MDPI - Publisher of Open Access Journals

19 pages, 2836 KB

Open AccessArticle

HL7 FHIR-Based Open-Source Framework for Real-Time Biomedical Signal Acquisition and IoMT Interoperability

by Felix-Constantin Adochiei, Florian-Alexandru Țoi, Ioana-Raluca Adochiei, Florin Ciprian Argatu, George Serițan and Gladiola-Gabriela Petroiu

Appl. Sci. 2025, 15(23), 12803; https://doi.org/10.3390/app152312803 (registering DOI) - 3 Dec 2025

Abstract

This study presents the design and validation of an open-source framework for biomedical signal acquisition and interoperable data exchange based on the Health Level Seven—Fast Healthcare Interoperability Resources (HL7 FHIR) standard. The proposed system enables secure, wireless transmission of physiological data from distributed [...] Read more.

This study presents the design and validation of an open-source framework for biomedical signal acquisition and interoperable data exchange based on the Health Level Seven—Fast Healthcare Interoperability Resources (HL7 FHIR) standard. The proposed system enables secure, wireless transmission of physiological data from distributed sensing nodes toward a locally hosted monitoring platform. The hardware architecture integrates ESP32-WROOM-32 microcontrollers for multi-parameter acquisition, the MQTT protocol for low-latency communication, and a Home Assistant (Nabu Casa, San Diego, CA, USA)–InfluxDB (InfluxData, San Francisco, CA, USA)–Grafana (Grafana Labs, New York, NY, USA) stack for real-time visualization. The novelty of this work lies in the full-stack implementation of HL7 FHIR Observations within a reproducible, open-source environment, ensuring semantic interoperability without reliance on proprietary middleware or cloud services. A case study involving multi-sensor acquisition of electrocardiographic (ECG), photoplethysmographic (PPG), temperature, and oxygen saturation signals was conducted to evaluate system performance. Validation results confirmed consistent end-to-end data flow, sub-second latency, zero packet loss, and accurate semantic preservation across all processing stages. These findings demonstrate the feasibility of implementing standardized, open, and scalable biomedical Internet of Medical Things (IoMT) systems using non-proprietary components. The proposed framework provides a reproducible foundation for future telemedicine and continuous patient-monitoring applications, aligning with FAIR data principles and the ongoing digital transformation of healthcare. Full article

(This article belongs to the Special Issue Evolutionary Computation in Biomedical Signal Processing)

► Show Figures

Figure 1

31 pages, 16657 KB

Open AccessArticle

Research on the Dynamic Characteristics of a New Bridge-and-Station Integrated Elevated Structure

by Kaijian Hu, Xiaojing Sun, Ruoteng Yang, Rui Han and Meng Ma

Vibration 2025, 8(4), 76; https://doi.org/10.3390/vibration8040076 (registering DOI) - 3 Dec 2025

Abstract

Elevated stations are essential auxiliary structures within the high-speed rail (HSR) network. The newly constructed integrated elevated station for bridge building possesses a distinctive construction and intricate force transmission pathways, complicating the assessment of the dynamic coupling of train vibrations. Consequently, it is [...] Read more.

Elevated stations are essential auxiliary structures within the high-speed rail (HSR) network. The newly constructed integrated elevated station for bridge building possesses a distinctive construction and intricate force transmission pathways, complicating the assessment of the dynamic coupling of train vibrations. Consequently, it is essential to examine the dynamic reaction of trains at such stations. This study utilises numerical simulation and field measurement techniques to examine the dynamic features of the newly constructed integrated elevated station for bridge building. Initially, vibration tests were performed on existing integrated elevated stations for bridge construction to assess their dynamic properties. The collected data were utilised to validate the modelling approach and parameter selection for the numerical model of existing stations, yielding a numerical solution method appropriate for bridge-station integrated stations. Secondly, utilising this technology, a numerical model of the newly integrated elevated station for bridge construction was developed to examine its dynamic features. Moreover, the impact of spatial configuration, train velocity, and operational organisation on the dynamic characteristics was analysed in greater depth. The vibration response level in the waiting hall was assessed. Research results indicate that structural joints alter the transmission path of train vibration energy, thereby significantly affecting the vibration characteristics of the station. The vibration response under double-track operation is notably greater than that under single-track operation. When two trains pass simultaneously at a speed of 200 km/h or higher, or a single train passes at 350 km/h, the maximum Z-vibration level of the waiting hall floor exceeds 75 dB, which goes beyond the specification limit. Full article

► Show Figures

Figure 1

14 pages, 236 KB

Open AccessArticle

Worlds Apart on Common Ground: Parent-Educator Perceptions of National Identity, Technology, and Collaboration in Hong Kong Kindergartens

by Jessie Ming Sin Wong

Educ. Sci. 2025, 15(12), 1626; https://doi.org/10.3390/educsci15121626 (registering DOI) - 3 Dec 2025

Abstract

Amid a policy mandate to foster national identity in Hong Kong’s early childhood education sector, this study explores the complex intersection of pedagogy, home–school collaboration, and technology integration. Navigating this value-laden topic depends fundamentally on a strong partnership between parents and educators, yet [...] Read more.

Amid a policy mandate to foster national identity in Hong Kong’s early childhood education sector, this study explores the complex intersection of pedagogy, home–school collaboration, and technology integration. Navigating this value-laden topic depends fundamentally on a strong partnership between parents and educators, yet the rapid push for artificial intelligence (AI) creates additional pressures. This qualitative study investigates the shared and conflicting perspectives of these key stakeholders. Eight focus groups were conducted with 21 parents and 26 educators from four diverse Hong Kong kindergartens. Data were analyzed using a novel human–AI collaborative thematic analysis to ensure analytical depth and reliability. The findings reveal a paradoxical consensus: while parents and educators agree on an experiential, play-based pedagogy, they hold divergent views on the division of responsibility. A further misalignment exists in communication ideals, with parents prioritizing efficiency and educators prioritizing relational nuance. Critically, a technology paradox emerges, pitting parents’ aspirational hopes for AI against educators’ pragmatic concerns over inadequate resources, training, and pedagogical suitability. The study concludes that a significant perception gap strains the home–school partnership. Simply introducing technology without first addressing these foundational human and resource-based misalignments risks widening, rather than bridging, the divide, offering important lessons for education systems globally. Full article

29 pages, 4084 KB

Open AccessArticle

Residents’ Satisfaction with Public Spaces in Old Urban Residential Communities: A PLS-SEM and IPMA-Based Case Study of Nankai District, Tianjin

by Jiahui Wang and Di Zhao

Land 2025, 14(12), 2363; https://doi.org/10.3390/land14122363 (registering DOI) - 3 Dec 2025

Abstract

With the acceleration of urbanization, urban renewal and the renovation of old residential communities have become important measures to enhance the quality of cities and improve the living conditions of residents. How to scientifically identify and evaluate the environmental factors of public spaces [...] Read more.

