MDPI - Publisher of Open Access Journals

21 pages, 6376 KB

Open AccessArticle

Effects of Sub-Inhibitory Rifampicin, Minocycline, and Dalbavancin on Early Biofilm Formation and Transcriptional Responses in Staphylococcus aureus SA113

by Adam Bieda, Sabine Illner, Volkmar Senz, Stefan Oschatz, Niels Grabow, Micha Löbermann, Emil Christian Reisinger and Martina Sombetzki

Microorganisms 2026, 14(9), 2101; https://doi.org/10.3390/microorganisms14092101 (registering DOI) - 19 Sep 2026

Abstract

Implant-associated infections are often caused by biofilm-forming pathogens such as Staphylococcus (S.) aureus. While local antibiotic delivery systems aim to prevent adhesion and biofilm establishment, declining drug concentrations may result in sub-inhibitory exposure, which can modulate bacterial adaptation linked to antibiotic resistance, [...] Read more.

Implant-associated infections are often caused by biofilm-forming pathogens such as Staphylococcus (S.) aureus. While local antibiotic delivery systems aim to prevent adhesion and biofilm establishment, declining drug concentrations may result in sub-inhibitory exposure, which can modulate bacterial adaptation linked to antibiotic resistance, biofilm formation, and regulatory responses. We investigated the effects of sub-inhibitory antibiotic exposure on biofilm formation in S. aureus SA113 and associated transcriptional responses. The biofilm-producing strain S. aureus SA113 was exposed to sub-inhibitory concentrations of rifampicin, minocycline, and dalbavancin during early biofilm formation. Phenotypic effects were assessed by crystal violet staining, enumeration of colony-forming units, and scanning electron microscopy, while transcriptional responses were analyzed by qPCR. Despite stable counts of culturable adherent bacteria, sub-inhibitory antibiotic exposure differentially altered biofilm formation and transcriptional responses. Rifampicin was associated with increased biomass at higher sub-inhibitory concentrations and increased early expression of icaA, icaD (+4.2 log₂) and fnbA (+2.7 log₂) at 1/2× MIC. Minocycline reduced biomass at lower concentrations with partial recovery toward control levels at 1/2× MIC, while transcriptional analysis at 1/4× MIC after 6 h showed increased expression of icaA, icaD (+2.1 log₂) and fnbA (+4.3 log₂). Dalbavancin induced a distinct transcriptional response characterized by increased expression of vraS (+1.1 log₂) and lrgA (+1.8 log₂), without induction of matrix-associated genes, while adherent biofilm biomass was reduced to 42.9% at 1/2× MIC. Morphologically, this was associated with compact aggregates rather than diffuse biofilm structures. Sub-inhibitory antibiotic exposure differentially modulated early biofilm formation in SA113 in a drug-specific manner. Overall, the distinct dalbavancin-associated response may be relevant for the development of preventive local drug-delivery systems. Full article

26 pages, 3654 KB

Open AccessArticle

Collaborative Optimization of Ladle Furnace Operating Parameters Using Prediction Models and Case-Guided Genetic–Tabu Search

by Yuhong Du, Xiaolong Li and Dongfeng He

Processes 2026, 14(18), 2994; https://doi.org/10.3390/pr14182994 (registering DOI) - 19 Sep 2026

Abstract

Intelligent control of the ladle furnace (LF) process and its endpoint is essential for product quality and stable continuous casting. Existing studies mainly address endpoint prediction or operating-parameter recommendation. Prediction models rarely provide multivariable operating schemes directly, whereas recommendation models often suffer from [...] Read more.

Intelligent control of the ladle furnace (LF) process and its endpoint is essential for product quality and stable continuous casting. Existing studies mainly address endpoint prediction or operating-parameter recommendation. Prediction models rarely provide multivariable operating schemes directly, whereas recommendation models often suffer from insufficient coordination among modules and complex commissioning. This study proposes a collaborative LF operating-parameter optimization method combining endpoint prediction with case-guided genetic–tabu search. Given the initial heat state and target endpoint temperature, the method treats electric energy input, power-on duration, and key material additions as decision variables, evaluates each candidate using temperature and composition prediction models, and coordinates the variables through a unified objective. A dynamic weighting mechanism coupling generational annealing with feasible-population temperature-error feedback balances endpoint quality against resource input. Case-based reasoning guides population initialization, while a real-coded genetic algorithm and tabu search strengthen global exploration and local exploitation. For 400 independent historical heats, the method obtained a recommendation satisfying all model constraints for every heat. Relative to the corresponding historical operations, the mean recommended quantities of lime, slag agent, aluminum granules, high-carbon ferromanganese, electric energy input, and power-on duration were reduced by 6.98%, 12.79%, 9.34%, 8.85%, 7.89%, and 8.95%, respectively. Case-guided initialization improved first-generation solution quality and early convergence, whereas tabu search enhanced mid-to-late local refinement. The method converts existing endpoint-prediction capability into coordinated multivariable operating recommendations. Full article

(This article belongs to the Section AI-Enabled Process Engineering)

► Show Figures

Figure 1

31 pages, 3787 KB

Open AccessArticle

Digital Predistortion of Wideband Power Amplifiers Using Functionally Decoupled Envelope-Assisted Attention-Guided Recurrent Architecture

by Bingwen Qiu, Xiaoyu Li and Yunjie Zhao

Sensors 2026, 26(18), 5945; https://doi.org/10.3390/s26185945 (registering DOI) - 19 Sep 2026

Abstract

Wideband power amplifiers (PAs) operating with high-order modulation signals exhibit strong nonlinear distortion and dynamic memory effects, making real-time digital predistortion (DPD) increasingly challenging under strict computational constraints. This work proposes a functionally decoupled neural DPD architecture, termed EA-CCF-AttGRU, which explicitly separates instantaneous [...] Read more.

Wideband power amplifiers (PAs) operating with high-order modulation signals exhibit strong nonlinear distortion and dynamic memory effects, making real-time digital predistortion (DPD) increasingly challenging under strict computational constraints. This work proposes a functionally decoupled neural DPD architecture, termed EA-CCF-AttGRU, which explicitly separates instantaneous nonlinear feature representation from temporal memory compensation within a unified end-to-end framework. Instead of introducing envelope features, cross-channel fusion, and recurrent attention as isolated modules, the proposed architecture assigns different compensation functions to dedicated components: envelope-assisted augmentation and point-wise cross-channel fusion enhance instantaneous nonlinear representation, while attention-guided recurrent modeling captures dynamic memory effects. A global linear bypass further reduces the burden of nonlinear compensation by preserving the linear transformation. Experimental results under a 160 MHz 1024-ary quadrature amplitude modulation (1024-QAM) baseband excitation with a 10.38 dB peak-to-average power ratio (PAPR) demonstrate that the proposed method achieves an adjacent channel leakage ratio (ACLR) of −65.91 dBc, a normalized mean square error (NMSE) of −57.84 dB, and an error vector magnitude (EVM) of 0.07% with only 6009 trainable parameters. The proposed architecture achieves an effective complexity–performance trade-off for wideband DPD applications and provides potential for future hardware-oriented implementation and synthesis validation. Full article

(This article belongs to the Special Issue Microwave, RF, and mm-Wave Circuits for Advanced Wireless Communications and Sensing)

► Show Figures

Figure 1

38 pages, 1683 KB

Open AccessArticle

Digital Transformation and Organizational Resilience in the Turkish Hospitality Sector: The Mediating Role of Strategic Agility and the Moderating Effect of Digital Leadership Toward Sustainable Tourism Management

by Ahmad Hamad and Ayşem Iyikal Çelebi

Sustainability 2026, 18(18), 9626; https://doi.org/10.3390/su18189626 (registering DOI) - 19 Sep 2026

Abstract

The rapid advancement of digital technologies and the increasing frequency of organizational disruptions have made digital transformation and organizational resilience central concerns for hospitality management research and practice. Drawing on Dynamic Capabilities Theory and upper echelons theory, this study develops and tests a [...] Read more.

