Search Results (1,307)

This work addresses the post-combustion CO₂ capture demand, employing a single-atom solvent as the absorbent to conduct systematic research on process modeling, energy consumption analysis, and multi-objective optimization. The energy consumption of the single-atom solvent-enhanced CO₂ capture process was reduced to 2.864 GJ/t, representing a 26.9% reduction compared with that of the conventional solution. The effects of solvent flow rate, gas flow rate, rich solvent temperature, reflux ratio, and extraction ratio on the energy consumption, annual total utility consumption, CO₂ equivalent emissions, and total annual cost were systematically investigated. The results indicate that rich solvent temperature and reflux ratio are the most sensitive parameters affecting system energy consumption variations; increasing solvent flow rate linearly elevates the reboiler duty, whereas gas flow rate variations exert negligible influence on system performance. The energy consumption was further reduced to 1.84 GJ/t CO₂ after process parameter optimization. A multi-objective optimization approach coupling the NSGA-II with Aspen Plus process simulation was developed for economic–energy–environmental optimization. Annual total utility consumption was reduced by 12.75%, CO₂ equivalent emissions per unit of product were reduced by 47.87%, and total annual cost was reduced by 13.09% after optimization. The optimal operating conditions under multi-objective optimization were determined simultaneously. This study provides an optimization strategy for the industrial application of CO₂ capture technology. Full article

Environmental, social, and governance (ESG) quality cannot be directly observed because substantial disagreement across rating providers contaminates the observed ESG scores with measurement error. This paper addresses this problem by modelling the true ESG state as a latent, vector-valued Itô diffusion defined on a filtered probability space whose information set is progressively enlarged by FinTech signals. We establish the well-posedness of the latent ESG process, prove the existence of an equivalent martingale measure under an explicit exponential-moment condition, and solve an ESG-constrained portfolio problem under a wealth-scaled sustainability constraint through a Hamilton–Jacobi–Bellman verification theorem. Computationally, raw sustainability information is extracted from SEC Form 10-K filings using a FinBERT transformer architecture and incorporated into a linear Gaussian state-space model, where the latent ESG state is recovered via Kalman filtering. Theoretical results are then linked to asset pricing, portfolio allocation, and systemic risk networks through a common filtered ESG factor. Using an unbalanced panel of 1086 U.S. listed firms over 2011–2023 and ESG information from MSCI, Refinitiv, and Sustainalytics, we document substantial provider disagreement and show that the observed ESG ratings contain significant transitory measurement noise. The filtered ESG state exhibits higher reliability, lower noise, and greater persistence than individual provider scores. In asset pricing tests, the latent ESG state predicts future excess returns, whereas a composite provider-based ESG measure does not; a one-standard-deviation increase in the latent ESG state is associated with approximately 0.35 percentage points higher monthly excess returns (about 4.3% annualised). When both measures are included simultaneously, only the filtered ESG state retains explanatory power. Out-of-sample portfolio tests show that a latent ESG strategy achieves a Sharpe ratio of 0.72, significantly exceeding both an unconstrained benchmark (0.59) and a composite ESG screen strategy (0.55). At the network level, ESG-adjusted weighting attenuates systemic fragility by reducing the spectral abscissa from 0.34 to 0.21, with the mitigating effect remaining significant under permutation-based placebo tests. Overall, the evidence supports the central hypothesis that ESG measurement error attenuates the observed pricing effects and that FinTech-enabled filtering recovers economically meaningful sustainability information relevant for asset pricing, portfolio construction, and systemic risk assessment. Full article

