Applied Sciences

16 pages, 14958 KB

Open AccessArticle

Spatiotemporal Evolution Characteristics and Influencing Factors of Urban Flood Resilience: The Case of Yangtze River Delta, East China

by Tiantian Gu, Hongtu Yan, Min Zhu, Zhi Kang and Peng Cui

Appl. Sci. 2025, 15(19), 10793; https://doi.org/10.3390/app151910793 - 7 Oct 2025

Abstract

Urban flood management is pivotal to the construction of resilient cities. However, investigation into the spatiotemporal evolution of urban flood resilience (UFR) and its influencing factors is insufficient. Aiming to address the challenge, this study establishes a multidimensional UFR indicator system grounded in [...] Read more.

Urban flood management is pivotal to the construction of resilient cities. However, investigation into the spatiotemporal evolution of urban flood resilience (UFR) and its influencing factors is insufficient. Aiming to address the challenge, this study establishes a multidimensional UFR indicator system grounded in the disaster resilience of place (DROP) model. Following the calculation of UFR through the entropy weighted technique for order preference by similarity to ideal solution (EW-TOPSIS) method, spatiotemporal evolution evaluation and factor detection are conducted. With panel data from the Yangtze River Delta Urban Agglomeration (YRDUA) over the period of 2012–2021, the results demonstrate overall UFR growth from a dominance of lowest-level and low-level cities to a more balanced distribution. Moreover, significant spatiotemporal heterogeneity is observed, with UFRs in cities adjacent to the Yangtze River higher than peripheral ones. Spatial clustering is significant until 2019, primarily manifested as High-High clusters along the Yangtze River and Low-Low clusters in northern Jiangsu and Anhui. Finally, factor detection identifies economic status, population size, environmental regulation, and drainage infrastructure as key influencing factors. These findings not only advance the understanding of UFR in urban agglomerations but also provide targeted recommendations for policymakers to enhance UFR. Full article

(This article belongs to the Special Issue Advances in Developing Urban Resilience Through Environmental and Earth Sciences)

► Show Figures

Figure 1

22 pages, 4797 KB

Open AccessArticle

Early Oral Cancer Detection with AI: Design and Implementation of a Deep Learning Image-Based Chatbot

by Pablo Ormeño-Arriagada, Gastón Márquez, Carla Taramasco, Gustavo Gatica, Juan Pablo Vasconez and Eduardo Navarro

Appl. Sci. 2025, 15(19), 10792; https://doi.org/10.3390/app151910792 - 7 Oct 2025

Abstract

Oral cancer remains a critical global health challenge, with delayed diagnosis driving high morbidity and mortality. Despite progress in artificial intelligence, computer vision, and medical imaging, early detection tools that are accessible, explainable, and designed for patient engagement remain limited. This study presents [...] Read more.

Oral cancer remains a critical global health challenge, with delayed diagnosis driving high morbidity and mortality. Despite progress in artificial intelligence, computer vision, and medical imaging, early detection tools that are accessible, explainable, and designed for patient engagement remain limited. This study presents a novel system that combines a patient-centred chatbot with a deep learning framework to support early diagnosis, symptom triage, and health education. The system integrates convolutional neural networks, class activation mapping, and natural language processing within a conversational interface. Five deep learning models were evaluated (CNN, DenseNet121, DenseNet169, DenseNet201, and InceptionV3) using two balanced public datasets. Model performance was assessed using accuracy, sensitivity, specificity, diagnostic odds ratio (DOR), and Cohen’s Kappa. InceptionV3 consistently outperformed the other models across these metrics, achieving the highest diagnostic accuracy (77.6%) and DOR (20.67), and was selected as the core engine of the chatbot’s diagnostic module. The deployed chatbot provides real-time image assessments and personalised conversational support via multilingual web and mobile platforms. By combining automated image interpretation with interactive guidance, the system promotes timely consultation and informed decision-making. It offers a prototype for a chatbot, which is scalable and serves as a low-cost solution for underserved populations and demonstrates strong potential for integration into digital health pathways. Importantly, the system is not intended to function as a formal screening tool or replace clinical diagnosis; rather, it provides preliminary guidance to encourage early medical consultation and informed health decisions. Full article

