractional-order PDEs, highlighting its potential for advancing data-driven scientific computing in complex physical systems. Full article

Electrochromic windows (ECWs) are promising smart façade technologies that can enhance indoor comfort and reduce energy demands, yet their performance in university classrooms remains underexplored in cold regions. This study evaluates the applicability of ECWs in classrooms in Xi’an, a representative cold-climate city, through dynamic simulations of three classroom types. Three control strategies—based on outdoor temperature, illuminance, and solar radiation—were tested under different thresholds. The results show that compared with static windows, ECWs can increase the annual mean indoor temperature by up to 1.4 °C, extend thermal comfort time ratio by 4.5%, improve visual comfort duration by 6.3%, and reduce heating and cooling demands by 11.6 and 14.3 kWh/m², respectively. These findings demonstrate both the feasibility and the differentiated benefits of ECWs in educational buildings, filling the research gap on their performance across different classroom types and offering practical guidance for sustainable classroom design and operation in cold climates. Full article

Coastal zones are critical areas of marine ecosystems, where biodiversity is a key ecological element for maintaining ecosystem stability and ensuring the sustainability of fishery resources. The Shengsi Ma’an Archipelago Marine Special Reserve features heterogeneous habitats such as rocky reefs, seaweed beds, and artificial aquaculture areas, which are significantly affected by human activities. This study focused on the nearshore waters of Lvhua Island within the reserve. Based on the degree of human disturbance, the study area was divided into five typical habitat types: cage culture area (A), intertidal seaweed bed (B), marine platform area (C), open waters (D), and mussel culture area (E). Environmental DNA (eDNA) technology was employed to analyze the characteristics of eukaryotic community structures across these habitats and their coupling mechanisms with environmental factors. The results showed that a total of 767,360 valid sequences were obtained from 15 seawater samples. Clustering into operational taxonomic units (OTUs) yielded 811 OTUs, taxonomically covering 50 phyla, 104 classes, 220 orders, 334 families, 435 genera, and 530 species. The number of OTUs shared across all habitats was 387. The intertidal seaweed bed (B) had the highest proportion of unique OTUs (4.8%) and showed significant differences (0.01 < p < 0.05) in OTU composition compared to the marine platform area (C) and the mussel culture area (E). Among the major dominant phyla, the abundance of Dinoflagellata across sites was A (74.56%) > E (68.32%) > B (62.15%) > C (58.74%) > D (55.21%). The abundance of Arthropoda across sites was D (27.34%) > C (19.98%) > B (17.89%) > E (9.17%) > A (8.25%). Each of the other sites had 1-2 dominant phyla. Among the major dominant genera, the abundance of an unclassified genus of Dinophyceae was B (41.39%) > C (23.31%) > D (22.03%) > E (19.27%) > A (18.56%). The genus Noctiluca was endemic to Site A, with an abundance of 39.98%. The genus Calanus was dominant in site D (26.17%). The genus Meganyctiphanes was unique to sites C (12.12%) and D (8.76%). The genus Ectopleura was unique to site A. The genus Botrylloides was unique to site E. The remaining genera were evenly distributed across sites without significant habitat specificity. Alpha diversity analysis revealed that the marine platform area (C) had the highest Shannon index (3.32 ± 0.22) and Pielou index (0.54 ± 0.04), while the mussel culture area (E) had the highest Chao1 index (578.96 ± 10.25). All diversity indices were lowest in the cage culture area (A). Principal coordinate analysis (PCoA) and ANOSIM tests indicated significant differences (p < 0.05) in eukaryotic community structures among different habitats. Samples from the seaweed bed clustered separately and were distant from other habitats. Redundancy analysis (RDA) showed that pH was the key environmental factor driving the differentiation of eukaryotic community structure. Temperature was negatively correlated with dissolved oxygen, while salinity was positively correlated with pH. The combined differences in environmental factors were the main drivers of eukaryotic community structure differentiation. In conclusion, this study clarifies the regulatory role of habitat type on the eukaryotic community structure in the nearshore waters of Ma’an Archipelago, confirming a negative correlation between human activity intensity and biodiversity, and a positive correlation between natural habitat complexity and biodiversity. The research findings provide scientific support for assessing the health of the marine ecosystem and formulating ecological conservation and management strategies in this region. Full article