Crude vegetable oils contain, in addition to glycerides, a range of co-extracted compounds whose composition depends strongly on the biological source. Their removal typically requires multi-stage refining, increasing process complexity as well as solvent and energy demands. In this study, a simple preparative chromatographic method was developed to purify crude algal lipid extracts and recover a glyceride-rich fraction. The process employed hexane as the mobile phase and cellulose-based materials as the stationary phase. The underlying separation mechanism was investigated experimentally using paper chromatography, thin-layer chromatography (TLC), and Fourier transform infrared spectroscopy (FTIR), and was further interpreted using Hansen solubility parameters (HSPs). Approximately 50% of the crude lipid extract was recovered as a glyceride-rich, colorless fraction. A secondary fraction enriched in sterols, chlorophylls, and carotenoids were also recovered, highlighting the potential for simultaneous purification and recovery of valuable lipid-associated compounds. To support potential process scale-up, regeneration of the stationary phase was evaluated, with each cycle required 0.225 L of fresh hexane and 0.1 L of fresh methanol. In addition, low-cost cellulose-rich residues, including used paperboard and wood dust, were successfully validated as alternative stationary phase materials. Overall, the proposed process provides a straightforward and potentially scalable approach to refining algal lipid extracts while enabling the recovery of additional value-added fractions and the use of inexpensive cellulose-based materials. Full article

The management of chronic wounds represents one of the major challenges in regenerative medicine, as the healing process can be compromised by infections, oxidative stress, and persistent inflammation. In this context, wound pH serves as an important biomarker of tissue status, highlighting the need for smart dressings capable of promoting regeneration while simultaneously monitoring the wound microenvironment. In this study, biomimetic silk fibroin films functionalized with curcuminoids extracted from Curcuma longa were developed and characterized through spectroscopic, swelling/degradation, antioxidant, colorimetric, and biological assays, with the aim of obtaining a multifunctional dressing with regenerative properties and pH responsiveness. The results showed that curcuminoids were physically incorporated into the protein matrix without altering its chemical structure. The films exhibited a high absorption ability and antioxidant activity in the initial stages, and a clear and reversible color change in response to pH. Biological assays on 3T3 fibroblasts further confirmed the high cytocompatibility of the materials and their ability to support cell migration and wound closure in vitro. The developed films represent a promising biomimetic platform for advanced wound dressings, capable of combining support for tissue regeneration, antioxidant protection, and visual monitoring of wound status through the detection of pH changes. Full article

Starch, a natural source of energy in the form of glucose chains, is widely utilized in various industrial and scientific fields. In its native state, starch is thermally unstable and undergoes gelatinization. Physicochemical modifications of starch aim to increase its thermal and structural stability while simultaneously enhancing its reactivity by introducing new functional groups. The primary objective of this study was to develop thermally stable and economically viable starch-based drug carriers capable of the controlled release of a negatively charged component sourced from aronia extract. Potato starch underwent a series of chemical modifications, specifically quaternary amine etherification, citric acid esterification, and/or hydrothermal modification. The characterization involved determining several parameters: the degree of amino substitution groups; starch particle size using a laser particle size analyzer; viscosity and pH; gelation temperature and heat capacity measured by scanning calorimetry (DSC); mass degradation analyzed via thermogravimetric analysis (TG); crystallinity determined by X-ray diffraction (XRD); potential intermolecular interactions studied by Fourier-Transform Infrared Spectroscopy with Attenuated Total Reflectance (FTIR-ATR); and the rate of chlorogenic acid release from aronia extract tablets quantified by spectrophotometry. The highest cationization results were achieved using citrate starches, reaching up to 86%. The combined application of citric acid esterification and cationization, coupled with an annealing process, resulted in increased viscosity, amorphousness, and enzyme resistance of the starch. Citric acid esterification significantly improved the thermal stability of the starch. Furthermore, FTIR studies revealed the formation of electrostatic interactions between the functional groups of the starch and the components of aronia extract. The amount of chlorogenic acid released showed significant variation (70–100%) depending on the type of starch modification. Collectively, these studies confirmed that both hydrothermal and chemical modifications influence the thermal and structural stability of the starch. Utilizing all combination modification strategies ensured the production of highly promising carriers for active substances. Full article