The rapid advancement of digital technologies and the increasing frequency of organizational disruptions have made digital transformation and organizational resilience central concerns for hospitality management research and practice. Drawing on Dynamic Capabilities Theory and upper echelons theory, this study develops and tests a dual-stage moderated mediation model examining the associations between digital transformation and organizational resilience through the mediating role of strategic agility, and the moderating role of digital leadership on both the digital transformation and strategic agility association and the strategic agility and organizational resilience association. Data were collected from 436 managers employed in 4- and 5-star hotels across major Turkish tourism destinations, and hypotheses were tested using Partial Least Squares Structural Equation Modeling (PLS-SEM) with SmartPLS 4. The results confirm that digital transformation is positively and significantly associated with both organizational resilience and strategic agility, and that strategic agility is positively associated with organizational resilience. Strategic agility partially mediates the digital transformation and organizational resilience association, while digital leadership significantly moderates both the digital transformation and strategic agility association and the strategic agility and organizational resilience association, amplifying these associations under high digital leadership conditions. A formal moderated mediation test confirmed that the indirect effect of digital transformation on organizational resilience through strategic agility varies significantly across levels of digital leadership (index of moderated mediation = 0.027, p = 0.023). These findings advance the application of Dynamic Capabilities Theory to hospitality resilience research, enrich the strategic agility and digital leadership studies, and offer actionable insights for hotel executives and policymakers navigating digital transformation in disruption-prone tourism environments. While sustainability was not directly measured, the findings theoretically position digitally enabled agility and organizational resilience as sustainability-relevant outcomes contributing to the long-term viability of hospitality operations. Full article

(This article belongs to the Special Issue AI-Driven Smart Business Strategies for Sustainable Innovation, Resilience, and Digital Transformation)

► Show Figures

Figure 1

20 pages, 43556 KB

Open AccessArticle

Contrasting Patterns of Soil Microbial Communities, Physicochemical Properties and Vegetation Characteristics Along Altitude and Soil Depths in an Alpine Ecosystem

by Xinyu Guo, Xue Yang, Xu Wang, Rui Chong, Shicheng Yan, Fujiang Hou, Peiqiang Yu, Zechen Peng and Tao Ran

Biology 2026, 15(18), 1663; https://doi.org/10.3390/biology15181663 (registering DOI) - 19 Sep 2026

Abstract

Altitude and soil depth were associated with vegetation, soil properties, and microbial communities in alpine ecosystems, yet a comprehensive understanding of how these gradients shape soil microbial and physicochemical characteristics on the Qinghai–Tibetan Plateau remains limited. Along altitudinal gradients (low, mid, and high) [...] Read more.

Altitude and soil depth were associated with vegetation, soil properties, and microbial communities in alpine ecosystems, yet a comprehensive understanding of how these gradients shape soil microbial and physicochemical characteristics on the Qinghai–Tibetan Plateau remains limited. Along altitudinal gradients (low, mid, and high) and soil depths (0–10 cm, 10–20 cm, 20–30 cm), we systematically investigated vegetation traits, soil physicochemical properties, microbial community composition, functional pathways, and co-occurrence networks. Aboveground biomass, community-weighted height, and soil pH declined with altitude. Soil pH and ammonium nitrogen were significantly associated with bacterial community variation in the RDA analyses. Surface and deeper soil layers differed in bacterial taxonomic composition and predicted functional potentials. Co-occurrence network properties also varied across altitude and soil-depth groups. These findings describe spatial associations among vegetation, soil conditions, and bacterial community patterns during the sampling period. This study improves our understanding of how interacting environmental structure soil microbial communities and influence ecosystem functioning in alpine ecosystems. Full article

(This article belongs to the Section Microbiology)

24 pages, 5218 KB

Open AccessArticle

Early Detection of Pine Wilt Disease at the Pre-Visual Stage Using UAV-Based Hyperspectral Imagery

by Tianteng Zhang, Lei Feng, Botian Zhou, Yuxin Zhao, Ben Yang, Wenhua Zeng, Xinchao Li, Yucai Li and Ling Wu

Remote Sens. 2026, 18(18), 3229; https://doi.org/10.3390/rs18183229 (registering DOI) - 19 Sep 2026

Abstract

Pine wilt disease (PWD) causes forest decline, but pre-visual spectral responses are weak and spatially heterogeneous within crowns. We developed a UAV hyperspectral framework combining targeted spectral-index construction with patch-based probability aggregation to enhance spectral discrimination and preserve localised spectral responses. We retrospectively [...] Read more.

Pine wilt disease (PWD) causes forest decline, but pre-visual spectral responses are weak and spatially heterogeneous within crowns. We developed a UAV hyperspectral framework combining targeted spectral-index construction with patch-based probability aggregation to enhance spectral discrimination and preserve localised spectral responses. We retrospectively labelled 477 Masson pine crowns from two survey plots using multi-temporal observations. Pre-visual candidates remained green at image acquisition and the first follow-up but subsequently developed visible discolouration. Sensitive bands were screened with emphasis on the healthy–pre-visual and pre-visual–early boundaries, and narrow-band indices were constructed using the training partition of Dataset A. Two three-band indices, VI2 (609, 699, and 737 nm) and VI3 (699, 737, and 765 nm), were retained. A random forest trained on crown-mean index features generated class-probability responses for non-overlapping 3 × 3 patches, which were aggregated into crown-level descriptors and classified using shrinkage linear discriminant analysis. On the held-out Dataset A validation set, the framework achieved 90.80% overall accuracy, 86.81% balanced accuracy, 88.45% macro-F1, and 77.78% pre-visual F1. Pre-visual F1 exceeded the NDVI baseline by 16.67 percentage points and improved by 12.78 and 7.19 points over crown-mean classification and pixel-level aggregation, respectively. For transfer from Dataset A to Dataset B, adaptation with three labelled crowns per class improved mean OA from 78.69% to 81.98% over the zero-shot baseline across 100 trials. These results indicate that targeted spectral indices and patch-based spatial representation can capture weak and localised spectral responses for crown-level screening of pre-visual PWD candidates under the evaluated conditions. Full article

(This article belongs to the Section Forest Remote Sensing)

25 pages, 1526 KB

Open AccessArticle

DNWI: A Drinking Water Quality Index for Sustainable Urban Distribution Networks, Complemented by an Analysis of User Perception

by Sonia Gonzaga-Vallejo, Adrián Ríos-Gonzaga, Daniel Cofre-Betancourt, Oscar Guerrero-Guarnizo and Nayeli Ramón-Chalán

Sustainability 2026, 18(18), 9625; https://doi.org/10.3390/su18189625 (registering DOI) - 19 Sep 2026

Abstract

This study evaluates drinking water quality in the urban distribution network of Loja, Ecuador, using the Distribution Network Water Quality Index (DNWI), complemented by citizen perception and expert-based weighting. Two sampling campaigns were conducted under low-rainfall conditions, collecting samples from household taps across [...] Read more.