► Show Figures

Figure 1

26 pages, 1244 KB

Open AccessReview

Neuroprotective Bioactive Compounds from Marine Algae and Their By-Products Against Cerebral Ischemia–Reperfusion Injury: A Comprehensive Review

by Joon Ha Park

Appl. Sci. 2025, 15(19), 10791; https://doi.org/10.3390/app151910791 - 7 Oct 2025

Abstract

Cerebral ischemia–reperfusion (I/R) injury is a leading cause of death and long-term disability worldwide, characterized by a complex interplay of pathophysiological mechanisms and currently limited therapeutic options. This critical unmet need underscores the importance of exploring novel multi-targeted neuroprotective agents. Marine algae represent [...] Read more.

Cerebral ischemia–reperfusion (I/R) injury is a leading cause of death and long-term disability worldwide, characterized by a complex interplay of pathophysiological mechanisms and currently limited therapeutic options. This critical unmet need underscores the importance of exploring novel multi-targeted neuroprotective agents. Marine algae represent a rich and underexplored source of structurally diverse bioactive compounds with promising therapeutic potential against cerebral I/R injury. This comprehensive review systematically summarizes the preclinical evidence on the neuroprotective effects and underlying mechanisms of key bioactive compounds found in marine algae, including polysaccharides (e.g., fucoidan, laminarin, porphyran), carotenoids (e.g., astaxanthin, fucoxanthin, lutein, zeaxanthin), polyphenols (e.g., dieckol, phlorotannins), and sterols (e.g., β-sitosterol). These compounds consistently demonstrate significant efficacy across various in vitro and in vivo models, primarily through multifaceted actions encompassing anti-excitotoxic, antioxidant, anti-inflammatory, and anti-apoptotic effects, as well as the modulation of crucial signaling pathways and preservation of blood–brain barrier integrity. While the existing preclinical evidence is highly promising, successful clinical translation necessitates further rigorous research to overcome challenges related to precise molecular understanding, translational relevance, pharmacokinetics, and safety. Beyond their pharmacological significance, the sustainable utilization of marine by-products as renewable sources of bioactive agents further highlights their dual value, offering not only novel therapeutic avenues for cerebral I/R injury but also contributing to marine resource valorization. Full article

(This article belongs to the Special Issue Utilization of Marine By-Products)

► Show Figures

Figure 1

14 pages, 1873 KB

Open AccessArticle

CycleGAN with Atrous Spatial Pyramid Pooling and Attention-Enhanced MobileNetV4 for Tomato Disease Recognition Under Limited Training Data

by Yueming Jiang, Taizeng Jiang, Chunyan Song and Jian Wang

Appl. Sci. 2025, 15(19), 10790; https://doi.org/10.3390/app151910790 - 7 Oct 2025

Abstract

To address the challenges of poor model generalization and suboptimal recognition accuracy stemming from limited and imbalanced sample sizes in tomato leaf disease identification, this study proposes a novel recognition strategy. This approach synergistically combines an enhanced image augmentation method based on generative [...] Read more.

To address the challenges of poor model generalization and suboptimal recognition accuracy stemming from limited and imbalanced sample sizes in tomato leaf disease identification, this study proposes a novel recognition strategy. This approach synergistically combines an enhanced image augmentation method based on generative adversarial networks with a lightweight deep learning model. Initially, an Atrous Spatial Pyramid Pooling (ASPP) module is integrated into the CycleGAN framework. This integration enhances the generator’s capacity to model multi-scale pathological lesion features, thereby significantly improving the diversity and realism of synthesized images. Subsequently, the Convolutional Block Attention Module (CBAM), incorporating both channel and spatial attention mechanisms, is embedded into the MobileNetV4 architecture. This enhancement boosts the model’s ability to focus on critical disease regions. Experimental results demonstrate that the proposed ASPP-CycleGAN significantly outperforms the original CycleGAN across multiple disease image generation tasks. Furthermore, the developed CBAM-MobileNetV4 model achieves a remarkable average recognition accuracy exceeding 97% for common tomato diseases, including early blight, late blight, and mosaic disease, representing a 1.86% improvement over the baseline MobileNetV4. The findings indicate that the proposed method offers exceptional data augmentation capabilities and classification performance under small-sample learning conditions, providing an effective technical foundation for the intelligent identification and control of tomato leaf diseases. Full article