With the acceleration of urbanization, urban renewal and the renovation of old residential communities have become important measures to enhance the quality of cities and improve the living conditions of residents. How to scientifically identify and evaluate the environmental factors of public spaces and their impacts from the perspective of residents’ demands and satisfaction remains an important issue that urgently needs to be addressed in the current research field. This research takes the urban renewal project in Tiantuo Area, Nankai District, Tianjin, as an example by using questionnaire surveys, PLS-SEM and IPMA, and other multivariate statistical analysis methods to systematically explore the influence mechanism factors such as space accessibility, spatial usability, spatial maintainability, environmental comfort, and site safety on residents’ satisfaction. These findings reveal the following: (1) Space Accessibility has a significant direct positive impact on residents’ satisfaction. (2) Emotional Perception plays a complete mediating role in the relationship between Site Safety and residents’ satisfaction. (3) Emotional Perception has a complementary mediating effect in the relationship between Space Usability, Space Maintainability, Environmental Comfort, and Resident Satisfaction. (4) The renovation of old urban residential communities should give priority to improving space maintainability, especially focusing on the green landscape maintenance status, life-supporting infrastructure maintenance degree, and the maintenance status of entertainment and fitness facilities. Secondly, the space accessibility should be optimized and improved. In the future, in terms of the Physical Space, we should focus on the rationality of road network layout and strengthen the maintenance and management of public facilities. In terms of the Perceptional Space, the flatness of pavement should be optimized and the construction of security systems should be strengthened. In terms of the Psychological Status, a multi-party resident participation mechanism can be established to encourage residents to participate in the decision-making and construction of community public affair. As has been noted, this research quantitatively analyzed the key factors influencing residents’ satisfaction and their respective impact intensities and proposed prioritized and targeted optimization strategies for the existing situation. The research results are expected to provide a theoretical basis and practical decision-making reference for the optimization of public space environmental quality. Full article

(This article belongs to the Special Issue Urban Planning for a Sustainable Future)

► Show Figures

Figure 1

16 pages, 2866 KB

Open AccessArticle

Bifunctionalized Polyethyleneimine-Based Sponge for Adsorption of Ibuprofen from Aqueous Solution

by Xiaoyi Gou, Zia Ahmad, Zaijin You and Zhou Ren

Polymers 2025, 17(23), 3221; https://doi.org/10.3390/polym17233221 (registering DOI) - 3 Dec 2025

Abstract

A quaternized and phenyl-functionalized hyperbranched PEI-based sponge (S_HPEI-QP) was successfully prepared, and its adsorption performance was investigated to evaluate its potential for removing the anionic non-steroidal anti-inflammatory drug (ibuprofen (IBU)). We reported that the synthesis of polyethyleneimine-based sponges was achieved through [...] Read more.

A quaternized and phenyl-functionalized hyperbranched PEI-based sponge (S_HPEI-QP) was successfully prepared, and its adsorption performance was investigated to evaluate its potential for removing the anionic non-steroidal anti-inflammatory drug (ibuprofen (IBU)). We reported that the synthesis of polyethyleneimine-based sponges was achieved through cryo-polymerization using 1,4-butanediol diglycidyl ether (BDDE) as the crosslinking agent. Subsequent functionalization with resorcinol diglycidyl ether (RDGE) and trimethylamine introduced quaternary ammonium cations, imparting strong basicity and hydrophilicity, as well as phenyl groups, conferring hydrophobic characteristics, respectively. The aforementioned sponge material, S_HPE-QPI, primarily facilitates the efficient adsorption of IBU in aqueous solutions through the anion exchange properties of quaternary ammonium groups and the π-π interactions associated with oxygen-activated benzene rings. Various characterizations, such as scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FT-IR), X-ray photoelectron spectroscopy (XPS), and specific surface area determination method (BET), confirmed the successful synthesis of the bifunctionalized S_HPEI-QP adsorbent. This adsorbent features a porous structure (specific surface area of 77.2 m² g⁻¹ and pore size distribution of 25–100 nm) and an isoelectric point (pH_pzc) of 9.38. The adsorption kinetics of the adsorbent for IBU were extremely rapid and conformed to a pseudo-second-order kinetic model, and the adsorption isotherm aligned with the Langmuir isotherm model. Noteworthily, S_HPEI-QP demonstrated an exceptionally high adsorption capacity for IBU, achieving a maximum uptake of 905.73 mg g⁻¹ at pH 7.0, which surpassed that of most of the previous reported adsorbents. Moreover, the sponge material can be chemically regenerated. After eight cycles of use, the adsorption efficiency decreased by only 4%. These findings suggest that the synthesized dendritic anion exchange adsorbent represents a promising candidate for the removal of IBU from contaminated water sources. Full article

(This article belongs to the Section Polymer Applications)

► Show Figures

Graphical abstract

15 pages, 2807 KB

Open AccessArticle

Flash Lamp Sintering and Optoelectronic Performance of Silver Nanowire Transparent Conductive Films

by Jiaqi Shan, Ye Hong, Kaixuan Cui, Yifan Xiao and Xingzhong Guo

Materials 2025, 18(23), 5456; https://doi.org/10.3390/ma18235456 (registering DOI) - 3 Dec 2025

Abstract

Silver nanowire transparent conductive films (AgNW TCFs), as a promising new generation of transparent electrode materials poised to replace ITO, have long been plagued by inadequate optoelectronic performance. Herein, flash lamp sintering was used to facilitate rapid welding of TCFs, and the effects [...] Read more.

Silver nanowire transparent conductive films (AgNW TCFs), as a promising new generation of transparent electrode materials poised to replace ITO, have long been plagued by inadequate optoelectronic performance. Herein, flash lamp sintering was used to facilitate rapid welding of TCFs, and the effects of process parameters and TCFs’ characteristics on the sintering outcomes were investigated. The leveraging of millisecond-scale intense light pulses of flash lamp sintering can achieve the rapid welding of AgNWs, thereby enhancing the optoelectronic performance of TCFs. The TCFs fabricated from 30 nm diameter AgNWs with an initial sheet resistance of 111 Ω/sq exhibited a reduced sheet resistance of 57 Ω/sq post-sintering, while maintaining a transmittance of 93.3%. The quality factor increased from 4.56 × 10⁻³ to 9.09 × 10⁻³ Ω⁻¹, and the surface roughness decreased from 6.12 to 5.19 nm after sintering. This work holds significant promise for advancing the continuous production of AgNW TCFs using flash lamp sintering technology, potentially paving the way for high-quality, low-cost, and rapid manufacturing of AgNW TCFs. Full article

(This article belongs to the Special Issue Advanced Thin Films: Structural, Optical, and Electrical Properties)

► Show Figures

Figure 1

33 pages, 628 KB

Open AccessReview

A Review of Pedestrian Trajectory Prediction Methods Based on Deep Learning Technology

by Xiang Gu, Chao Li, Long Gao and Xuefen Niu

Sensors 2025, 25(23), 7360; https://doi.org/10.3390/s25237360 (registering DOI) - 3 Dec 2025

Abstract

Pedestrian trajectory prediction is a critical component of autonomous driving and intelligent urban systems, with deep learning now dominating the field by overcoming the limitations of traditional models in handling multi-modal behaviors and complex social interactions. This survey provides a systematic review and [...] Read more.