This study evaluates drinking water quality in the urban distribution network of Loja, Ecuador, using the Distribution Network Water Quality Index (DNWI), complemented by citizen perception and expert-based weighting. Two sampling campaigns were conducted under low-rainfall conditions, collecting samples from household taps across three supply systems. Physicochemical and microbiological parameters were analyzed following standard methods, and normalized sub-indices were calculated to quantify deviations from regulatory thresholds. Additionally, a perception survey (n = 1015) and a Delphi-based expert evaluation were incorporated. Results revealed marked spatial variability, with critical DNWI values observed in distal sections of the CWS network (DNWI = 44.34), mainly driven by residual chlorine depletion and microbiological contamination. Non-parametric analysis (Kruskal–Wallis, p < 0.001) confirmed significant differences in perception across systems. Spearman correlations indicated that sensory water quality parameters, particularly color, strongly influence user satisfaction. The internal consistency of the perception scale was acceptable (Cronbach’s α = 0.634). The DNWI enables the identification of critical sectors, supports risk-based management, and provides an objective assessment of service management. Its comparison with citizen perception reveals gaps between sensory acceptability and technical quality, thereby strengthening sustainability, water governance, and progress toward SDG 6. Full article

(This article belongs to the Section Sustainable Water Management)

46 pages, 793 KB

Open AccessReview

Maritime GNSS Threats and Resilience: A Layered Review of Vulnerabilities and Defences

by Dimitrios Piromalis, Efthymios Tserepas and Panagiotis Papageorgas

Appl. Syst. Innov. 2026, 9(9), 197; https://doi.org/10.3390/asi9090197 (registering DOI) - 19 Sep 2026

Abstract

Global Navigation Satellite Systems (GNSS) underpin maritime Positioning, Navigation, and Timing (PNT), supporting bridge functions such as the Automatic Identification System (AIS), the Electronic Chart Display and Information System (ECDIS), radar overlays, and track-control systems. This dependence exposes vessels to non-malicious degradation, unintentional [...] Read more.

Global Navigation Satellite Systems (GNSS) underpin maritime Positioning, Navigation, and Timing (PNT), supporting bridge functions such as the Automatic Identification System (AIS), the Electronic Chart Display and Information System (ECDIS), radar overlays, and track-control systems. This dependence exposes vessels to non-malicious degradation, unintentional interference, jamming, spoofing, and replay-based (meaconing-type) deception. Using a structured scoping review with narrative synthesis, this paper synthesises evidence on maritime GNSS threats and resilience across the receiver-to-bridge chain. The review develops a maritime-oriented taxonomy and maps disruption across the radio-frequency (RF) front-end, acquisition and tracking, navigation-estimation, and integrated-bridge layers. It examines suppression-oriented resilience, including front-end mitigation, spatial anti-jamming, and multi-sensor fusion, alongside deception-oriented detection, integrity monitoring, and authentication, including correlator-domain monitoring, navigation-engine integrity monitoring, Galileo Open Service Navigation Message Authentication (OSNMA), and bridge-level cross-checks. The analytical novelty lies in integrating disruption mechanism, security objective, evidence type, layer of action, and residual protection gap within a maritime cross-layer decision framework spanning the receiver-to-bridge chain. This framework enables resilience measures to be interpreted and compared according to what they protect and what remains unprotected, supporting complementary rather than interchangeable defences. The synthesis shows that maritime GNSS resilience extends beyond the RF receiver to navigation-solution trust and integrated-bridge safety. Full article

(This article belongs to the Special Issue Intelligent Communication, Positioning and IoT Systems for Smart Infrastructure and Industrial Applications)

22 pages, 35697 KB

Open AccessArticle

Establishment and Characterization of a Brain-Derived Cell Line from Japanese Eel (Anguilla Japonica) and Transcriptomic Analysis of Its Responses to Sex Steroid Hormones

by Yajuan Huang, Xu Yan and Songlin Chen

Animals 2026, 16(18), 2951; https://doi.org/10.3390/ani16182951 (registering DOI) - 19 Sep 2026

Abstract

The Japanese eel (Anguilla japonica) is an economically important aquaculture species, but cellular mechanisms underlying sex steroid responses remain poorly characterized. In this study, we established and characterized a continuous brain-derived cell line from Japanese eel, designated AJBC. The cells exhibited [...] Read more.

The Japanese eel (Anguilla japonica) is an economically important aquaculture species, but cellular mechanisms underlying sex steroid responses remain poorly characterized. In this study, we established and characterized a continuous brain-derived cell line from Japanese eel, designated AJBC. The cells exhibited stable fibroblast−like morphology and continuous proliferation. Molecular characterization showed high expression of col1a1, a marker associated with fibroblast−like stromal populations, whereas neuronal and glial markers (map2, gfap, and sox2) showed low expression. The neural-associated marker nestin was detectable, indicating that AJBC cells retained certain brain-associated characteristics. AJBC cells were successfully transfected with a GFP−expressing plasmid, demonstrating their potential for genetic manipulation. Transcriptome analysis identified 3212 and 3522 differentially expressed genes following 17β−estradiol (E2) and 11-ketotestosterone (11−KT) treatment, respectively. KEGG analysis revealed enrichment of PI3K−Akt and MAPK signaling pathways, as well as cell adhesion and cytoskeletal-related processes. Representative reproductive-related genes, including sox9 and foxl2, showed differential expression following sex steroid hormone treatment. E2 increased esr1 and esr2 expression, whereas 11−KT increased ar expression, confirming the steroid-responsive characteristics of AJBC cells. This brain-derived cell line provides a useful in vitro model for investigating sex steroid−responsive molecular mechanisms and neuroendocrine-related processes in Japanese eel. Full article

► Show Figures

Figure 1

40 pages, 1680 KB

Open AccessArticle

Teleoperated Quadruped Robot-Based SLAM LiDAR for Standardized Tree Stem Diameter Derivation in Spanish Mediterranean Savanna-like Woodlands: A Field Feasibility Study

by Leon Vehlken, Jan Wolf, Diana López Serrano, Marie Gröbner, Victor Rolo, Gerardo Moreno, M. Pilar Martín, Arnaud Carrara, Sung-Ching Lee, Georg Bareth and Alexander Jenal

Robotics 2026, 15(9), 176; https://doi.org/10.3390/robotics15090176 (registering DOI) - 19 Sep 2026

Abstract

Mediterranean Dehesa woodlands require monitoring of tree structural change, but conventional stem-diameter inventories are labor-intensive. Mobile simultaneous localization and mapping (SLAM) LiDAR can acquire three-dimensional tree structure, although inconsistent trajectories and uneven stem coverage may reduce measurement repeatability. This study evaluated whether a [...] Read more.

Mediterranean Dehesa woodlands require monitoring of tree structural change, but conventional stem-diameter inventories are labor-intensive. Mobile simultaneous localization and mapping (SLAM) LiDAR can acquire three-dimensional tree structure, although inconsistent trajectories and uneven stem coverage may reduce measurement repeatability. This study evaluated whether a teleoperated quadruped carrying a Hovermap ST sensor along standardized circular trajectories could acquire stem-level point clouds suitable for diameter-at-breast-height (DBH) estimation in open Dehesa terrain. Backpack-mounted surveys were conducted in 2025 at Majadas de Tiétar, followed by quadruped-mounted surveys in 2026 at two Spanish sites. DBH was derived from horizontal slices at 1.3 m using Circle Fitting, Convex Hull Peeling, and Kalman-filter reconstruction, then compared with height-matched manual references. Backpack estimates for 186 caliper-referenced observations achieved R² values of 0.853–0.869, Lin’s concordance coefficients of 0.787–0.833, and root-mean-square errors of 5.22–6.11 cm. After excluding one bifurcation-affected observation, 29 robot estimates achieved R² values of 0.930–0.943, concordance coefficients of 0.941–0.956, and root-mean-square errors of 3.17–3.84 cm. Robot acquisitions showed more balanced radial point distributions and fewer reconstruction errors. However, dataset differences preclude controlled platform comparison. Quadruped-mounted SLAM LiDAR therefore shows promise for standardized one-time DBH inventories, while paired same-tree repeatability studies remain necessary before application to multitemporal growth monitoring. Full article

(This article belongs to the Special Issue Multi-Robot Systems for Environmental Monitoring and Intervention)

23 pages, 3387 KB

Open AccessArticle

Rapid and Interpretable Wheat Seed Variety Identification Using Morphology-Guided Feature Engineering and Ensemble Learning

by Li Wang, Jingyuan Yun, Chunmei Wang, Xueqiang Gao, Jianbo Liu, Limiao Deng and Tianyu Zhu

AgriEngineering 2026, 8(9), 397; https://doi.org/10.3390/agriengineering8090397 (registering DOI) - 19 Sep 2026

Abstract

Confirming seed variety identity is important in certification, breeding-material management and grain trade, but visually similar cultivars remain difficult to distinguish consistently. We present GAFE-Stack, a morphology-guided classifier that expands seven kernel measurements into twenty-one interpretable descriptors and combines five complementary learners. The [...] Read more.