(This article belongs to the Section Agricultural Science and Technology)

19 pages, 1878 KB

Open AccessArticle

The Potential of Bilberry and Blackcurrant Juices as a Source of Colorants in Intelligent Pectin Films

by Anna Pakulska, Magdalena Mikus, Magdalena Karwacka and Sabina Galus

Appl. Sci. 2025, 15(19), 10789; https://doi.org/10.3390/app151910789 - 7 Oct 2025

Abstract

The aim of this study was to develop biodegradable pectin films enriched with anthocyanin-rich fruit juices and evaluate their functional properties. Films were prepared with bilberry and blackcurrant juices, and their color response to pH, mechanical performance, thermal stability, and water vapor permeability [...] Read more.

The aim of this study was to develop biodegradable pectin films enriched with anthocyanin-rich fruit juices and evaluate their functional properties. Films were prepared with bilberry and blackcurrant juices, and their color response to pH, mechanical performance, thermal stability, and water vapor permeability were analyzed. The incorporation of juices significantly affected the films’ color, with ΔE values ranging from 8.41 to 39.24 for blackcurrant and 36.60 to 59.59 for wild bilberry juice, showing clear visual differences. Increasing juice concentration from 5% to 10% enhanced color intensity and opacity, with the highest opacity (12.90 a.u./mm) observed for films containing 2% pectin and 10% bilberry juice. Mechanical testing indicated reduced tensile strength after juice addition, with the lowest elongation (11.90%) noted for films with 2% pectin and 5% blackcurrant juice. The lowest water vapor permeability (7.43·10⁻¹¹ g/m·s·Pa) was recorded for films with 2% pectin. Thermal analysis revealed greater mass loss in juice-enriched films (40–44.5%) compared to controls (37.6%), reflecting the presence of volatile compounds. pH testing confirmed the films’ indicator function, with red coloration at pH 2 and shifts toward blue-grey (bilberry) or orange-green (blackcurrant) at pH 8. These findings demonstrate that pectin films enriched with dark red fruit juices exhibit promising potential for smart food packaging applications. Full article

(This article belongs to the Special Issue Functional Food: From Discovery to Application)

► Show Figures

Figure 1

10 pages, 1117 KB

Open AccessArticle

Effects of Dual-Tasking on Center-of-Pressure Dynamics and Spectral Balance Control

by Lei Zhang, Qingjie Wang, Yuanyuan Ren and Aming Lu

Appl. Sci. 2025, 15(19), 10788; https://doi.org/10.3390/app151910788 - 7 Oct 2025

Abstract

Cognitive tasks play a pivotal role in posture control among young adults. This study examined how concurrent cognitive tasks alter balance stability and sensory integration during single-leg stance by analyzing center-of-pressure trajectories and wavelet spectra to elucidate the neurobehavioral mechanisms underlying dual-task balance [...] Read more.

Cognitive tasks play a pivotal role in posture control among young adults. This study examined how concurrent cognitive tasks alter balance stability and sensory integration during single-leg stance by analyzing center-of-pressure trajectories and wavelet spectra to elucidate the neurobehavioral mechanisms underlying dual-task balance degradation. A cohort of 24 young adults completed both single postural control tasks and dual cognitive–postural tasks on a force plate. COP data and wavelet decomposition energy were computed and analyzed. The results revealed significant differences between the dual-task and single-task groups for Lxy, Ly, Vxy, and Vy (p < 0.05). Energy content analysis showed that the dual-task group had significantly different energy ratios across four frequency bands along the x-axis (p < 0.05). Our findings showed that dual-task conditions impair postural control in young adults, increasing anteroposterior sway and altering mediolateral energy patterns. This suggests a shift toward proprioceptive reliance during cognitive division, revealing cognitive–postural interference. These results support using dual-task assessments for fall risk evaluation and inform interventions for populations requiring cognitive–motor integration. Full article