Pedestrian trajectory prediction is a critical component of autonomous driving and intelligent urban systems, with deep learning now dominating the field by overcoming the limitations of traditional models in handling multi-modal behaviors and complex social interactions. This survey provides a systematic review and critical analysis of deep learning-based approaches, offering a structured examination of four key model families: RNNs, GANs, GCNs, and Transformer. Unlike previous reviews, we introduce a comparative analytical framework that evaluates each method’s strengths and limitations across standardized criteria. The review also presents a comprehensive taxonomy of datasets and evaluation metrics, highlighting both established practices and emerging trends. Finally, we derive future research directions directly from our critical assessment, focusing on semantic scene understanding, model transferability, and the precision–efficiency trade-off. Our work provides both a historical perspective on methodological evolution and a forward-looking analysis to guide future research development. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

26 pages, 3868 KB

Open AccessArticle

Tourism-Driven Land Use Transitions and Rural Livelihood Resilience: A Spatial Production Approach to Sustainable Development in China’s Heritage Areas

by Lijie Liu, Xinmin Liu and Yanan Zhang

Sustainability 2025, 17(23), 10839; https://doi.org/10.3390/su172310839 (registering DOI) - 3 Dec 2025

Abstract

Enhancing farmers’ livelihood resilience is a cornerstone of sustainable rural development and poverty alleviation consolidation in developing countries. While tourism has emerged as a prominent rural revitalization strategy, the mediating role of tourism-induced land use transitions in building resilience—and the underlying spatial mechanisms [...] Read more.

Enhancing farmers’ livelihood resilience is a cornerstone of sustainable rural development and poverty alleviation consolidation in developing countries. While tourism has emerged as a prominent rural revitalization strategy, the mediating role of tourism-induced land use transitions in building resilience—and the underlying spatial mechanisms through which these transformations operate—remains inadequately understood. This study integrates Henri Lefebvre’s spatial production theory with land systems analysis to examine how tourism-driven land use transitions influence farmers’ livelihood resilience in rural China. Using provincial panel data and three waves (2018, 2020, 2022) of nationally representative household survey data from the China Family Panel Studies (CFPS), we construct a comprehensive tourism development index emphasizing land transformation dimensions and employ panel regression models with instrumental variables and threshold analysis. The findings reveal that tourism-induced land use transitions significantly enhance farmers’ livelihood resilience through three distinct spatial mechanisms: land-based rural infrastructure investment, industrial land structure rationalization, and cultural facility land development. Importantly, this relationship exhibits a double-threshold effect with diminishing marginal returns, and the positive impact is substantially stronger in heritage-rich regions with comparative policy advantages. By establishing land use transitions as a critical spatial production pathway linking tourism to sustainable livelihood outcomes, this study advances land systems science, offering a novel theoretical framework for integrating people–nature interactions in heritage-rich rural areas and practical guidance for strategic land use planning in support of the Sustainable Development Goals (SDGs). Full article

► Show Figures

Figure 1

28 pages, 3783 KB

Open AccessReview

Exploring the Links Between Clean Energies and Community Actions in Remote Areas: A Literature Review

by Alessandra Longo, Matteo Basso, Giulia Lucertini and Linda Zardo

Energies 2025, 18(23), 6350; https://doi.org/10.3390/en18236350 (registering DOI) - 3 Dec 2025

Abstract

In the fight against growing energy poverty in Europe, remote and rural areas are most affected but play a crucial role in promoting a fair and sustainable transition. Furthermore, energy communities have been recognized as cost-efficient options and opportunities to enhance the active [...] Read more.

In the fight against growing energy poverty in Europe, remote and rural areas are most affected but play a crucial role in promoting a fair and sustainable transition. Furthermore, energy communities have been recognized as cost-efficient options and opportunities to enhance the active participation of citizens in electricity markets. Despite the wide recognition of their potential in alleviating energy poverty, evidence is still limited. This paper investigates the ‘missing links’ in producing clean energy through community-based practices in remote areas. This study presents a literature review aimed at identifying case studies at the European level to build a knowledge base on the state of the art in the context of the Green Deal. Of the 4422 publications found, we identified and analyzed 266 publications with one or more European cases. Of these, only 67 publications used keywords relevant to our research objective, which we further explored and categorized according to the primary purpose of the study, i.e., assessment, barriers and gaps, implementation, management and planning, modeling, and public opinion. Our results show that publications serve mainly to test a methodology for potential use and not to recount an experience, lacking practical application and policy integration. Nevertheless, we noticed a tendency to activate citizen engagement forms or gather perceptions to increase social acceptability. Full article

(This article belongs to the Section B2: Clean Energy)

► Show Figures

Figure 1

20 pages, 652 KB

Open AccessReview

The Evolving Role of Cine MRI in Crohn’s Disease: From Functional Motility Analysis to Precision Management: A Review of the Last 10 Years

by Ali S. Alyami

Diagnostics 2025, 15(23), 3078; https://doi.org/10.3390/diagnostics15233078 (registering DOI) - 3 Dec 2025

Abstract

Cine (dynamic) MRI is a non-invasive MRI technique that captures moving images and can be valuable in evaluating inflammatory bowel disease (IBD). This sequence shows emerging potential in providing functional data to assess bowel motility patterns, to aid in the differentiation between predominantly [...] Read more.

Cine (dynamic) MRI is a non-invasive MRI technique that captures moving images and can be valuable in evaluating inflammatory bowel disease (IBD). This sequence shows emerging potential in providing functional data to assess bowel motility patterns, to aid in the differentiation between predominantly inflammatory (showing reduced peristalsis) and fibrotic strictures (rigid, non-motile segments) and detecting functional obstructions in Crohn’s disease (CD). Unlike static MRI, cine MRI enables clinicians to observe peristaltic movements, aiding in disease characterization and treatment monitoring. Its non-invasive nature and lack of ionizing radiation make it especially useful for repeated assessments in CD. Studies indicate it improves diagnostic accuracy when used with conventional MRI sequences, providing a complementary, functional dimension to the comprehensive management of this chronic condition. While the functional assessment offered by cine MRI presents a significant advantage over conventional static imaging, its clinical translation is currently challenged by high technical variability. Specifically, there is a distinct lack of standardized acquisition protocols (such as field strength, sequence parameters), post-processing software, and universally validated quantitative motility metrics (such as motility index). Therefore, a primary objective of this review is not only to summarize the evolving diagnostic and monitoring applications of cine MRI but also to critically address the methodological inconsistencies and reproducibility hurdles that must be overcome before this technique can be fully integrated into clinical guidelines for precision management of CD. Full article

(This article belongs to the Special Issue Advancements in the Diagnosis and Treatment of Inflammatory Bowel Diseases)

► Show Figures

Figure 1

27 pages, 2900 KB

Open AccessArticle

Graph-SENet: An Unsupervised Learning-Based Graph Neural Network for Skeleton Extraction from Point Cloud

by Jie Li, Wei Guo and Wenli Zhang

Future Internet 2025, 17(12), 558; https://doi.org/10.3390/fi17120558 (registering DOI) - 3 Dec 2025

Abstract

Extracting 3D skeletons from point clouds is a challenging task in computer vision. Most existing deep learning methods rely heavily on supervised data requiring extensive manual annotation. Consequently, re-labeling is often necessary for cross-category applications, while the process of 3D point cloud annotation [...] Read more.