Coastal zones are subject to a range of natural and anthropogenic processes that result in coastal erosion and accretion, threatening essential infrastructure and straining livelihoods. Analysis of shoreline changes is thus crucial for informing coastal zone planning and management to avert the ramifications of erosion and accretion. Despite a range of literature examining coastline changes worldwide, there is a paucity of literature focusing on Southern Africa, particularly within small harbours. The paper, therefore, aims to analyse shoreline changes at four small harbour zones along the Atlantic Ocean in the Western Cape province, South Africa, over the period from 1985 to 2025. To acquire an accurate shoreline position, four spectral criteria were applied simultaneously: the Automated Water Extraction Index (AWEI), the Modified Normalised Difference Water Index (MNDWI), the Normalised Difference Vegetation Index (NDVI), and the Near Infrared (NIR). Four statistics were then used to measure shoreline changes in the USGS Digital Shoreline Analysis System (DSAS): Net Shoreline Movement (NSM), Shoreline Change Envelope (SCE), End Point Rate (EPR), and Weighted Linear Regression (WLR). Considerable variability was observed within and among the four small harbour study areas, with several erosion and accretion hotspots identified. The 20-year forecast indicated that future shoreline positions would largely maintain the 2025 curvature. Although the study did not reveal significant threats, authorities are encouraged to pay particular attention to erosion and accretion hotspots through appropriate mitigation and adaptation efforts. Full article

Accurate semantic segmentation of large-scale outdoor LiDAR point clouds remains a challenging endeavor, primarily due to ambiguous class transitions at object interfaces, non-uniform sampling density across the surveyed area, and shared geometric signatures among distinct object categories. This paper proposes GFE-Net (Geometry-Enhanced Feature Extraction Network), a hierarchical encoder–decoder architecture that systematically improves per-point feature characterization through three complementary design contributions: First, to mitigate the shortcomings of conventional fixed-neighborhood queries in regions of variable point density, a Structure-Guided Neighborhood Adaptation (SGNA) module is devised. At its core lies a morphology-driven contextual gating (MCG) unit that synthesizes neighbor-wise calibration weights from hierarchical shape descriptors fused with elevation difference statistics, allowing the network to preferentially amplify morphologically congruent neighbors while dampening spurious or cross-boundary contributions. Second, to strengthen semantic discrimination beyond what spatial locality alone affords, a Local–Global Interactive Enhancement (LGIE) module is presented. The LGIE module simultaneously distills precise local structure through Euclidean-space neighborhood graphs and captures scene-wide co-activation patterns via compact bilinear factorization of the latent feature space, merging both streams through a residual refinement mechanism that markedly improves inter-class separability. Third, to enforce label consistency at object interfaces without relying on post-processing heuristics, a Neighborhood Prediction Consistency (NPC) loss is introduced. Built upon a Gaussian distance-decay weighting kernel, the NPC loss assigns progressively stronger penalties to label mismatches between a query point and its geometrically proximate neighbors, thereby promoting spatially coherent predictions and attenuating boundary noise. GFE-Net is rigorously benchmarked on two widely adopted large-scale datasets—S3DIS and SensatUrban—yielding OA/mIoU of 89.6%/73.1% and 93.3%/61.1%, respectively. These results demonstrate competitive performance under the reported protocols. Detailed ablation studies and computational profiling further substantiate the efficacy of each individual component. Full article

The global seafood industry generates substantial quantities of processing by-products such as heads, viscera, skin, bones, scales, and shells. These residues represent an underutilized resource rich in proteins, lipids, minerals, enzymes, and polysaccharides. Conventional valorization approaches such as chemical extraction, wet rendering, and enzymatic hydrolysis have been used to recover valuable compounds from seafood waste. However, conventional methods often involve high chemical consumption, long processing times, and environmental concerns. Green extraction technologies have emerged as promising alternatives, with subcritical water extraction (SWE) gaining significant attention due to its unique properties and ability to simultaneously extract and convert biomass components. This review provides a comprehensive overview of the valorization of seafood processing wastes using SWE. Particular emphasis is placed on the physicochemical properties of subcritical water, the reaction mechanisms governing the hydrolysis and transformation of proteins, lipids, and polysaccharides, and the key parameters influencing extraction performance. Recent advances in the recovery of value-added products such as amino acids, bioactive peptides, protein hydrolysates, omega-3-rich oils, chitin derivatives, and mineral-rich materials are summarized. In addition, the integration of SWE with complementary technologies such as supercritical CO₂ extraction, enzymatic hydrolysis, and hydrothermal carbonization is examined as a strategy for developing integrated seafood biorefineries. Full article