(This article belongs to the Section Applied Biosciences and Bioengineering)

► Show Figures

Figure 1

24 pages, 1207 KB

Open AccessArticle

A Reputation-Enhanced Shapley–FAHP Method for Multi-Dimensional Food Safety Evaluation

by Xiaobo Yang, Hanning Wei, Binghui Guo, Min Zuo, Lipo Mo and Haiwei Gao

Appl. Sci. 2025, 15(19), 10787; https://doi.org/10.3390/app151910787 - 7 Oct 2025

Abstract

Ensuring food safety in complex supply chains requires evaluation frameworks that integrate multiple indicators, account for their interdependencies, and incorporate historical performance. This study proposes a novel RM–Shapley–FAHP framework that combines the Fuzzy Analytic Hierarchy Process, Shapley value contribution analysis, and a reputation [...] Read more.

Ensuring food safety in complex supply chains requires evaluation frameworks that integrate multiple indicators, account for their interdependencies, and incorporate historical performance. This study proposes a novel RM–Shapley–FAHP framework that combines the Fuzzy Analytic Hierarchy Process, Shapley value contribution analysis, and a reputation decay mechanism to construct a dynamic, multi-year assessment model. The framework evaluates six governance subsystems, mitigates indicator redundancy, and links past performance to current risk posture. Applied to a leading food enterprise over three years, the method demonstrated superior consistency, interpretability, and operational relevance compared to FAHP, entropy weighting, and equal-weight baselines. The results demonstrate that RM–Shapley–FAHP framework effectively supports balanced development in food safety governance by capturing temporal dynamics and interdependencies, offering interpretable and operationally relevant guidance for decision makers. In future work, this framework may be extended with machine learning to improve adaptability for multi-dimensional and time-series evaluations, noted here as a research prospect rather than a present contribution. Full article

22 pages, 5534 KB

Open AccessArticle

GIS-Based Assessment of Photovoltaic and Green Roof Potential in Iași, Romania

by Otilia Pitulac, Constantin Chirilă, Florian Stătescu and Nicolae Marcoie

Appl. Sci. 2025, 15(19), 10786; https://doi.org/10.3390/app151910786 - 7 Oct 2025

Abstract

Urban areas are increasingly challenged by the combined effects of climate change, rapid population growth, and high energy demand. The integration of renewable energy systems, such as photovoltaic (PV) panels, and nature-based solutions, such as green roofs, represents a key strategy for sustainable [...] Read more.

Urban areas are increasingly challenged by the combined effects of climate change, rapid population growth, and high energy demand. The integration of renewable energy systems, such as photovoltaic (PV) panels, and nature-based solutions, such as green roofs, represents a key strategy for sustainable urban development. This study evaluates the spatial potential for PV and green roof implementation in Iași, Romania, using moderate to high-resolution geospatial datasets, including the ALOS AW3D30 Digital Surface Model (DSM) and the Copernicus Urban Atlas 2018, processed in ArcMap 10.8.1 and ArcGIS Pro 2.6.0. Solar radiation was computed using the Area Solar Radiation tool for the average year 2023, while roof typology (flat vs. pitched) was derived from slope analysis. Results show significant spatial heterogeneity. The Copou neighborhood has the highest PV-suitable roof share (73.6%) and also leads in green roof potential (46.6%). Integrating PV and green roofs can provide synergistic benefits, improving energy performance, mitigating urban heat islands, managing stormwater, and enhancing biodiversity. These findings provide actionable insights for urban planners and policymakers aiming to prioritize green infrastructure investments and accelerate the local energy transition. Full article

(This article belongs to the Special Issue Applications of Digital Modeling, 3D Modeling, and Simulation Visualization for Urban Development)

► Show Figures

Figure 1

21 pages, 2942 KB

Open AccessArticle

A Real-Time Six-Axis Electromagnetic Field Monitoring System with Wireless Transmission and Intelligent Vector Analysis for Power Environments

by Xiran Zheng, Xuecong Li, Yucheng Mai, Wendong Li, Meiqi Chen, Gengjie Huang, Zheng Zhang and Yue Wang

Appl. Sci. 2025, 15(19), 10785; https://doi.org/10.3390/app151910785 - 7 Oct 2025

Abstract

Accurate and real-time monitoring of low-frequency electromagnetic field (EMF) is essential in power and industrial environments, yet most conventional approaches still suffer from limited spatial coverage, manual operation, and insufficient digitization. To address these challenges, this paper proposes an intelligent EMF monitoring system [...] Read more.