Extracting 3D skeletons from point clouds is a challenging task in computer vision. Most existing deep learning methods rely heavily on supervised data requiring extensive manual annotation. Consequently, re-labeling is often necessary for cross-category applications, while the process of 3D point cloud annotation is inherently time-consuming and expensive. Simultaneously, existing unsupervised methods often suffer from significant skeleton point deviations due to limited capabilities in modeling local structures. To address these limitations, we propose Graph-SENet, an unsupervised learning-based graph neural network method for skeleton extraction. This method integrates dynamic graph convolution with a multi-level feature fusion mechanism to more comprehensively capture local geometric relationships. Through a multi-dimensional unsupervised feature loss, it learns the structural representation of skeleton points, significantly improving the precision and stability of skeleton point localization under annotation-free conditions. Furthermore, we propose a graph autoencoder structure optimized by cosine similarity to predict topological connections between skeleton points, thereby recovering semantically consistent and structurally complete 3D skeleton representations in an end-to-end manner. Experimental results on multiple datasets, including ShapeNet, ITOP, and Soybean-MVS, demonstrate that Graph-SENet outperforms existing mainstream unsupervised methods in terms of Chamfer Distance and F1-score. It exhibits superior accuracy, robustness, and cross-category generalization capabilities, effectively reducing manual annotation costs while enhancing the completeness and semantic consistency of skeleton recovery. These results validate the application potential and practical value of Graph-SENet in 3D structure understanding and downstream 3D analysis tasks. Full article

(This article belongs to the Special Issue Algorithms and Models for Next-Generation Vision Systems)

► Show Figures

Figure 1

30 pages, 3730 KB

Open AccessArticle

Deep Learning Analysis of CBCT Images for Periodontal Disease: Phenotype-Level Concordance with Independent Transcriptomic and Microbiome Datasets

by Ștefan Lucian Burlea, Călin Gheorghe Buzea, Florin Nedeff, Diana Mirilă, Valentin Nedeff, Maricel Agop, Lăcrămioara Ochiuz and Adina Oana Armencia

Dent. J. 2025, 13(12), 578; https://doi.org/10.3390/dj13120578 (registering DOI) - 3 Dec 2025

Abstract

Background: Periodontitis is a common inflammatory disease characterized by progressive loss of alveolar bone. Cone-beam computed tomography (CBCT) can visualize 3D periodontal bone defects, but its interpretation is time-consuming and examiner-dependent. Deep learning may support standardized CBCT assessment if performance and biological relevance [...] Read more.

Background: Periodontitis is a common inflammatory disease characterized by progressive loss of alveolar bone. Cone-beam computed tomography (CBCT) can visualize 3D periodontal bone defects, but its interpretation is time-consuming and examiner-dependent. Deep learning may support standardized CBCT assessment if performance and biological relevance are adequately characterized. Methods: We used the publicly available MMDental dataset (403 CBCT volumes from 403 patients) to train a 3D ResNet-18 classifier for binary discrimination between periodontitis and healthy status based on volumetric CBCT scans. Volumes were split by subject into training (n = 282), validation (n = 60), and test (n = 61) sets. Model performance was evaluated using area under the receiver operating characteristic curve (AUROC), area under the precision–recall curve (AUPRC), and calibration metrics with 95% bootstrap confidence intervals. Grad-CAM saliency maps were used to visualize the anatomical regions driving predictions. To explore phenotype-level biological concordance, we analyzed an independent gingival transcriptomic cohort (GSE10334, n ≈ 220 arrays after quality control) and an independent oral microbiome cohort based on 16S rRNA amplicon sequencing, using unsupervised clustering, differential expression/abundance testing, and pathway-level summaries. Results: On the held-out CBCT test set, the model achieved an AUROC of 0.729 (95% CI: 0.599–0.850) and an AUPRC of 0.551 (95% CI: 0.404–0.727). At a high-sensitivity operating point (sensitivity 0.95), specificity was 0.48, yielding an overall accuracy of 0.62. Grad-CAM maps consistently highlighted the alveolar crest and furcation regions in periodontitis cases, in line with expected patterns of bone loss. In the transcriptomic cohort, inferred periodontitis samples showed up-regulation of inflammatory and osteoclast-differentiation pathways and down-regulation of extracellular-matrix and mitochondrial programs. In the microbiome cohort, disease-associated samples displayed a dysbiotic shift with enrichment of classic periodontal pathogens and depletion of health-associated commensals. These omics patterns are consistent with an inflammatory–osteolytic phenotype that conceptually aligns with the CBCT-defined disease class. Conclusions: This study presents a proof-of-concept 3D deep learning model for CBCT-based periodontal disease classification that achieves moderate discriminative performance and anatomically plausible saliency patterns. Independent transcriptomic and microbiome analyses support phenotype-level biological concordance with the imaging-defined disease class, but do not constitute subject-level multimodal validation. Given the modest specificity, single-center imaging source, and inferred labels in the omics cohorts, our findings should be interpreted as exploratory and hypothesis-generating. Larger, multi-center CBCT datasets and prospectively collected paired imaging–omics cohorts are needed before clinical implementation can be considered. Full article

► Show Figures

Figure 1

33 pages, 10355 KB

Open AccessArticle

S²GL-MambaResNet: A Spatial–Spectral Global–Local Mamba Residual Network for Hyperspectral Image Classification

by Tao Chen, Hongming Ye, Guojie Li, Yaohan Peng, Jianming Ding, Huayue Chen, Xiangbing Zhou and Wu Deng

Remote Sens. 2025, 17(23), 3917; https://doi.org/10.3390/rs17233917 (registering DOI) - 3 Dec 2025

Abstract

In hyperspectral image classification (HSIC), each pixel contains information across hundreds of contiguous spectral bands; therefore, the ability to perform long-distance modeling that stably captures and propagates these long-distance dependencies is critical. A selective structured state space model (SSM) named Mamba has shown [...] Read more.

In hyperspectral image classification (HSIC), each pixel contains information across hundreds of contiguous spectral bands; therefore, the ability to perform long-distance modeling that stably captures and propagates these long-distance dependencies is critical. A selective structured state space model (SSM) named Mamba has shown strong capabilities for capturing cross-band long-distance dependencies and exhibits advantages in long-distance modeling. However, the inherently high spectral dimensionality, information redundancy, and spatial heterogeneity of hyperspectral images (HSI) pose challenges for Mamba in fully extracting spatial–spectral features and in maintaining computational efficiency. To address these issues, we propose S²GL-MambaResNet, a lightweight HSI classification network that tightly couples Mamba with progressive residuals to enable richer global, local, and multi-scale spatial–spectral feature extraction, thereby mitigating the negative effects of high dimensionality, redundancy, and spatial heterogeneity on long-distance modeling. To avoid fragmentation of spatial–spectral information caused by serialization and to enhance local discriminability, we design a preprocessing method applied to the features before they are input to Mamba, termed the Spatial–Spectral Gated Attention Aggregator (SS-GAA). SS-GAA uses spatial–spectral adaptive gated fusion to preserve and strengthen the continuity of the central pixel’s neighborhood and its local spatial–spectral representation. To compensate for a single global sequence network’s tendency to overlook local structures, we introduce a novel Mamba variant called the Global_Local Spatial_Spectral Mamba Encoder (GLS²ME). GLS²ME comprises a pixel-level global branch and a non-overlapping sliding-window local branch for modeling long-distance dependencies and patch-level spatial–spectral relations, respectively, jointly improving generalization stability under limited sample regimes. To ensure that spatial details and boundary integrity are maintained while capturing spectral patterns at multiple scales, we propose a multi-scale Mamba encoding scheme, the Hierarchical Spectral Mamba Encoder (HSME). HSME first extracts spectral responses via multi-scale 1D spectral convolutions, then groups spectral bands and feeds these groups into Mamba encoders to capture spectral pattern information at different scales. Finally, we design a Progressive Residual Fusion Block (PRFB) that integrates 3D residual recalibration units with Efficient Channel Attention (ECA) to fuse multi-kernel outputs within a global context. This enables ordered fusion of local multi-scale features under a global semantic context, improving information utilization efficiency while keeping computational overhead under control. Comparative experiments on four publicly available HSI datasets demonstrate that S²GL-MambaResNet achieves superior classification accuracy compared with several state-of-the-art methods, with particularly pronounced advantages under few-shot and class-imbalanced conditions. Full article