The sufficient capacity of railway lines is a key prerequisite for stable and sustainable transport, not only on main or high-speed lines, but also on lines of regional importance that complement the network. Their indispensable role is manifested not only daily, but especially in the event of incidents on the backbone network. One of the main characteristics of these support lines is that they are largely single-track. Another important characteristic is that they alternate between sections with different traffic loads, which significantly changes the capacity requirements along the whole line. Existing modernisation approaches are frequently implemented in a non-differentiated manner, thereby lacking segment-specific prioritisation. The present paper introduces a novel methodology for systematic identification and the ranking of line sections for capacity upgrades. The approach is comprised of three distinct steps: first, the line is segmented using traffic homogeneity criteria; second, limiting journey times are determined through analytical capacity calculations based on the ninth decile of train volumes; and third, infrastructure measures are identified when the actual journey times exceed these thresholds. Potential interventions encompass the introduction of additional block sections, the implementation of passing loops, or the introduction of double-tracking. The methodology was applied to the Havlíčkův Brod–Jihlava–Znojmo line, thereby demonstrating its ability to detect bottlenecks and propose targeted measures. The findings indicate that there is considerable potential for enhancing capacity while concomitantly improving operational safety and cost efficiency. Consequently, this will serve to reinforce the role of diversionary lines within the broader context of the rail network. The proposed framework provides infrastructure managers with a generalisable tool with which to prioritise investments and support the long-term development of resilient and sustainable railway systems. Full article

This paper investigates the role of digital marketing signals as alternative data for understanding financial and operational dynamics in the beverage supply chain. Drawing on web analytics covering multiple actors across a five-month horizon, we analyze traffic composition, user engagement, and acquisition channels through a panel econometric framework. Descriptive statistics reveal pronounced heterogeneity in channel reliance, with some firms emphasizing organic search visibility while others depend more on paid campaigns or social referrals. Correlation patterns indicate strong substitution between organic and paid search, while display advertising is positively associated with session depth, suggesting that differentiated digital strategies influence user engagement. Analysis of variance confirms significant structural differences across firms, with an effect size exceeding 0.90. A two-way fixed-effects regression demonstrates that brand-specific factors explain the vast majority of variation in digital visibility, overshadowing short-term fluctuations. These results highlight the potential of web-derived marketing metrics to serve as leading indicators of supply chain finance outcomes such as revenue growth, working-capital efficiency, and investor sentiment. By integrating digital signals into financial econometrics, this study contributes to emerging research on alternative data in supply chain contexts and offers practical implications for managers, investors, and policymakers. Full article

This study applied a multi-analytical methodology involving Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR) spectroscopy, protein secondary structure determination, colorimetry, and texture analysis of milk images at a microscopic level to characterize 47 commercial cow and goat milk samples of different fat content (whole and light). Colorimetric measurements showed that hue values were significantly higher in light than in whole milks, providing a rapid marker of fat level, while microscopic image analysis indicated that whole milks had more heterogeneous textures with larger fat globules, whereas light milks were more homogeneous. ATR-FTIR spectra revealed lipids, proteins, and carbohydrates as the main constituents; lipid-associated bands were more intense in whole milks, whereas carbohydrate-associated bands, particularly at 1026–1028 cm⁻¹, were stronger in cow milk. Protein secondary structure analysis confirmed β-parallel sheet as the predominant motif, with cow milk showing higher random coil and α-helix proportions and goat milk enriched in β-turn structures. Chemometric modeling using PCA and PLS-DA achieved robust classification of samples by species and fat content, while Receiver Operation Characteristics (ROC) analysis validated markers of differentiation. The combination of the above methodologies enables effective classification of cow’s and goat’s milk, offering a thorough product description. Full article

Mealybugs are high-priority quarantine pests in fresh-produce trade due to cryptic habits, broad host ranges, and market-access risks. Phytosanitary irradiation (PI) provides a non-residual, process-controlled option that is increasingly integrated with modified-atmosphere (MA/MAP) logistics. Because molecular oxygen enhances indirect radiation damage (oxygen enhancement ratio, OER), oxygen limitation may modulate PI outcomes in mealybugs. The Jack Beardsley mealybug (Pseudococcus jackbeardsleyi) has an IPPC-adopted PI treatment of 166 Gy (ISPM 28, PT 45). We exposed adult females to 166 Gy under air and 1% O₂ and generated whole-transcriptome profiles across treatments. Differentially expressed genes and co-differentially expressed genes (co-DEGs) were integrated with protein–protein interaction (PPI) and regulatory networks, and ten hubs were validated by reverse transcription quantitative PCR (RT-qPCR). Hypoxia attenuated irradiation-induced transcriptional disruption. Expression programs shifted toward transport, redox buffering, and immune readiness, while morphogen signaling (Wnt, Hedgehog, BMP) was coherently suppressed; hubs including wg, hh, dpp, and ptc showed stronger down-regulation under hypoxia + irradiation than under irradiation alone. Despite these molecular differences, confirmatory bioassays at 166 Gy under both atmospheres (air and 1% O₂) achieved complete control. These results clarify how oxygen limitation modulates PI responses in a quarantine mealybug while confirming the operational efficacy of the prescribed 166 Gy dose. Practically, they support the current international standard and highlight the value of documenting oxygen atmospheres and managing dose margins when PI is applied within MA/MAP supply chains. Full article