Accurate and real-time monitoring of low-frequency electromagnetic field (EMF) is essential in power and industrial environments, yet most conventional approaches still suffer from limited spatial coverage, manual operation, and insufficient digitization. To address these challenges, this paper proposes an intelligent EMF monitoring system that integrates six-axis magnetic field sensing, temperature compensation, vector synthesis, Sub-1 GHz wireless communication, and real-time data visualization. The system supports simultaneous measurement of both AC and DC magnetic fields across the 30 Hz–100 kHz range, with specific optimization for power-frequency conditions (50/60 Hz). Designed with modular integration and low power consumption, it is suitable for portable deployment in field scenarios. Comprehensive laboratory and substation tests demonstrate high accuracy, with maximum measurement errors of 1.17% under zero-field and 1.42% under applied-field conditions—well below the ±5% tolerance defined by international standards. Wireless performance tests further confirm stable long-distance communication, achieving ranges of up to 5 km without significant transmission errors, while overall system measurement error reached as low as 0.015%. These results verify the system’s robustness, fidelity, and compliance with international safety standards. Overall, the proposed platform provides a practical and scalable solution for intelligent EMF monitoring, offering strong potential for deployment in industrial environments and infrastructure-critical applications. Full article

► Show Figures

Figure 1

29 pages, 1231 KB

Open AccessReview

X-Ray Absorption and Emission Spectroscopy in Pharmaceutical Applications: From Local Atomic Structure Elucidation to Protein-Metal Complex Analysis—A Review

by Klaudia Wojtaszek, Krzysztof Tyrała and Ewelina Błońska-Sikora

Appl. Sci. 2025, 15(19), 10784; https://doi.org/10.3390/app151910784 - 7 Oct 2025

Abstract

X-ray absorption spectroscopy (XAS) and X-ray emission spectroscopy (XES) are analytical techniques enabling precise analysis of the electronic structure and local atomic environment in chemical compounds and materials. Their application spans materials science, chemistry, biology, and environmental sciences, supporting studies on catalytic mechanisms, [...] Read more.

X-ray absorption spectroscopy (XAS) and X-ray emission spectroscopy (XES) are analytical techniques enabling precise analysis of the electronic structure and local atomic environment in chemical compounds and materials. Their application spans materials science, chemistry, biology, and environmental sciences, supporting studies on catalytic mechanisms, redox processes, and metal speciation. A key advantage of both techniques is element selectivity, allowing the analysis of specific elements without matrix interference. Their high sensitivity to chemical state and coordination enables determination of oxidation states, electronic configuration, and local geometry. These methods are applicable to solids, liquids, and gases without special sample preparation. Modern XAS and XES studies are typically performed using synchrotron radiation, which provides an intense, monochromatic X-ray source and allows advanced in situ and operando experiments. Sub-techniques such as XANES (X-ray absorption near-edge structure), EXAFS (Extended X-ray Absorption Fine Structure), and RIXS (resonant inelastic X-ray scattering) offer enhanced insights into oxidation states, local structure, and electronic excitations. Despite their broad scientific use, applications in pharmaceutical research remain limited. Nevertheless, recent studies highlight their potential in analyzing crystalline active pharmaceutical ingredients (APIs), drug–biomolecule interactions, and differences in drug activity. This review introduces the fundamental aspects of XAS and XES, with an emphasis on practical considerations for pharmaceutical applications, including experimental design and basic spectral interpretation. Full article

(This article belongs to the Special Issue Contemporary Pharmacy: Advances and Challenges)