(This article belongs to the Special Issue Deep Learning for Spectral-Spatial Hyperspectral Image Classification (2nd Edition))

► Show Figures

Figure 1

16 pages, 6207 KB

Open AccessCommunication

AI-Guided Dual Strategy for Peptide Inhibitor Design Targeting Structural Polymorphs of α-Synuclein Fibrils

by Jinfang Duan, Haoyu Zhang and Chuanqi Sun

Cells 2025, 14(23), 1921; https://doi.org/10.3390/cells14231921 (registering DOI) - 3 Dec 2025

Abstract

One of the most important events in the pathogenesis of Parkinson’s disease and related disorders is the formation of abnormal fibrils via the aggregation of α-synuclein (α-syn) with β-sheet-rich organization. The use of Cryo-EM has uncovered different polymorphs of the fibrils, each having [...] Read more.

One of the most important events in the pathogenesis of Parkinson’s disease and related disorders is the formation of abnormal fibrils via the aggregation of α-synuclein (α-syn) with β-sheet-rich organization. The use of Cryo-EM has uncovered different polymorphs of the fibrils, each having unique structural interfaces, which has made the design of inhibitors even more challenging. Here, a structure-guided framework incorporating AI-assisted peptide generation was set up with the objective of targeting the conserved β-sheet motifs that are present in various forms of α-syn fibrils. The ProteinMPNN, then, AlphaFold-Multimer, and PepMLM were employed to create short peptides that would interfere with the growth of the fibrils. The two selected candidates, T1 and S1, showed a significant inhibition of α-syn fibrillation, as measured by a decrease in the ThT fluorescence and the generation of either amorphous or fragmented aggregates. The inhibitory potency of the peptides was in line with the predicted interface energies. This research work illustrates that the integration of cryo-EM structural knowledge with the computational design method leads to the quick discovery of the wide-spectrum peptide inhibitors, which is a good strategy for the precision treatment of neurodegenerative diseases. Full article

(This article belongs to the Special Issue α-Synuclein in Parkinson’s Disease)

► Show Figures

Figure 1

34 pages, 11986 KB

Open AccessArticle

High-Speed Die Bond Quality Detection Using Lightweight Architecture DSGβSI-SECS-Yolov7-Tiny

by Bao Rong Chang, Hsiu-Fen Tsai and Wei-Shun Chang

Sensors 2025, 25(23), 7358; https://doi.org/10.3390/s25237358 (registering DOI) - 3 Dec 2025

Abstract

The die bonding process significantly impacts the yield and quality of IC packaging, and its quality detection is also a critical image sensing technology. With the advancement of machine automation and increased operating speeds, the misclassification rate in die bond image inspection has [...] Read more.

The die bonding process significantly impacts the yield and quality of IC packaging, and its quality detection is also a critical image sensing technology. With the advancement of machine automation and increased operating speeds, the misclassification rate in die bond image inspection has also risen. Therefore, this study develops a high-speed intelligent vision inspection model that slightly improves classification accuracy and adapts to the operation of new-generation machines. Furthermore, by identifying the causes of die bonding defects, key process parameters can be adjusted in real time during production, thereby improving the yield of the die bonding process and substantially reducing manufacturing cost losses. Previously, we proposed a lightweight model named DSGβSI-YOLOv7-tiny, which integrates depthwise separable convolution, Ghost convolution, and a Sigmoid activation function with a learnable β parameter. This model enables real-time and efficient detection and prediction of die bond quality through image sensing. We further enhanced the previous model by incorporating an SE layer, ECA-Net, Coordinate Attention, and a Small Object Enhancer to accommodate the faster operation of new machines. This improvement resulted in a more lightweight architecture named DSGβSI-SECS-YOLOv7-tiny. Compared with the previous model, the proposed model achieves an increased inference speed of 294.1 FPS and a Precision of 99.1%. Full article

(This article belongs to the Special Issue AI-Enhanced Industrial Sensors: From Adaptive Detection to Smart Manufacturing)

► Show Figures

Figure 1

26 pages, 1005 KB

Open AccessArticle

A Context-Aware Lightweight Framework for Source Code Vulnerability Detection

by Yousef Sanjalawe, Budoor Allehyani and Salam Al-E’mari

Future Internet 2025, 17(12), 557; https://doi.org/10.3390/fi17120557 (registering DOI) - 3 Dec 2025

Abstract

As software systems grow increasingly complex and interconnected, detecting vulnerabilities in source code has become a critical and challenging task. Traditional static analysis methods often fall short in capturing deep, context-dependent vulnerabilities and adapting to rapidly evolving threat landscapes. Recent efforts have explored [...] Read more.

As software systems grow increasingly complex and interconnected, detecting vulnerabilities in source code has become a critical and challenging task. Traditional static analysis methods often fall short in capturing deep, context-dependent vulnerabilities and adapting to rapidly evolving threat landscapes. Recent efforts have explored knowledge graphs and transformer-based models to enhance semantic understanding; however, these solutions frequently rely on static knowledge bases, exhibit high computational overhead, and lack adaptability to emerging threats. To address these limitations, we propose DynaKG-NER++, a novel and lightweight framework for context-aware vulnerability detection in source code. Our approach integrates lexical, syntactic, and semantic features using a transformer-based token encoder, dynamic knowledge graph embeddings, and a Graph Attention Network (GAT). We further introduce contrastive learning on vulnerability–patch pairs to improve discriminative capacity and design an attention-based fusion module to combine token and entity representations adaptively. A key innovation of our method is the dynamic construction and continual update of the knowledge graph, allowing the model to incorporate newly published CVEs and evolving relationships without retraining. We evaluate DynaKG-NER++ on five benchmark datasets, demonstrating superior performance across span-level F1 (89.3%), token-level accuracy (93.2%), and AUC-ROC (0.936), while achieving the lowest false positive rate (5.1%) among state-of-the-art baselines. Sta tistical significance tests confirm that these improvements are robust and meaningful. Overall, DynaKG-NER++ establishes a new standard in vulnerability detection, balancing accuracy, adaptability, and efficiency, making it highly suitable for deployment in real-world static analysis pipelines and resource-constrained environments. Full article

(This article belongs to the Topic Addressing Security Issues Related to Modern Software)

► Show Figures

Figure 1

17 pages, 5437 KB

Open AccessArticle

Battery Parameter Identification and SOC Estimation Based on Online Parameter Identification and MIUKF

by Liteng Zeng, Lei Zhao, Youwei Song, Yuli Hu and Guang Pan

Batteries 2025, 11(12), 445; https://doi.org/10.3390/batteries11120445 (registering DOI) - 3 Dec 2025

Abstract

Accurate state of charge (SOC) estimation is crucial for the safety, reliability, and energy efficiency of lithium-ion battery systems. However, variations in battery parameters and the loss of historical information during the update steps of traditional unscented Kalman filters (UKFs) often lead to [...] Read more.