Fractional heat conduction models extend classical formulations by incorporating fractional differential operators that capture multiscale relaxation effects. In this work, we introduce an electrical analogy that represents the action of these operators via generalized longitudinal impedance and admittance elements, thereby clarifying their physical role in energy transfer: fractional derivatives account for the redistribution of heat accumulation and dissipation within micro-scale heterogeneous structures. This analogy unifies different classes of fractional models—diffusive, wave-like, and mixed—as well as distinct fractional operator types, including the Caputo and Atangana–Baleanu forms. It also provides a general computational methodology for solving heat conduction problems through the concept of thermal impedance, defined as the ratio of surface temperature variations (relative to ambient equilibrium) to the applied heat flux. The approach is illustrated for a semi-infinite sample, where different models and operators are shown to generate characteristic spectral patterns in thermal impedance. By linking these spectral signatures of microstructural relaxation to experimentally measurable quantities, the framework not only establishes a unified theoretical foundation but also offers a practical computational tool for identifying relaxation mechanisms through impedance analysis in microscale thermal transport. Full article

This study investigates how process manufacturing enterprises can optimize semi-finished inventory (SFI) distribution in delayed production models, with particular attention to differences in cost volatility between single- and multi-period planning scenarios. To address this research gap, we develop a mixed-integer programming model that determines optimal customer order decoupling point (CODP)/product differentiation point (PDP) positions and SFI quantities (both generic and dedicated) for each production period, employing particle swarm optimization for solution derivation and validating findings through a comprehensive case study of a steel manufacturer with characteristic long-period production processes. The analysis yields two significant findings: (1) single-period operations demonstrate marked cost sensitivity to service level requirements and delay penalties, necessitating end-stage inventory buffers, and (2) multi-period optimization generates a distinctive cost-smoothing effect through strategic order deferrals and cross-period inventory reuse, resulting in remarkably stable total costs (≤2% variation observed). The study makes seminal theoretical contributions by revealing the convex cost sensitivity of short-term inventory decisions versus the near-flat cost trajectories achievable through multi-period planning, while establishing practical guidelines for process industries through its empirically validated two-period threshold for optimal order deferral and inventory positioning strategies. Full article

This main focus of this work is the fractional-order nonlinear Schrödinger equation with wave operators. First, a conservative difference scheme is constructed. Then, the discrete energy and mass conservation formulas are derived and maintained by the difference scheme constructed in this paper. Through rigorous theoretical analysis, it is proved that the constructed difference scheme is unconditionally stable and has second-order precision in both space and time. Due to the completely implicit property of the differential scheme proposed, a linearized iterative algorithm is proposed to implement the conservative differential scheme. Numerical experiments including one example with the fractional boundary conditions were studied. The results effectively demonstrate the long-term numerical behaviors of the fractional nonlinear Schrödinger equations with wave operators. Full article

In order to solve the problem encountered by traditional potato–stem separation devices, that is, they cannot meet the requirements when installed in small-scale harvesters, a new type of vertical differential roller potato–stem separation device was developed. The device features a compact structure and simultaneously possesses both separating and conveying functions. Through the analysis of the separation force between potato and stem, the structure and parameters of the separation device were determined. The simulation and the field test of the potato–stem separation process were carried out with the vertical differential roller speed, the vertical differential roller gap width and the conveyor chain speed as the influencing factors. The simulation test analysed the influence law of different working parameters on the performance of potato–stem separation. The field test revealed the order of the effects of various factors on the impurity rate and skin-breaking rate, concluding that the optimal combination of operational parameters was a vertical differential roller rotational speed of 6 s⁻¹, a vertical differential roller gap width of 7 mm, and a conveyor chain speed of 1.4 m·s⁻¹. This experiment fills the research gap in the study of potato–stem separation devices suitable for small-scale potato harvesters and promotes the development of compact potato harvesters. Full article