Confirming seed variety identity is important in certification, breeding-material management and grain trade, but visually similar cultivars remain difficult to distinguish consistently. We present GAFE-Stack, a morphology-guided classifier that expands seven kernel measurements into twenty-one interpretable descriptors and combines five complementary learners. The method is evaluated on the small, balanced public UCI Seeds benchmark (

b l a c k N = 210

;

b l a c k 70

kernels per variety), whose measurements were extracted from soft X-ray images. Across ten repeats of stratified five-fold cross-validation, GAFE-Stack achieved

96.33 b l a c k \pm 2.29 %

accuracy and

96.67 %

under leave-one-out validation. Its observed mean differences from seven re-implemented references ranged from

b l a c k 0.48

to

b l a c k 4.29

percentage points; after accounting for dependence among repeated folds and applying Holm adjustment, none of the comparisons was significant at

b l a c k α = 0.05

. The strongest individual member, LightGBM, achieved a slightly higher mean accuracy (

b l a c k 96.76 %

), whereas GAFE-Stack had lower fold-level dispersion (

b l a c k 2.29 %

versus

b l a c k 2.60 %

) and fewer pooled Kama–Canadian confusions than the RBF-SVM baseline. The complete pipeline achieved

b l a c k 93.33 %

and

b l a c k 94.52 %

accuracy with

b l a c k 30

and

b l a c k 60

labelled training kernels, respectively, compared with

b l a c k 96.33 %

using the full training folds. From stored morphometric inputs, single-thread CPU training required

b l a c k 0.85

s and batch prediction processed approximately

b l a c k 52, 600

kernels/s without a GPU. Applying the same dimension-typed construction rules to Raisin, Rice and Dry Bean datasets produced small positive mean changes of

b l a c k 0.09

–

b l a c k 0.37

points, with corrected intervals including zero. Lot-purity results are reported only as an exploratory resampling analysis of UCI observations. The present evidence therefore supports GAFE-Stack as an interpretable proof-of-concept approach to seed screening under benchmark conditions; validation on independently acquired kernels, measurement systems and physical seed lots remains future work. Full article

(This article belongs to the Section Computer Applications and Artificial Intelligence in Agriculture)

25 pages, 12662 KB

Open AccessArticle

Toughening PLA via Diisocyanate-Induced Dynamic Vulcanization of Carboxylated Oleic Acid: Role of Crosslinked Network Topology

by Dongmei Xie, Xiao Li, Jiaqi Cai, Xiaodi Mao, Hongzhi Liu and Ping Zhang

Polymers 2026, 18(18), 2297; https://doi.org/10.3390/polym18182297 (registering DOI) - 19 Sep 2026

Abstract

Diisocyanate-induced dynamic vulcanization of difunctional fatty acids has emerged as a universal strategy to efficiently improve the impact resistance of polylactic acid (PLA). However, how the crosslinked network topology of the in situ formed polyamide elastomer (COPA) affects its toughening efficacy on PLA [...] Read more.

Diisocyanate-induced dynamic vulcanization of difunctional fatty acids has emerged as a universal strategy to efficiently improve the impact resistance of polylactic acid (PLA). However, how the crosslinked network topology of the in situ formed polyamide elastomer (COPA) affects its toughening efficacy on PLA remains unknown. Here, we synthesized two carboxylated fatty acids with distinct molecular architectures (TCOA and NCOA) from technical-grade and high-purity oleic acids via UV-initiated thiol-ene click chemistry. These two diacids, along with tetradecanedioic acid (TA) without a dangling chain, were dynamically vulcanized with hexamethylene diisocyanate (HDI) to toughen PLA. By varying NCO/COOH molar ratios between HDI and TCOA, their effects on gel content, crosslinking density, phase morphology, and mechanical properties of resulting blends were systematically investigated. With increasing the ratio, both gel content and interfacial adhesion with PLA in the blends were enhanced, accompanied by the transformation of phase structure from “sea-island” morphology to a partially or fully co-continuous one. At a molar ratio of 1.8:1, the notched impact strength value of the blend reached 86.5 kJ/m². By substituting TCOA with high-purity NCOA or TA, comparable gel content and interfacial compatibilization level, and co-continuous morphologies were achieved. Notably, NCOA yielded a PLA blend with a remarkably higher impact toughness (131.0 kJ/m²). The linear chain structure of TA led to a higher crosslinking density of the formed TAPA domains, which unfavorably suppressed their cavitation during impact fracture and thus resulted in an inferior impact strength (7.0 kJ/m²). These findings provide valuable insights into the toughening mechanism of PLA via dynamic vulcanization of monomers. Full article

(This article belongs to the Special Issue Advanced Poly(lactic Acid) Materials and PLA-Based Composites: From Fundamentals to Applications)

30 pages, 3786 KB

Open AccessReview

The Role of Diet in Functional Dyspepsia

by Luisa Bertin, Cedric Van de Bruaene, Elena Formisano, Marcella Pesce, Stefania Piccirelli, Daniele Salvi, Karen Routhiaux, Andrea Pasta, Francesco Calabrese, Federico Caldart, Salvatore Crucillà, Giovanni Sarnelli, Elisa Marabotto, Javier Chahuan, Tom van Gils and Edoardo Vincenzo Savarino

Nutrients 2026, 18(18), 3064; https://doi.org/10.3390/nu18183064 (registering DOI) - 19 Sep 2026

Abstract

Functional dyspepsia (FD) is a disorder of gut–brain interaction affecting approximately 7% of the global population under Rome IV criteria, characterised by postprandial fullness, early satiation, epigastric pain, and/or epigastric burning in the absence of identifiable organic disease. Meal ingestion is the dominant [...] Read more.

Functional dyspepsia (FD) is a disorder of gut–brain interaction affecting approximately 7% of the global population under Rome IV criteria, characterised by postprandial fullness, early satiation, epigastric pain, and/or epigastric burning in the absence of identifiable organic disease. Meal ingestion is the dominant symptom trigger in the majority of patients, yet the relationship between diet and FD is bidirectional: dietary exposures drive symptoms through established pathophysiological mechanisms, while FD itself can promote maladaptive eating behaviours and progressive dietary restriction with nutritional and psychological consequences. The pathophysiological substrate linking diet to symptom generation encompasses impaired gastric accommodation and emptying, visceral hypersensitivity to mechanical and chemical stimuli, low-grade duodenal mucosal immune activation and epithelial barrier impairment, alterations of the small intestinal microbiota, and gut–brain axis dysregulation. Dietary fat is the macronutrient most consistently implicated, acting through cholecystokinin-mediated chemosensory and mechanosensory pathways. Cross-sectional population data suggest that a high protein intake may also be associated with epigastric pain, although this association did not persist after multivariable adjustment, while the evidence linking total carbohydrate intake to symptoms remains sparse and inconsistent, so that no recommendation on total carbohydrate intake can currently be made. Cognitive factors, including learned food beliefs, expectancy, and nocebo effects, further modulate symptom perception independently of actual nutrient content. Among dietary patterns, Mediterranean-style eating offers the most coherent mechanistic rationale and the strongest observational support, while a diet low in fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAP) shows selective benefit in patients with postprandial distress syndrome features or prominent bloating. Common triggers including spicy foods, coffee, and alcohol show heterogeneous evidence driven by individual sensitivity rather than dose-related effects. Nutritional concerns are substantial: clinically significant weight loss affects approximately 35% of tertiary-care patients, and nearly 40% screen positive for avoidant/restrictive food intake behaviour using validated screening tools. Dietary management should therefore be individualised, nutritionally adequate, and integrated within a broader biopsychosocial framework, with screening for disordered eating before any elimination strategy is initiated. Full article