Accurate state of charge (SOC) estimation is crucial for the safety, reliability, and energy efficiency of lithium-ion battery systems. However, variations in battery parameters and the loss of historical information during the update steps of traditional unscented Kalman filters (UKFs) often lead to decreased estimation accuracy under dynamic operating conditions. To address these issues, this paper proposes a variable forgetting factor recursive least squares (VFFRLS) algorithm combined with a multi-innovation unscented Kalman filter (MIUKF) algorithm. First, a second-order RC equivalent circuit model is established, and the battery parameters are identified online using the VFFRLS method, enabling the model to dynamically adapt to changing operating conditions. Then, multi-innovation theory is incorporated into the standard UKF, extending the single-innovation matrix to a multi-innovation matrix, effectively enhancing the utilization of historical residuals and improving robustness to measurement noise and model uncertainty. Experimental validation under four typical dynamic operating conditions (FUDS, DST, BJDST, and US06) demonstrates that the proposed method significantly improves SOC estimation accuracy. Compared to the traditional UKF, MIUKF reduces MAE and RMSE by 25–30% while maintaining real-time performance, with single-step computation time reaching the microsecond level. Robustness tests under different initial SOC errors further validate MIUKF’s strong robustness to initial biases. In summary, the proposed method provides an effective solution for high-precision SOC estimation of batteries and has the potential for application in battery management systems. Full article

► Show Figures

Figure 1

27 pages, 1696 KB

Open AccessReview

Biotechnologies for Promoting Germplasm Resource Utilization and Preservation of the Coconut and Important Palms

by Ke Deng, Shuya Yang, Sisunandar Sisunandar, Binh-Minh Tran, Mridula Kottekate, Nancy Shaftang and Zhihua Mu

Horticulturae 2025, 11(12), 1461; https://doi.org/10.3390/horticulturae11121461 (registering DOI) - 3 Dec 2025

Abstract

Coconut (Cocos nucifera L.) plays a vital economic and cultural role in many tropical and subtropical regions. A comprehensive review of the existing literature underscores that advanced biotechnologies are pivotal in unlocking the full potential of coconut germplasm exchange, which is crucial [...] Read more.

Coconut (Cocos nucifera L.) plays a vital economic and cultural role in many tropical and subtropical regions. A comprehensive review of the existing literature underscores that advanced biotechnologies are pivotal in unlocking the full potential of coconut germplasm exchange, which is crucial for the future sustainability of this crop. While traditional exchange methods are hampered by phytosanitary risks and logistical burdens, biotechnological interventions such as in vitro conservation and cryopreservation present targeted solutions to overcome these bottlenecks. The exchange, facilitated by these technologies, allows for the efficient introduction of desirable traits. We indicate that diversification and germplasm exchange hold the key to improving coconut quality and yield, developing varieties resistant to pests and diseases, and ensuring long-term conservation of coconut genetic diversity. This review highlights the potential to overcome the challenges faced by regional breeding programs often hindered by restricted genetic resources. Furthermore, by examining past successes and challenges in coconut germplasm identification and exchange, we offer perspectives on optimizing strategies to conserve diversity. This work emphasizes that germplasm exchange paves the way for coconut varieties that can thrive under changing environmental conditions, securing the future of this highly valuable crop. Full article

(This article belongs to the Special Issue Multi-Omics-Driven Breeding for Tropical Horticultural Crops)

► Show Figures

Figure 1

19 pages, 1319 KB

Open AccessArticle

Effects of Corn Steep Liquor on the Fermentation Quality, Bacterial Community and Ruminal Degradation Rate of Corncob Silage

by Xinyi Wang, Xinfeng Wang, Tengyu Wang, Xiaoping Chen, Zuoxing Huang, Rui Yang, Shuai Liu, Xinwen Sun and Dengke Hua

Animals 2025, 15(23), 3487; https://doi.org/10.3390/ani15233487 (registering DOI) - 3 Dec 2025

Abstract

This study aims to investigate the effect of varying addition levels of corn steep liquor (CSL) on the fermentation quality, bacterial community, and ruminal degradation rate of corncob silage. The experiment included a control group (CON) and four treatment groups: L1 with 5% [...] Read more.

This study aims to investigate the effect of varying addition levels of corn steep liquor (CSL) on the fermentation quality, bacterial community, and ruminal degradation rate of corncob silage. The experiment included a control group (CON) and four treatment groups: L1 with 5% CSL (50 g·kg⁻¹ fresh matter), L2 with 10% CSL (100 g·kg⁻¹ fresh matter), L3 with 15% CSL (150 g·kg⁻¹ fresh matter), and L4 with 20% CSL (200 g·kg⁻¹ fresh matter). The water content was controlled at 65% during fermentation for a period of 45 days. The results showed that the addition of CSL significantly increased the contents of dry matter (DM), crude protein (CP), and lactic acid (LA), while decreasing the pH, neutral detergent fiber (NDF), acid detergent fiber (ADF), and ammonia nitrogen (NH₃-N). Furthermore, the addition of CSL altered the relative abundance of microbial genera. While Pediococcus was the dominant bacterium in the CON group, Lactobacillus became the prevalent species upon the addition of CSL, and its relative abundance increased in accordance with the supplemental amount. These findings suggest that CSL provides a favorable environment for lactic acid bacteria. It is worth noting that CSL addition did not significantly alter the phylum-level bacterial community structure. The dominant bacterial taxa across all treatments were Bacillota, Proteobacteria, and Bacteroidota, with their cumulative relative abundance accounting for over 95%. The rumen degradation of the tested feedstuff was determined using the in situ nylon bag method. Results revealed that incorporating CSL into corncob silage significantly enhanced the effective degradation rates of DM, CP, NDF, and ADF in the rumen of Kazakh sheep. Specifically, the effective degradation rate of DM in the CON group was only 49.10%, which increased to 53.12% following the addition of 20% CSL, along with corresponding improvements in the degradation rates of CP, NDF, and ADF. In summary, as a valuable feed additive, corn steep liquor supports the proliferation of beneficial microorganisms in fermentation systems by supplying essential growth substrates. Additionally, it improves the nutritional balance of corncob feed and further enhances the absorption and utilization of nutrients from this feed by animals. Full article

(This article belongs to the Special Issue Alternative Protein Sources for Animal Feeds)

► Show Figures

Figure 1

19 pages, 5227 KB

Open AccessArticle

Exploring the Role of Tamarind Seed Polysaccharides in Modulating the Structural, Digestive, and Emulsion Stability Properties of Waxy Corn Starch Composites

by Xiangyu Ya, Yongshuai Ma, Zibo Song, Yongli Jiang, Chaofan Guo and Junjie Yi

Foods 2025, 14(23), 4152; https://doi.org/10.3390/foods14234152 (registering DOI) - 3 Dec 2025

Abstract

This study investigated the effects of tamarind seed polysaccharide (TSP) on the structural characteristics, digestibility, and emulsifying properties of waxy maize starch (WMS), as well as their interaction mechanisms. WMS-TSP complexes were prepared via complexes to improve starch’s physical and functional properties. Native [...] Read more.