Background: Passive brain–computer interface (pBCI) systems use a combination of electroencephalography (EEG) and machine learning (ML) to evaluate a user’s cognitive and physiological state, with increasing applications in both clinical and non-clinical scenarios. pBCI systems have been limited by their traditional reliance on sensor technologies that cannot easily be integrated into non-laboratory settings where pBCIs are most needed. Advances in textile-electrode-based EEG show promise in overcoming the operational limitations; however, no study has demonstrated their use in pBCIs. This study presents the first application of fully textile-based EEG for pBCIs in differentiating cognitive states. Methods: Cognitive state comparisons between eyes-open (EO) and eyes-closed (EC) conditions were conducted using publicly available data for both novel textile and traditional dry-electrode EEG. EO vs. EC differences across both EEG sensor technologies were assessed in delta, theta, alpha, and beta EEG power bands, followed by the application of a Support Vector Machine (SVM) classifier. The SVM was applied to each EEG system separately and in a combined setting, where the classifier was trained on dry EEG data and tested on textile EEG data. Results: The textile EEG system accurately captured the characteristic increase in alpha power from EO to EC (p < 0.01), but power values were lower than those of dry EEG across all frequency bands. Classification accuracies for the standalone dry and textile systems were 96% and 92%, respectively. The cross-sensor generalizability assessment resulted in a 91% classification accuracy. Conclusions: This study presents the first use of textile-based EEG for pBCI applications. Our results indicate that textile-based EEG can reliably capture changes in EEG power bands between EO and EC, and that a pBCI system utilizing non-traditional textile electrodes is both accurate and generalizable. Full article

We present a tunable diode-laser absorption spectroscopy (TDLAS) system operating at 1.5711 µm for CO₂ gas concentration measurements. The system can operate in either a traditional direct-mode (dTDLAS) sawtooth wavelength scan or a recently demonstrated wavelength-toggled single laser differential-absorption lidar (WTSL-DIAL) mode using precompensated current pulses. The use of such precompensated pulses offsets the slow thermal constants of the diode laser, leading to fast toggling between ON and OFF-resonance wavelengths. A short measurement time is indeed pivotal for atmospheric sensing, where ambient factors, such as turbulence or mechanical vibrations, would otherwise deteriorate sensitivity, precision and accuracy. Having a system able to operate in both modes allows us to benchmark the novel experimental procedure against the well-established dTDLAS method. The theory behind the new WTSL-DIAL method is also expanded to include the periodicity of the current modulation, fundamental for the calculation of the OFF-resonance wavelength. A two-detector scheme is chosen to suppress the influence of laser intensity fluctuations in time (1/f noise), and its performance is eventually benchmarked against a one-detector approach. The main difference between dTDLAS and WTSL-DIAL, in terms of signal processing, lies in the fact that while the former requires time-consuming data processing, which limits the maximum update rate of the instrument, the latter allows for computationally simpler and faster concentration readings. To compare other performance metrics, the update rate was kept at 2 kHz for both methods. To analyze the dTDLAS data, a four-parameter Lorentzian fit was performed, where the fitting function comprised the six main neighboring absorption lines centered around 1.5711 µm. Similarly, the spectral overlap between the same lines was considered when analyzing the WTSL-DIAL data in real time. Our investigation shows that, for the studied time intervals, the WTSL-DIAL approach is 3.65 ± 0.04 times more precise; however, the dTDLAS-derived CO₂ concentration measurements are less subject to systematic errors, in particular pressure-induced ones. The experimental results are accompanied by a thorough explanation and discussion of the models used, as well as their advantages and limitations. Full article

This study evaluates media-level filtration behaviour and short-term fuel consumption outcomes for five spin-on lubricating oil filters operated under real driving conditions at high altitude. To improve interpretability, filters are reported using parameter-based identifiers (media descriptors and equivalent circular diameter, ECD) rather than internal codes. Pore-scale morphology was quantified by microscopy and expressed as ECD, and bulk fluid cleanliness was summarised using ISO 4406 codes. Trials were conducted over representative urban and extra-urban routes at altitude; fuel consumption was analysed using ANCOVA. The results indicated clear media-level differences (tighter pore envelopes and cleaner ISO codes, particularly for two OEM units). However, fuel-consumption differences were not statistically significant (ANCOVA, p = 0.29). Accordingly, findings are reported as short-term cleanliness and media characterisation under high-altitude duty rather than durability or efficiency claims. The parameter-based framing clarifies trade-offs across metrics and avoids over-generalisation from brand or part numbers. The work highlights the value of ECD as a comparative pore metric and underscores limitations of microscopy/cleanliness data for inferring engine wear or long-term consumption. Future work will incorporate formal multi-pass testing (ISO 4548-12), direct differential-pressure instrumentation, used-oil viscosity tracking, and wear-metal spectrometry to enable cross-vendor benchmarking and causal interpretation. Findings are presented as short-term cleanliness and media characterisation; no durability claims are made in the absence of direct wear measurements. Full article