(This article belongs to the Special Issue Nutrition in Neurogastroenterology)

25 pages, 508 KB

Open AccessArticle

Hybrid Graph Retrieval-Augmented Language Agents for Collaborative Recommendation

by Ivan Bulychev and Andrey Savchenko

AI 2026, 7(9), 380; https://doi.org/10.3390/ai7090380 (registering DOI) - 19 Sep 2026

Abstract

Recent advances in large language model (LLM) agents have shown promise for autonomous decision-making in recommender systems. However, existing approaches suffer from two fundamental limitations: flat agent memories that conflate different information modalities and prohibitive computational costs that prevent scaling beyond a few [...] Read more.

Recent advances in large language model (LLM) agents have shown promise for autonomous decision-making in recommender systems. However, existing approaches suffer from two fundamental limitations: flat agent memories that conflate different information modalities and prohibitive computational costs that prevent scaling beyond a few hundred users. We propose Hybrid-GraphRAG, a recommender system that integrates hierarchical agent memory structures, graph-based retrieval-augmented generation (Graph RAG), and knowledge distillation for scalable deployment. Our approach extends agent-based collaborative filtering by structuring agent memories into intrinsic, collaborative, and interaction tiers that disentangle different information types; performing multi-hop retrieval over a dynamically constructed heterogeneous interaction graph to enable relational reasoning; and distilling LLM-generated memory dynamics into efficient graph neural encoders with adaptive gating between full and efficient inference paths. Experiments on Amazon review datasets (CDs and Vinyl, Office Products) demonstrate that Hybrid-GraphRAG achieves recommendation quality comparable to full LLM-based agents while reducing computational cost by 85% and improving NDCG@10 by 12.7% over flat-memory agent baselines. Our results establish a principled bridge between semantic agent reasoning and scalable graph-based recommendation. Full article

(This article belongs to the Topic AI Agents: Progress, Architecture, and Applications)

► Show Figures

Figure 1

19 pages, 8106 KB

Open AccessArticle

Effects of Wood Vinegar Application and Planting Density on Cadmium Uptake and Accumulation in a Maize–Wheat Rotation System Under Straw Removal Management

by Jing Duo, Ying Zhao, Yuhao Liu, Jiarun Ye, Zhenzhu Cao, Haitao Liu, Shiliang Liu, Guiying Jiang and Fang Liu

Agronomy 2026, 16(18), 1851; https://doi.org/10.3390/agronomy16181851 (registering DOI) - 19 Sep 2026

Abstract

Cadmium (Cd) contamination in agricultural soils threatens food safety and human health. To address this issue, wood vinegar (WV) and high-density planting were combined in our field study. The results revealed that the combined application significantly increased maize (Zea mays) grain [...] Read more.

Cadmium (Cd) contamination in agricultural soils threatens food safety and human health. To address this issue, wood vinegar (WV) and high-density planting were combined in our field study. The results revealed that the combined application significantly increased maize (Zea mays) grain and straw biomass while enhancing Cd accumulation in aboveground tissues. Notably, grain Cd concentrations remained relatively low, ranging from 0.032 to 0.047 mg·kg⁻¹, whereas the Cd enrichment coefficients in stems and leaves rose by 70.03% and 67.42%, respectively. Additionally, WV application altered soil Cd speciation and reduced the exchangeable Cd fraction. The ZT2 treatment reduced the exchangeable Cd fraction by 24.56%, while the ZT3 treatment reduced total soil Cd by 6.91%, indicating its potential to enhance Cd removal. In the subsequent wheat (Triticum aestivum) season, maize straw removal combined with high-density planting considerably decreased Cd concentrations across wheat tissues, with maximum reductions of 25.59% in grain and 45.63% in straw. However, wheat grain Cd concentrations remained above the applicable food safety limit. These management practices were also associated with shifts in the soil microbial community, including changes in the relative abundance of Actinobacteria and Acidobacteria. Partial least squares path modeling (PLS-PM) indicated that Cd accumulation in maize straw showed the strongest negative association with Cd concentrations in wheat grain. Overall, integrating WV application with high-density planting represents a promising agronomic strategy for enhancing Cd phytoextraction by maize and mitigating Cd accumulation in subsequent wheat. Full article

(This article belongs to the Section Soil and Plant Nutrition)

► Show Figures

Figure 1

24 pages, 4652 KB

Open AccessArticle

An Unbiased Drug Screen in a Drosophila Model of LMNA-Muscular Dystrophy Identifies Calcium Channel Blockers as Potential Treatments

by Nathaniel P. Mohar, Brenna A. Powers, Benjamin E. Hinz, Alex R. Bock, Hailey McCoy-Munger, Zachary Darr, Maxwell W. Shumaker, Alexander T. Wemmie, Amya Saxena and Lori L. Wallrath

Int. J. Mol. Sci. 2026, 27(18), 8367; https://doi.org/10.3390/ijms27188367 (registering DOI) - 19 Sep 2026

Abstract

Mutations in the LMNA gene encoding A-type lamins cause multiple muscular dystrophies (LMNA-MD). Lamins are intermediate filaments with three conserved domains: an N-terminal head, a coiled-coil rod, and a C-terminal domain possessing an Ig-like fold. Missense mutations in all three domains [...] Read more.

Mutations in the LMNA gene encoding A-type lamins cause multiple muscular dystrophies (LMNA-MD). Lamins are intermediate filaments with three conserved domains: an N-terminal head, a coiled-coil rod, and a C-terminal domain possessing an Ig-like fold. Missense mutations in all three domains can cause LMNA-MD by mechanisms that are not well understood. Currently, there are limited treatments for LMNA-MD beyond symptom management. In this study, our goal was to identify candidate treatments through an in vivo drug repurposing screen using Drosophila models of LMNA-MD that recapitulate aspects of the disease phenotype. Expression of LamC R264Q (Drosophila orthologue of human LMNA R249Q) in fly muscles causes muscle defects and premature death. We fed 1520, mostly FDA/EMA-approved, drugs to larvae and screened for rescue of lethality. This screen resulted in 68 positive hit drugs that partially restored viability. Molecular fingerprinting clustered a subset of these compounds into six structurally related groups, including L-type voltage-gated calcium channel inhibitors. We tested these inhibitors using multiple fly models expressing different mutant lamins and discovered that only flies expressing amino acid substitutions in the rod domain benefited, supporting a need for personalized treatments. Collectively, our findings support L-type voltage-gated calcium channel blockers as candidate treatments that warrant testing in pre-clinical models. Full article

(This article belongs to the Special Issue Molecular Mechanisms and Therapies in Skeletal Muscle Diseases)

► Show Figures

Figure 1

25 pages, 1099 KB

Open AccessReview

Microstructured Optical Sensors: Design, Fabrication, and Applications

by Victor Argueta-Diaz

Photonics 2026, 13(9), 889; https://doi.org/10.3390/photonics13090889 (registering DOI) - 19 Sep 2026

Abstract

Microstructured optical sensors (MOS) have emerged as transformative photonic devices offering unprecedented sensitivity, miniaturization, and multiplexing capability. This review synthesizes the state of the field for specialists in photonics and optics, covering principal design architectures—photonic crystal fibers, microresonators, plasmonic nanostructures, and integrated waveguide [...] Read more.