This study investigated the effects of tamarind seed polysaccharide (TSP) on the structural characteristics, digestibility, and emulsifying properties of waxy maize starch (WMS), as well as their interaction mechanisms. WMS-TSP complexes were prepared via complexes to improve starch’s physical and functional properties. Native WMS showed smooth spherical granules, while WMS-TSP samples formed freeze-drying-induced honeycomb structures (~200–250 μm). In vitro digestion indicated that WMS-TSP systems (5–15%) reduced RDS by 20.1–24.11% relative to native WMS (41% ± SD), suggesting a potential to attenuate postprandial glycemic responses. Fourier-transform infrared (FT-IR) spectroscopy revealed that TSP interacted with WMS mainly through non-covalent bonds such as hydrogen bonding, while influencing the degree of crystallinity without generating new crystalline polymorphs. In corn oil-based emulsions, the WMS-TSP composites showed strong viscoelastic behavior, with elevated storage (G′) and loss (G″) moduli, together with improved storage stability. These findings highlight the synergistic potential of WMS and TSP in enhancing the functionality of starch-based systems and provide insights into the role of polysaccharides in food structure and digestion regulation. Full article

(This article belongs to the Section Food Physics and (Bio)Chemistry)

► Show Figures

Figure 1

12 pages, 1061 KB

Open AccessArticle

The Premature Infants’ Gut Microbiota Assembly and Neurodevelopment (PIGMAN) Cohort Study: Protocol for a Prospective, Longitudinal Cohort Study

by Tingting Li, Liangfeng Fang, Xianhong Chen, Youming He, Xiaoyuan Pang, Ling Lin, Heng Chen, Yajie Su, Yan Huang, Yanping Guo, Tiantian Xiao, Aiping Liu, Yanli Wang, Hanhua Yang, Chuan Nie, Wei Zhou, Guang Yang, Chunquan Cai, Xiaoguang Zhou, Shujuan Zeng, Yongfu Yu, Long Li, Huifeng Zhang, Lijun Yu, Guoqiang Cheng, Wenhao Zhou, Cheng Chen, Zhangbin Yu, Mingbang Wang and Yingmei Xie Show full author list Hide full author list

Children 2025, 12(12), 1644; https://doi.org/10.3390/children12121644 (registering DOI) - 3 Dec 2025

Abstract

Background: Early-life gut microbiota colonization plays a significant role in the neurodevelopment of infants and young children. However, the causal relationship between early-life gut microbiota colonization and neurodevelopment in preterm infants has not yet been conclusively established. Our research will initiate the PIGMAN [...] Read more.

Background: Early-life gut microbiota colonization plays a significant role in the neurodevelopment of infants and young children. However, the causal relationship between early-life gut microbiota colonization and neurodevelopment in preterm infants has not yet been conclusively established. Our research will initiate the PIGMAN (Premature Infants Gut Microbiota Assembly and Neurodevelopment) cohort study to systematically examine the dynamic interplay between gut microbiota developmental trajectories and neurodevelopmental processes in preterm infants. Methods: This study will employ a longitudinal cohort design and utilize data from the PIGMAN cohort, examining the interplay between gut microbiota metabolism and neurodevelopmental outcomes. The study design incorporates longitudinal stool sample collection, which will be analyzed through 16S rRNA gene sequencing and metagenomic shotgun sequencing, enabling comprehensive characterization of microbial community dynamics and functional metabolic pathways. Anticipated Results: Advanced analytical approaches incorporating causal inference methodologies will be implemented to identify significant microbial and metabolic biomarkers associated with neurodevelopmental outcomes in preterm neonates, and to establish causal pathways between these biomarkers and neurodevelopment. These analytical advancements will facilitate the construction of predictive models that utilize temporal microbial signatures and metabolite trajectories as prognostic indicators for neurodevelopmental outcomes. Causal inference method evaluations will further reveal that specific gut-derived metabolites, particularly those involved in cholesterol metabolism and neural signaling pathways—such as bile acids and GABA (gamma-aminobutyric acid)—exhibit superior predictive capacity for cognitive development trajectories. Anticipated Conclusions: The findings will collectively suggest that longitudinal metabolic profiling of the gut ecosystem, when combined with causal network analysis, provides a novel paradigm for developing clinically actionable predictive models of neurodevelopment in vulnerable preterm populations. Full article

(This article belongs to the Special Issue Advances in Neonatal Resuscitation and Intensive Care)

► Show Figures

Figure 1

11 pages, 1719 KB

Open AccessReview

Lymphatic Spread of Non-Small-Cell Lung Cancer: Mechanisms, Patterns, Staging, and Diagnosis

by Mohamed Salih Makawi, Stephen Ciaccio, Asad Khan, Alireza Nathani and Ronaldo Ortiz-Pacheco

Lymphatics 2025, 3(4), 43; https://doi.org/10.3390/lymphatics3040043 (registering DOI) - 3 Dec 2025

Abstract

Lung cancer is the leading cause of cancer-related death worldwide. Lymph node involvement affects staging and, therefore, prognosis. Understanding lymph node drainage, metastatic patterns, and different sampling techniques contributes to the overall care of lung cancer patients. Non-small-cell lung cancer is the most [...] Read more.

Lung cancer is the leading cause of cancer-related death worldwide. Lymph node involvement affects staging and, therefore, prognosis. Understanding lymph node drainage, metastatic patterns, and different sampling techniques contributes to the overall care of lung cancer patients. Non-small-cell lung cancer is the most common type of lung cancer; appropriate staging is vital to determine treatment modalities which includes surgery, radiation therapy, chemotherapy, or a combination of these. In this review, we aim to describe the pathogenesis of lymph node metastasis, current guidelines for lymph node sampling, patterns of lymph node spread, new and novel lymph node sampling techniques, and their diagnostic yields. Full article

► Show Figures

Figure 1

14 pages, 285 KB

Open AccessArticle

Knowledge and Preventive Practices Toward COVID-19 Among Sex Workers in Chiang Mai, Thailand

by Sameen Ashfaq, Kriengkrai Srithanaviboonchai, Patumrat Sripan, Arunrat Tangmunkongvorakul and Natthapol Kosashunhanan

Int. J. Environ. Res. Public Health 2025, 22(12), 1814; https://doi.org/10.3390/ijerph22121814 (registering DOI) - 3 Dec 2025

Abstract

Sex workers were disproportionately affected by the COVID-19 pandemic due to precarious working conditions. This cross-sectional study was conducted in 2022 among 264 sex workers in Chiang Mai, Thailand, during the transition to the endemic phase, to evaluate their COVID-19 knowledge and preventive [...] Read more.

Sex workers were disproportionately affected by the COVID-19 pandemic due to precarious working conditions. This cross-sectional study was conducted in 2022 among 264 sex workers in Chiang Mai, Thailand, during the transition to the endemic phase, to evaluate their COVID-19 knowledge and preventive practices. Face-to-face interviews were used. Descriptive statistics were used to describe sample characteristics. Factors associated with knowledge and preventive practices were identified using the Mann–Whitney U test or Kruskal–Wallis test as appropriate. Independent factors associated with preventive practices were assessed through linear regression. The median scores for knowledge and preventive practices were 10 (interquartile range (IQR) = 9–10) and 5 (IQR = 3–5), respectively. In univariate analysis, females scored higher in knowledge than males. For preventive practices, females vs. males, older vs. younger, heterosexual vs. homosexual/bisexual, longer vs. shorter career, worked in massage parlors vs. pubs/bars, and having child vs. none showed higher rates. In multivariate analysis, being male (β = −1.87; 95%CI; −0.87 to −0.88) and single (β = −1.15; 95%CI; −2.28 to −0.02) were independent predictors of lower rates of preventive practices. Despite having good knowledge, certain groups of sex workers’ COVID-19 preventive behaviors remain inadequate, emphasizing the need for targeted interventions to enhance pandemic preparedness. Full article

(This article belongs to the Section Global Health)

25 pages, 8383 KB

Open AccessArticle

MemLoTrack: Enhancing TIR Anti-UAV Tracking with Memory-Integrated Low-Rank Adaptation

by Jae Kwan Park and Ji-Hyeong Han

Sensors 2025, 25(23), 7359; https://doi.org/10.3390/s25237359 (registering DOI) - 3 Dec 2025

Abstract

Tracking small, fast-moving unmanned aerial vehicles (UAVs) in thermal infrared (TIR) imagery is a significant challenge due to low-resolution targets, Dynamic Background Clutter, and frequent occlusions. To address this, we introduce MemLoTrack, a novel onestream Vision Transformer tracker that integrates a memory mechanism [...] Read more.