Yellowfin tuna (Thunnus albacares) processing by-products are underutilized resources rich in high-quality proteins that can be valorized into bioactive ingredients. In this study, proteins from yellowfin tuna tail bones were hydrolyzed using either an endogenous protease extract recovered from tuna viscera or commercial trypsin. The endogenous extract was characterized and exhibited optimal proteolytic activity at pH 8 and 40 °C. Comparative bioactivity assessment showed that trypsin-derived hydrolysates exhibited significantly higher antioxidant capacity, with ABTS and DPPH values of 11.8 ± 0.04 and 13.3 ± 0.40 mg Trolox/g, respectively, compared with 10.1 ± 0.12 and 7.3 ± 0.21 mg Trolox/g for hydrolysates produced with endogenous enzymes. Trypsin hydrolysates also showed stronger dipeptidyl peptidase-IV (DPP-IV) inhibition (IC₅₀ = 0.83 ± 0.37 mg/mL) than endogenous-enzyme hydrolysates (IC₅₀ = 1.60 ± 0.47 mg/mL). Remarkably, hydrolysates generated with the endogenous enzymatic consortium exhibited exceptionally potent angiotensin-converting enzyme (ACE) inhibitory activity (IC₅₀ = 0.008 ± 0.004 mg/mL), outperforming trypsin hydrolysates (IC₅₀ = 0.01 ± 0.007 mg/mL). Neither hydrolysate showed evidence of acute toxicity in the Artemia salina model at the tested concentrations, supporting their favorable performance in this preliminary toxicity screening assay. Considering that nearly two-thirds of total fish biomass is discarded during processing, this integrated bioprocess demonstrates a sustainable strategy for the simultaneous valorization of skeletal and visceral waste streams. By employing endogenous enzymes as biocatalysts, this approach reduces dependence on commercial proteases, mitigates environmental burdens, and supports the development of a circular marine bioeconomy through the production of highly potent, peptide-rich functional ingredients with promising cardiovascular health applications. Further toxicological evaluation is required to confirm their safety for food and nutraceutical applications. Full article

Hybrid CNN–Transformer (HCT) synthesis for automated neurodevelopmental diagnostics is an effective approach to constructing intelligent detection systems that are not merely oriented toward feature classification but primarily toward solving spatiotemporal pattern recognition problems in motor disorder assessment. In neurodevelopmental diagnostics, existing automated methods rely on fixed, single-model architectures that process spatial or temporal motion features independently, failing to adapt to the heterogeneous motor irregularities characteristic of developmental coordination disorder and degrading detection sensitivity and generalization across diverse patient populations. There is therefore a pressing need for models capable of simultaneously capturing intra-frame spatial coordination patterns and inter-frame temporal movement dependencies against interrelated diagnostic criteria including accuracy, sensitivity, and motor irregularity specificity. To address this challenge, this paper proposes HCT, a novel framework that integrates ResNet-based spatial feature extraction from optical flow maps and pose estimation skeletons with multi-head self-attention Transformer encoding for modeling long-range temporal dependencies across multi-frame motion sequences. Unlike conventional single-stream approaches, where spatial and temporal processing remain confined to independent architectures, HCT decouples spatiotemporal feature learning through a cross-modal fusion pipeline, constructing a unified discriminative architecture that captures motor coordination dependencies between motion imaging inputs and multiple diagnostic criteria simultaneously. The convolutional encoder generates diverse joint displacement features, which are consolidated through cross-modal attention fusion into a robust, unified embedding with enhanced generalization and resilience to inter-individual motor variability. Integration within neurodevelopmental assessment frameworks facilitates reliable developmental coordination disorder classification, motor irregularity prediction, and interpretable diagnostic decision support, advancing the accuracy, flexibility, and clinical validity of intelligent motor disorder diagnostic systems. Full article