Microstructured optical sensors (MOS) have emerged as transformative photonic devices offering unprecedented sensitivity, miniaturization, and multiplexing capability. This review synthesizes the state of the field for specialists in photonics and optics, covering principal design architectures—photonic crystal fibers, microresonators, plasmonic nanostructures, and integrated waveguide platforms—alongside the fabrication methodologies underpinning their realization. We examine four high-impact application domains: environmental monitoring, biomedical diagnostics, structural health monitoring, and food safety. A central argument runs through the review: the dominant bottleneck constraining MOS adoption has shifted from sensitivity—where many platforms now approach the physical detection limit—to the engineering triad of packaging difficulty, calibration drift, and manufacturing reproducibility. Against this backdrop, we provide frank comparative assessments of platform readiness for each application domain, distinguishing proof-of-concept demonstrations from deployable systems. A dedicated Critical Perspective section addresses the translation of laboratory sensitivity figures to field performance, the genuine and overstated contributions of machine learning, and a commercialization-readiness assessment for principal MOS platform families. Full article

(This article belongs to the Special Issue Microstructured Optical Sensors: Design, Fabrication and Applications)

44 pages, 1458 KB

Open AccessFeature PaperArticle

A Multiscale Approach to Foodborne Disease Dynamics: Unifying Within-Host Scale and Environmental Pathogen Load Using Escherichia coli O157:H7 as a Case Study

by Azwindini Delinah Maphiri, Kizito Muzhinji, Dephney Mathebula and Rendani Netshikweta

Mathematics 2026, 14(18), 3403; https://doi.org/10.3390/math14183403 (registering DOI) - 19 Sep 2026

Abstract

Foodborne diseases remain a major public health concern because of their high incidence and the substantial number of deaths they cause worldwide. Consequently, there is a need to develop multiscale mathematical models that improve our understanding of the biological mechanisms driving disease transmission [...] Read more.

Foodborne diseases remain a major public health concern because of their high incidence and the substantial number of deaths they cause worldwide. Consequently, there is a need to develop multiscale mathematical models that improve our understanding of the biological mechanisms driving disease transmission and identify key factors influencing disease persistence. Traditional mathematical models typically investigate within-host and between-host dynamics separately, overlooking the interactions between processes occurring at these two scales. This study develops and analyses a nested multiscale mathematical model for Escherichia coli O157:H7 that integrates within-host bacterial replication with between-host environmental transmission. The two scales are explicitly coupled through pathogen shedding, allowing the influence of within-host bacterial dynamics on environmental contamination and disease transmission to be quantified. A nonstandard finite difference (NSFD) scheme is constructed to preserve the essential dynamical properties of the continuous model, and the resulting discrete model is analysed mathematically. Numerical simulations demonstrate that once E. coli O157:H7 becomes established within infected cattle, continued bacterial replication and pathogen shedding sustain environmental contamination, thereby promoting the persistence of disease transmission within the cattle population. These findings highlight the important role of within-host bacterial dynamics in driving environmental transmission and provide a quantitative framework for evaluating intervention strategies targeting both within-host bacterial replication and between-host disease transmission. Full article

(This article belongs to the Special Issue Mathematical Modeling and Computation in Systems Biology)

16 pages, 3950 KB

Open AccessFeature PaperArticle

Effects of Repeated Monopolar Pulsed Capacitive Radiofrequency Energy Treatments on Mechanical Nociceptive Thresholds and Skin Temperature in Show-Jumping Horses with Thoracolumbar Pain

by Benedetta Brenna, Maria Chiara Pressanto, Marco Pepe, Bruno Boccioli, Stefanie Rouge and Francesca Beccati

Animals 2026, 16(18), 2943; https://doi.org/10.3390/ani16182943 (registering DOI) - 19 Sep 2026

Abstract

This study evaluated the effects of repeated monopolar Capacitive Resistive Electrical Transfer (CRET) therapy on mechanical nociceptive thresholds (MNT) and skin surface temperature in horses with perceived chronic thoracolumbar pain. Fifteen show-jumping horses were enrolled in a randomized sham-controlled clinical trial. Ten horses [...] Read more.

This study evaluated the effects of repeated monopolar Capacitive Resistive Electrical Transfer (CRET) therapy on mechanical nociceptive thresholds (MNT) and skin surface temperature in horses with perceived chronic thoracolumbar pain. Fifteen show-jumping horses were enrolled in a randomized sham-controlled clinical trial. Ten horses received treatment, and five underwent a sham procedure. Five treatment sessions were performed at 72 h intervals using a monopolar capacitive radiofrequency device operating at 500 kHz. Mechanical nociceptive thresholds (MNT) were assessed before treatment, after each session, and during a 45-day follow-up period. Infrared thermography was performed before and after each treatment session. MNT data were analyzed using linear mixed-effects models including treatment group, time, anatomical site and their interactions as fixed effects. CRET treatment resulted in a progressive increase in MNT, with significant effects of treatment group, time and their interaction (p < 0.0001). Higher ΔMNT values were observed in treated horses from the second treatment session onwards and remained above baseline throughout the follow-up period. Skin surface temperature increased after both CRET and sham procedures, with no significant difference between groups. Repeated monopolar capacitive CRET increased MNT in show-jumping horses with perceived chronic thoracolumbar pain, while producing no greater immediate thermographic response. Full article

(This article belongs to the Special Issue Equine Physiotherapy and Functional Recovery: Advances in Diagnostics and Therapeutic Approaches)

► Show Figures

Figure 1

20 pages, 7729 KB

Open AccessArticle

Comparative Study on Taste and Odor Flavor Compounds of Chrysanthemums from Five Different Varieties Using UPLC-Q-TOF-MS, GC-MS, E-Tongue, E-Nose, and Chemometrics

by Guizhou Hu, Xinqi Li, Geng Qin, Ailing Ben and Jin Qi

Foods 2026, 15(18), 3328; https://doi.org/10.3390/foods15183328 (registering DOI) - 19 Sep 2026

Abstract

A practical strategy was established for distinguishing five varieties of Chrysanthemums by combining E-tongue, E-nose, GC-MS, and LC-Q-TOF-MS. A total of 19 low molecule fraction compounds and 26 volatile fraction compounds were identified as the common peaks. Chemometrics were used to screen the [...] Read more.

A practical strategy was established for distinguishing five varieties of Chrysanthemums by combining E-tongue, E-nose, GC-MS, and LC-Q-TOF-MS. A total of 19 low molecule fraction compounds and 26 volatile fraction compounds were identified as the common peaks. Chemometrics were used to screen the characteristic compounds of the five varieties of Chrysanthemums. E-tongue and E-nose were used to distinguish the taste and odor of the five varieties of Chrysanthemum, respectively. The Hangju variety had the richest taste flavor and the weakest aroma, while the Huaiju variety had the strongest odor of pine, woody, floral, and fruity aroma. The study found that 3, 5-dicaffeoylquinic acid, 1, 5-dicaffeoylquinic acid, luteolin-7-β-D-glucoside, and 4, 5-dicaffeoylquinic acid were the principal taste flavor components based on Gray Relational Analysis (GRA) and Partial Least Squares Regression (PLSR). α-pinene, camphene, γ-terpinene, camphor, etc. were identified as the principal volatile flavor components based on Spearman’s rank correlation. The combination of chemometrics and intelligent sensory technology presents a potential technique for analyzing the flavor of chrysanthemums and other food products. Full article

(This article belongs to the Section Food Analytical Methods)

21 pages, 56304 KB

Open AccessArticle

Biomechanical Effects of Horizontal, Vertical, and Combined Misfits in Full-Arch Implant-Supported Titanium Frameworks: A Three-Dimensional Finite Element Analysis

by Hale Arikan Kalayci and Mustafa Baris Guncu

J. Funct. Biomater. 2026, 17(9), 478; https://doi.org/10.3390/jfb17090478 (registering DOI) - 19 Sep 2026

Abstract

This study evaluated the effects of misfit type, magnitude, and location on the stress distribution in full-arch screw-retained implant frameworks. A maxillary finite element model with four implants was analyzed under 16 scenarios, characterized by horizontal misfits of 10, 50, 100, and 200 [...] Read more.