Tracking small, fast-moving unmanned aerial vehicles (UAVs) in thermal infrared (TIR) imagery is a significant challenge due to low-resolution targets, Dynamic Background Clutter, and frequent occlusions. To address this, we introduce MemLoTrack, a novel onestream Vision Transformer tracker that integrates a memory mechanism into a parameterefficient LoRA framework. MemLoTrack enhances a baseline tracker (LoRAT) with two key components: (i) a gated First-In, First-Out (FIFO) memory bank (MB) for temporal context aggregation and (ii) a lightweight Memory Attention Layer (MAL) for effective information retrieval. A key component of our method is a selective memory update policy, which commits a frame to the memory bank only when it satisfies both a classification confidence threshold (

τ

) and a Kalman filter-based motion consistency check. This gating mechanism robustly prevents memory contamination due to distractors, occlusions, and reappearance events. Our training is highly efficient, updating only the LoRA adapters, MAL, and prediction head while the pretrained DINOv2 backbone remains frozen. Evaluated on the challenging Anti-UAV410 benchmark, MemLoTrack (L_mem = 7,

τ

= 0.8) achieves an AUC of 63.6 and a State Accuracy (SA) of 64.0, representing a significant improvement over the LoRAT baseline by +1.4 AUC and +1.5 SA. Compared to the state-of-the-art method FocusTrack, MemLoTrack demonstrates superior robustness with higher AUC (63.6 vs. 62.8) and SA (64.0 vs. 63.9), while trading lower precision (P/P-Norm) scores. Furthermore, MemLoTrack operates at 153 FPS on a single RTX 4070 Ti SUPER, demonstrating that parameter-efficient fine-tuning with a selective memory mechanism is a powerful and deployable strategy for real-time Anti-UAV tracking in demanding TIR environments. Full article

(This article belongs to the Special Issue Vision Sensors for Object Detection and Tracking)

► Show Figures

Figure 1

30 pages, 6337 KB

Open AccessArticle

Geochemical and Geochronological Constraints on the Provenance and Heavy Metal Contamination of Beach Sediments Along the Gulf of Mexico, Mexico

by Itzamna Zaknite Flores-Ocampo, John Selvamony Armstrong-Altrin, Gloria Daniela Fernández-Guevara, Jayagopal Madhavaraju, Inna Valeria Acevedo-Granados, Barbara Yaneth Pérez-Alvarado, Sandra Elizabeth Ibarra-Rueda, Mayte Flores-Cortés and Isis Allanah Guadalupe-Díaz

Minerals 2025, 15(12), 1277; https://doi.org/10.3390/min15121277 (registering DOI) - 3 Dec 2025

Abstract

This study investigates the textural characteristics, mineral composition, and U–Pb ages of detrital zircon grains from the Playa Norte (PN) and Playa Tamiahua (PT) beach sediments along the Gulf of Mexico (GoM). The objective is to trace the sediment origin and to identify [...] Read more.

This study investigates the textural characteristics, mineral composition, and U–Pb ages of detrital zircon grains from the Playa Norte (PN) and Playa Tamiahua (PT) beach sediments along the Gulf of Mexico (GoM). The objective is to trace the sediment origin and to identify the possible environmental impacts in the coastal ecosystem. This work represents the first integrated provenance and geochemical analysis performed in these beaches, contributing to a broader regional sedimentological and geochronological database for the GoM. The results reveal distinct compositional and provenance signatures: PN sediments are rich in quartz (57.7% avg.), feldspars (15.7% avg.), and carbonate minerals (8.6% avg.), with zircon populations dominated by Proterozoic ages (~820–2200 Ma) and minor anthropogenic enrichment. In contrast, PT exhibits higher contents of quartz (78.6% avg.), andesine (9.6% avg.), and anorthite (8.5% avg.), with zircons mainly of Oligocene age (~32 Ma) and minimal contamination. Comparison with potential source regions indicates that PN sediments were derived primarily from the Sierra Madre Oriental, while PT sediments were originated from the Mesa Central and Eastern Mexican Alkaline Provinces. Overall, the findings demonstrate that, beyond littoral mixing and sediment recycling, the composition of GoM coastal sediments reflects the region’s complex tectono-sedimentary evolution and variable natural versus anthropogenic influences. PN is enriched in arsenic content, which is associated with agricultural activities and oil industries, while PT exhibits low values with no evidence of contamination. Meanwhile, Cr in PN suggests an anthropogenic input, which is linked to oil exploration activities in the GoM. Full article

(This article belongs to the Special Issue Tectonic Setting and Provenance of Sedimentary Rocks)

► Show Figures

Figure 1

49 pages, 6479 KB

Open AccessArticle

IoT-Driven Destination Prediction in Smart Urban Mobility: A Comparative Study of Markov Chains and Hidden Markov Models

by João Batista Firmino Junior, Francisco Dantas Nobre Neto, Bruno Neiva Moreno and Tiago Brasileiro Araújo

IoT 2025, 6(4), 75; https://doi.org/10.3390/iot6040075 (registering DOI) - 3 Dec 2025

Abstract

The increasing availability of IoT-enabled mobility data and intelligent transportation systems in Smart Cities demands efficient and interpretable models for destination prediction. This study presents a comparative analysis between Markov Chains and Hidden Markov Models applied to urban mobility trajectories, evaluated through mean [...] Read more.

The increasing availability of IoT-enabled mobility data and intelligent transportation systems in Smart Cities demands efficient and interpretable models for destination prediction. This study presents a comparative analysis between Markov Chains and Hidden Markov Models applied to urban mobility trajectories, evaluated through mean precision values. To ensure methodological rigor, the Smart Sampling with Data Filtering (SSDF) method was developed, integrating trajectory segmentation, spatial tessellation, frequency aggregation, and 10-fold cross-validation. Using data from 23 vehicles in the Vehicle Energy Dataset (VED) and a filtering threshold based on trajectory recurrence, the results show that the HMM achieved 61% precision versus 59% for Markov Chains (p = 0.0248). Incorporating day-of-week contextual information led to statistically significant precision improvements in 78.3% of cases for precision (95.7% for recall, 87.0% for F1-score). The remaining 21.7% indicate that model selection should balance model complexity and precision-efficiency trade-off. The proposed SSDF method establishes a replicable foundation for evaluating probabilistic models in IoT-based mobility systems, contributing to scalable, explainable, and sustainable Smart City transportation analytics. Full article

(This article belongs to the Special Issue IoT-Driven Smart Cities)

► Show Figures

Figure 1

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