Polymer-based three-dimensional (3D) printing has evolved from a prototyping approach toward a manufacturing strategy with emerging clinical relevance for individualized dosage forms, local drug depots, microneedle systems, microfluidic cartridges, biosensor housings and integrated theranostic platforms. Its value arises from the simultaneous control of polymer chemistry, device architecture and process history: infill, porosity, shell thickness, crosslink density, swelling, degradation and surface chemistry can be used as design variables rather than incidental manufacturing outcomes. This review critically synthesizes recent progress in polymer-enabled additive manufacturing for drug delivery and diagnostic applications, with emphasis on thermoplastic and biodegradable polymers, hydrogels, photopolymers, elastomers, conductive composites, stimuli-responsive networks and bioinks. Fused deposition modelling, hot-melt extrusion, semi-solid extrusion, vat photopolymerization, two-photon polymerization, selective laser sintering, binder jetting and inkjet/aerosol jet approaches are compared in relation to drug stability, diagnostic compatibility, feature resolution, scalability and regulatory risk. Particular attention is given to geometry-controlled release, multi-drug printlets, microneedles, implants, scaffold-based local therapy, microfluidic diagnostics, electrochemical biosensors and wearable or closed-loop systems. Translation is discussed through quality-by-design, critical material attributes, critical process parameters, process analytical technology, extractables/leachables, sterilization, point-of-care manufacturing, data integrity and clinical evidence requirements. Future advances should connect polymer–process–property relationships with clinically meaningful use cases, verified quality attributes and realistic regulatory pathways. Full article

Accurate estimation of subsurface acoustic impedance is fundamental to quantitative reservoir characterization in seismic exploration. Nevertheless, a single network architecture cannot adequately represent both the local details and the global trends of seismic records within a unified framework, while the severe scarcity of annotated well-log data substantially constrains the generalization capability and predictive accuracy of deep-learning-based inversion approaches. To overcome these limitations, a semi-supervised acoustic impedance inversion framework based on a hybrid deep learning architecture is proposed. The framework employs a cascaded architecture consisting of a multi-scale depthwise separable convolution with channel attention (MSDSE) module and a convolution-augmented Transformer encoder. Seismic data are first processed by the MSDSE module to extract local multi-scale temporal features, and are subsequently passed to the convolution-augmented Transformer encoder, which captures global long-range sequence dependencies while retaining complementary local temporal information. The two modules progress hierarchically and jointly achieve a feature representation that spans from local details to global trends, and the initial low-frequency model is fused with the network output via channel-wise concatenation. Meanwhile, an initial-model constraint together with a physical-consistency constraint are simultaneously imposed within the loss function, thereby improving training stability while fully leveraging the physical information embedded in unlabeled traces. Experiments on both synthetic and field data confirm the effectiveness of the proposed method. The results show that, even with a small number of labels, the method produces stable impedance estimates and outperforms conventional deep learning methods in both generalization and prediction accuracy. Full article

High-resolution remote sensing object detection is a fundamental task in Earth observation. However, the storage, transmission, and downstream processing of high-resolution images impose substantial bandwidth, memory, and computational burdens. Compressed sensing (CS) samples and compresses the signals simultaneously, thereby reducing data transmission and storage overhead. Unfortunately, most existing CS-based pipelines require explicit image reconstruction before downstream inference, leading to heavy computational overheads and poor scalability for high-resolution remote sensing images (RSIs). This work focuses on post-acquisition image compression and explores algorithmically, rather than from a physical hardware implementation perspective, whether explicit image reconstruction is an indispensable intermediate step prior to object detection. To this end, we propose HR²SIOD-CL, an end-to-end compressed learning (CL) framework that performs object detection directly on CS measurements of high-resolution RSIs without explicit image reconstruction. HR²SIOD-CL integrates an entropy-driven content-aware adaptive sampling strategy and a measurement-domain detection backbone for multi-scale feature extraction. Although jointly optimized during training, the adaptive sampling and detection modules can be decoupled for flexible deployment. For fair comparison, an extra lightweight reconstruction network equipped with a single-step data-consistency correction is constructed as the baseline. Extensive experiments on the NWPU VHR-10 and DIOR datasets show that across various sampling ratios, HR²SIOD-CL surpasses the reconstruction-based detection method when integrated into two-stage detectors, and achieves comparable or superior detection performance to the reconstruction-based counterparts when integrated into single-stage detectors. Meanwhile, its computational overhead and GPU memory consumption are merely 6.97% and 35.74% of those of the reconstruction-based counterpart, respectively. These results indicate that CS measurements can function as an effective intermediate representation for object detection, and explicit image reconstruction is not a prerequisite when detection is the primary objective. Full article