This study evaluated the effects of misfit type, magnitude, and location on the stress distribution in full-arch screw-retained implant frameworks. A maxillary finite element model with four implants was analyzed under 16 scenarios, characterized by horizontal misfits of 10, 50, 100, and 200 µm; vertical misfits of 10 and 100 µm; and combined 10–10 and 100–100 µm misfits, each positioned anteriorly or posteriorly. Forced seating was simulated using prescribed displacement; no occlusal load or screw preload was applied. Von Mises stresses were evaluated in the framework, occlusal screws, and implants, and principal stresses were assessed in peri-implant bone. Framework stress increased with horizontal misfit magnitude. When the same numerical misfit value (10 or 100 µm) was applied at the same location under otherwise identical model conditions, vertical misfit produced higher framework stress than horizontal misfit, indicating a direction-dependent response associated with different seating deformation modes. The posterior 100–100 µm combined misfit produced the maximum framework (356 MPa), occlusal screw (193 MPa), and implant (250 MPa) stresses. In contrast, the maximum principal stress peaked at the posterior bone site with the anterior 200 µm horizontal misfit (V0 H200 A; 77 MPa), while the most negative minimum principal stress occurred at the posterior bone site with the posterior 200 µm horizontal misfit (V0 H200 P; −52 MPa). Among the tested scenarios, combined misfits yielded the highest framework and screw stresses, whereas implant and bone responses depended on the magnitude and location of the misfit. These findings indicate that the types, magnitudes, and locations of misfits should be considered during framework-fit assessments, with particular attention to combined misfits, before definitive seating. Full article

(This article belongs to the Section Dental Biomaterials)

► Show Figures

Figure 1

22 pages, 2636 KB

Open AccessArticle

Assessment of Soil Injection Treatments in Street Tree Beds: Impacts on Soil Microclimate and Foliar NDVI

by Jindřiška Jeřábková, Viera Paganová, Jana Burgová, Petr Salaš and Patrik Burg

Forests 2026, 17(9), 1120; https://doi.org/10.3390/f17091120 (registering DOI) - 19 Sep 2026

Abstract

This study evaluated the effects of soil injection treatments, including aeration, hydrogel application, and mycorrhizal inoculation, on the soil hydrothermal regime and foliar NDVI of young street trees (Prunus serrulata) under real urban conditions. The experiment was conducted during three growing [...] Read more.

This study evaluated the effects of soil injection treatments, including aeration, hydrogel application, and mycorrhizal inoculation, on the soil hydrothermal regime and foliar NDVI of young street trees (Prunus serrulata) under real urban conditions. The experiment was conducted during three growing seasons (2023–2025) at two sites with contrasting soil structural properties. Soil moisture, temperature, and leaf-level NDVI were monitored to assess treatment effects and temporal variability. Soil moisture responses to the treatments were site-dependent, with Aeration and the combined Mycorrhiza + Hydrogel treatment showing contrasting responses between the two sites. Aeration showed site-dependent effects, improving soil moisture conditions in compacted clay soils, but providing limited benefits in coarse-textured soils. The amendment treatments, all of which included Aeration, showed variable responses relative to Aeration alone. Soil temperature responses were less consistent, with no significant effects of Site, Treatment or their interaction. Differences in NDVI associated with the combined Mycorrhiza + Hydrogel treatment were detected during the following two growing seasons. The results highlight that soil injection treatments can contribute to urban tree management; however, their effects are site-dependent and should be evaluated through long-term monitoring. Full article

(This article belongs to the Special Issue Urban Forests and Greening for Sustainable Cities, 2nd Edition)

18 pages, 2275 KB

Open AccessArticle

TOCNF/Nisin-Stabilized Oregano Oil Pickering Emulsions Enhance Chitosan Quaternary Ammonium Salt/Propolis Films for Green Cherry Tomato Preservation

by Keying Hou, Shengsi Hu, Chenfeng Yu, Kang Tu, Xiaodong Zheng, Ye Song and Leiqing Pan

Foods 2026, 15(18), 3327; https://doi.org/10.3390/foods15183327 (registering DOI) - 19 Sep 2026

Abstract

Essential oils are difficult to disperse in hydrophilic films and may separate or volatilize during processing. We prepared chitosan quaternary ammonium salt/propolis films containing a TEMPO-oxidized cellulose nanofibril (TOCNF)/nisin-stabilized Pickering emulsion loaded with oregano essential oil. Two independent formulation variables were examined: the [...] Read more.

Essential oils are difficult to disperse in hydrophilic films and may separate or volatilize during processing. We prepared chitosan quaternary ammonium salt/propolis films containing a TEMPO-oxidized cellulose nanofibril (TOCNF)/nisin-stabilized Pickering emulsion loaded with oregano essential oil. Two independent formulation variables were examined: the nisin fraction in the TOCNF/nisin particles (0–35% of particle mass) was first screened to select the emulsion stabilizer composition, whereas the selected TNO emulsion was subsequently added at 20%, 30%, 40%, or 50% (v/v) of the initial film-forming solution. At 25% nisin, encapsulation efficiency reached 88.0%, and the polydispersity index was 0.15, so this formulation was used in the films. As emulsion loading increased, the films became lighter and opacity fell from 19.75 to 4.81 A mm⁻¹. HCP/TNO30 reached a tensile strength of 45.18 MPa, but elongation at break decreased from 65.35% to 28.11% across the series. Water vapor and oxygen permeability were lowest at 40% loading. HCP/TNO50 had the highest radical-scavenging and antibacterial activities. It also reduced weight loss and quality deterioration during 12 days of green cherry tomato storage. No loading optimized every property. Mechanical strength and barrier performance peaked at 30% and 40%, respectively, while active preservation continued to improve up to 50%. Full article

(This article belongs to the Special Issue Recent Research on the Processing Technology and Quality Control of Fruit and Vegetables)

34 pages, 8078 KB

Open AccessArticle

RS-CARES: Context-Aware Cross-Modal Alignment with Semantic Spatial Prior for Referring Remote Sensing Image Segmentation

by Hui Xiong, Wen Luo and Bing He

J. Imaging 2026, 12(9), 453; https://doi.org/10.3390/jimaging12090453 (registering DOI) - 19 Sep 2026

Abstract

Remote sensing referring image segmentation faces critical challenges including arbitrary target rotation, drastic scale variation, cluttered complex backgrounds, and large visual-language semantic gaps. Existing mainstream segmentation models adopt fixed-receptive-field backbones, coarse unidirectional cross-modal interaction and static learnable object queries, which easily cause small-object [...] Read more.