The rapid development of large model-driven agent applications, such as digital assistants and robots, requires 6G radio access networks (RAN) to deliver enhanced flexibility, adaptability, and low-latency capabilities. However, the existing RAN user plane (UP) architecture suffers from coarse decoupling granularity and significant cross-layer functional redundancy. These limitations severely hinder the on-demand orchestration and dynamic reconfiguration required by heterogeneous agent services. To address these challenges, this paper proposes a ComBERT-driven service-based RAN UP decoupling method, specifically targeting the functional coupling and redundancy between the PDCP and RLC sublayers. First, we develop a domain-specific language model, ComBERT, by pre-training a BERT model on a 3GPP protocol corpus and fine-tuning it on text-matching tasks to deeply comprehend protocol semantics. Subsequently, ComBERT is utilized to extract semantic features from UP functional components, employing a sliding window mechanism to overcome truncation in lengthy protocol texts and using cosine similarity to measure functional relevance. Finally, a threshold-based fusion algorithm is designed to identify and merge cross-layer redundant functions, thereby forming independent service units with distinct responsibilities. These fused units serve as the basic building blocks for scenario-specific orchestration. Simulation results demonstrate that the proposed method reduces the number of UP components by 12.5%, 18.7%, and 18.2% in eMBB, URLLC, and mMTC scenarios, respectively. Simultaneously, it decreases average processing delays by 7.9%, 10.2%, and 11.0% across these respective scenarios. Ultimately, this approach effectively improves the lightweight deployment, processing efficiency, and reconfiguration capabilities of the service-based UP, providing a crucial foundation for on-demand service orchestration in 6G networks tailored to agent services. Full article

Manual fruit thinning is labor-intensive and inefficient, making the development of intelligent visual detection systems a crucial approach for improving the quality and production efficiency of the pear industry. However, in natural orchard environments, young pear fruits are small in size with slender fruit stalks, and their texture and color characteristics are highly similar to those of tender branches. Furthermore, variations in illumination and occlusions caused by branches and leaves make it difficult for existing detection models to simultaneously and accurately identify fruits and fruit stalks, limiting their application in automated thinning equipment. In this study, Yuluxiang pear was selected as the research object, and image data were collected under diverse field conditions, including forward-lighting, backlighting, close-range shooting, long-range shooting and fruit overlapping. A dedicated dataset containing 3057 images was established. Based on the RT-DETR-R18 network, a lightweight and high-precision fruit–stalk synchronous detection model was proposed. Specifically, the backbone network was reconstructed by integrating GCConv with C2f modules to enhance global feature extraction for slender fruit stalks. The bottleneck structure was optimized using GCConvC3 to reduce feature degradation under occlusion conditions, and an additional 4× down-sampling P2 detection head was introduced to improve the detection capability for small targets. To fully validate the model performance and stability, three types of experiments were conducted in this study: ablation experiments, repeated experiments with different random seeds, and comparative experiments. Ablation experiments verified the cumulative performance improvements brought by the introduced modules. Repeated experiments with different random seeds were performed to explore training randomness-induced performance fluctuations, and the results demonstrated that the proposed model maintains stable overall detection accuracy with minor metric fluctuations. Comparative experiments demonstrated that the proposed model achieved a compact parameter size of only 15.97 M, with a precision of 95.0% for young pear fruit detection and an mAP50 of 83.0% for fruit stalk detection, outperforming all comparative models in overall mAP50. The training convergence curves and Grad-CAM++ visualization results further confirmed the stable optimization process and enhanced feature attention capability of the proposed model. By achieving a favorable balance between detection accuracy and model lightweightness, this approach provides effective technical support for the development of intelligent fruit thinning equipment and vision-based systems for smart pear orchards. Full article