Remote sensing referring image segmentation faces critical challenges including arbitrary target rotation, drastic scale variation, cluttered complex backgrounds, and large visual-language semantic gaps. Existing mainstream segmentation models adopt fixed-receptive-field backbones, coarse unidirectional cross-modal interaction and static learnable object queries, which easily cause small-object missed detection, blurred boundary segmentation and fragmented predictions. This paper proposes a context-aware referring expression segmentation model for remote sensing to tackle the above limitations. We construct a dual-stream feature extraction backbone with InternImage and CLIP text encoder, and design a semantic prior localization map module to generate spatial heatmaps for spatial inductive bias and improve small-object localization recall. A cross-modal context aggregator performs multi-scale bidirectional visual-text alignment, while a dynamic query initialization strategy and a language-guided Transformer decoder progressively improve target localization and mask refinement. Experiments are conducted on two standard RRSIS benchmarks, RefSegRS and RRSIS-D. On RefSegRS, the proposed model achieves an mIoU of 70.43% and an oIoU of 77.34%, and obtains significant improvements on small vehicles, buildings and slender road markings. On the larger-scale RRSIS-D dataset, our model achieves an mIoU of 63.71% and an oIoU of 73.68%. The proposed model provides a solution for accurate multi-scale target segmentation guided by natural language descriptions in complex remote sensing scenes. Full article

(This article belongs to the Special Issue Unified and Multimodal Segmentation: Foundations, Methods, and Applications)

► Show Figures

Figure 1

33 pages, 6390 KB

Open AccessArticle

Graphene Oxide-Doped Polycarbonate Track-Etched Membranes Grafted with Poly(acrylic acid) as Gel Polymer Electrolytes for All-Solid-State Supercapacitors

by Alisher M. Zhumabayev, Semiha Duygu Sutekin, Alimzhan A. Almanov, Anastassiya A. Mashentseva, Igor A. Ivanov, Saeideh Alipoori and Murat Barsbay

Batteries 2026, 12(9), 376; https://doi.org/10.3390/batteries12090376 (registering DOI) - 19 Sep 2026

Abstract

Track-etched polycarbonate membranes (PC-TeMs) offer a structurally supported and well-defined nanoporous architecture, but their intrinsically low ionic conductivity limits their direct application as solid-state electrolytes in supercapacitors. Here, we grafted poly(acrylic acid) (PAA) into the cylindrical nanopores of PC-TeMs via reversible addition–fragmentation chain-transfer [...] Read more.

Track-etched polycarbonate membranes (PC-TeMs) offer a structurally supported and well-defined nanoporous architecture, but their intrinsically low ionic conductivity limits their direct application as solid-state electrolytes in supercapacitors. Here, we grafted poly(acrylic acid) (PAA) into the cylindrical nanopores of PC-TeMs via reversible addition–fragmentation chain-transfer (RAFT) polymerization, incorporating graphene oxide (GO) at 0–1 wt% to enhance ion transport and structural stability. Structural, spectroscopic, and morphological analyses indicated effective PAA grafting along the nanochannels and homogeneous GO dispersion within the membrane matrix. Electrochemical tests revealed a non-linear dependence of performance on GO content: an optimal GO content of 0.5 wt% with PAA grafting delivered the highest ionic conductivity (0.57 mS·cm⁻¹) and specific capacitance (1.14 F·g⁻¹ at 5 mV·s⁻¹). This corresponds to improvements of approximately ~6× and ~800× compared to the ungrafted GO-free membrane (0.09 mS·cm⁻¹ and 1.42 mF·g⁻¹), and ~3× and ~42× compared to the ungrafted GO-containing membrane (0.18 mS·cm⁻¹ and 26.96 mF·g⁻¹), respectively. In contrast, at a GO content of 1 wt%, pore blockage and a decline in performance were observed. These results demonstrate that finely tuned GO incorporation into RAFT-grafted PAA/PC-TeMs enables dimensionally stable solid-state electrolytes for high-performance solid-state supercapacitors. Full article

15 pages, 2947 KB

Open AccessArticle

Selected Physical and Mechanical Properties of Wood of Robinia pseudoacacia L. Across Hungarian Counties

by Róbert Németh, Mátyás Báder and Fath Alrhman A. A. Younis

Forests 2026, 17(9), 1119; https://doi.org/10.3390/f17091119 (registering DOI) - 19 Sep 2026

Abstract

Robinia pseudoacacia L. wood is globally recognized for its high density, superior mechanical performance and low dimensional changes. Although widely cultivated across Hungary under diverse environmental conditions, potential regional variations in its wood properties remain insufficiently characterized. This study aimed to investigate the [...] Read more.

Robinia pseudoacacia L. wood is globally recognized for its high density, superior mechanical performance and low dimensional changes. Although widely cultivated across Hungary under diverse environmental conditions, potential regional variations in its wood properties remain insufficiently characterized. This study aimed to investigate the density, linear and volumetric shrinkage and swelling, modulus of rupture (MOR), bending modulus of elasticity (MOE) and compressive strength parallel to the grain (CS) of Hungarian R. pseudoacacia samples across five counties and under three growth conditions. Results indicated significant regional variations (p < 0.05) across some physical and mechanical parameters studied. The mixed stand in poor growth conditions showed significantly higher linear and volumetric shrinkage than pure stands under good and poor growth conditions. Also, the results revealed that samples from Bács-Kiskun county showed the highest density (0.84 g/cm³) and MOE (13.33 GPa). However, samples from poor growth conditions revealed the greatest density (0.81 g/cm³), MOR (131 MPa), and CS (71.90 MPa). These findings suggest that regional site conditions and environmental stressors play a critical role in defining the quality of R. pseudoacacia wood, offering valuable insights for localized forest management and enabling targeted site selection for high-quality timber production. Full article

(This article belongs to the Section Wood Science and Forest Products)

21 pages, 546 KB

Open AccessArticle

VQ-CycleDiffusion: Vector Quantized Cycle-Consistent Diffusion Models for Voice Conversion

by Dongsuk Yook, Semin Kim and Hyung-Pil Chang

Appl. Sci. 2026, 16(18), 9314; https://doi.org/10.3390/app16189314 (registering DOI) - 19 Sep 2026

Abstract

This paper proposes a novel hybrid voice conversion framework, termed VQ-CycleDiffusion, which integrates vector quantization (VQ) with CycleDiffusion to address the inherent limitations of continuous latent representations in diffusion-based models. While conventional diffusion-based voice conversion (VC) models provide superior speech quality, they often [...] Read more.

This paper proposes a novel hybrid voice conversion framework, termed VQ-CycleDiffusion, which integrates vector quantization (VQ) with CycleDiffusion to address the inherent limitations of continuous latent representations in diffusion-based models. While conventional diffusion-based voice conversion (VC) models provide superior speech quality, they often suffer from speaker information leakage due to their reliance on continuous latent spaces, which fail to completely separate linguistic content from source speaker characteristics. To overcome this issue, we introduce a discrete bottleneck via VQ to extract multi-speaker linguistic representations, which are then used as conditional inputs for CycleDiffusion after a carefully designed codeword transformation between the source and target speakers. Experimental results on the VCTK corpus demonstrate that the proposed model significantly outperforms the conventional method in terms of speaker similarity, as confirmed by both i-vector and x-vector cosine similarity metrics, while maintaining spectral reconstruction performance comparable to that of the baseline model, as evidenced by stable Mel-cepstral distance (MCD) values. Furthermore, the quality of the converted speech was not degraded, as demonstrated by the predicted mean opinion score (MOS) evaluation. Ablation studies reveal that initializing the codebook using k-means clustering and focusing on diffusion model fine-tuning play key roles in maximizing performance. These results validate that the proposed method effectively disentangles linguistic content from speaker characteristics while preserving the high-fidelity generation capability of diffusion models. Full article

(This article belongs to the Section Computing and Artificial Intelligence)

Recent Articles

Open Access Journals

Highly Accessed Articles

Latest Books

News

Blog Posts

Topics

Selected Special Issues

Selected Collections

Journals by Subject

Further Information

Guidelines

MDPI Initiatives

Follow MDPI