28 pages, 1558 KB

Open AccessArticle

Multi-Fidelity Neural Network-Aided Multi-Objective Optimization Framework for Shell Structure Dynamic Analysis

by Bartosz Miller and Leonard Ziemiański

Appl. Sci. 2025, 15(19), 10783; https://doi.org/10.3390/app151910783 - 7 Oct 2025

Abstract

We address surrogate-assisted multi-objective optimization for computationally expensive structural designs. The testbed is an axisymmetric laminated composite shell whose geometry, ply angles, and plywise materials are optimized to simultaneously (i) maximize separation of selected natural frequencies from a known excitation and (ii) minimize [...] Read more.

We address surrogate-assisted multi-objective optimization for computationally expensive structural designs. The testbed is an axisymmetric laminated composite shell whose geometry, ply angles, and plywise materials are optimized to simultaneously (i) maximize separation of selected natural frequencies from a known excitation and (ii) minimize material cost. To reduce high-fidelity (HF) finite element evaluations, we develop a deep neural network surrogate framework with three variants: an HF-only baseline; a multi-fidelity (MF) pipeline using an auxiliary refinement network to convert abundant low-fidelity (LF) data into pseudo-HF labels for a single-fidelity evaluator; and a cascaded ensemble that emulates HF responses and then maps them to pseudo-experimental targets. During optimization, only surrogates are queried—no FEM calls—while final designs are verified by FEM. Pareto-front quality is quantified primarily by a normalized relative hypervolume indicator computed against an envelope approximation of the True Pareto Front, complemented where appropriate by standard indicators. A controlled training protocol and common validation regime isolate the effect of fidelity strategy from architectural choices. Results show that MF variants markedly reduce HF data requirements and improve Pareto-front quality over the HF-only baseline, offering a practical route to scalable, accurate design under strict computational budgets. Full article

(This article belongs to the Special Issue Innovations in Artificial Neural Network Applications)

► Show Figures

Figure 1

18 pages, 5236 KB

Open AccessArticle

Three-Dimensional Accuracy of Clear Aligner Attachment Reproduction Using a Standardized In-House Protocol: An In Vitro Study

by U-Hyeong Cho and Hyo-Sang Park

Appl. Sci. 2025, 15(19), 10782; https://doi.org/10.3390/app151910782 - 7 Oct 2025

Abstract

This in vitro study aimed to quantitatively evaluate the accuracy of reproducing attachments for clear aligner therapy (CAT) using a standardized in-house fabrication protocol and to analyze discrepancies across maxillary tooth types. A custom attachment was designed on a symmetrical master model, and [...] Read more.

This in vitro study aimed to quantitatively evaluate the accuracy of reproducing attachments for clear aligner therapy (CAT) using a standardized in-house fabrication protocol and to analyze discrepancies across maxillary tooth types. A custom attachment was designed on a symmetrical master model, and 30 experimental models were fabricated by three-dimensional (3D) printing, template construction, and bonding. Following scanning and superimposition, dimensional, angular, and positional deviations were quantified and statistically analyzed (p < 0.05). Results showed minor mean discrepancies but a consistent pattern of under-reproduction, most evident in the mesial and distal wall angles, as well as in the gingival bevel angle and attachment height. A significant trend was observed in the occlusal bevel, demonstrating marked extrusion in the anterior region that decreased posteriorly. Positional errors were minimal mesiodistally but substantial in the lingual and occlusal directions, with magnitudes varying by tooth type. In conclusion, this study identified consistent, predictable inaccuracies in a simulated in-house attachment reproduction protocol. These findings indicate that similar deviations may occur clinically, potentially affecting the predictability of CAT. Full article

(This article belongs to the Special Issue Advances in Orthodontics and Dentofacial Orthopedics)

23 pages, 1934 KB

Open AccessArticle

INTU-AI: Digitalization of Police Interrogation Supported by Artificial Intelligence

by José Pinto Garcia, Carlos Grilo, Patrício Domingues and Rolando Miragaia

Appl. Sci. 2025, 15(19), 10781; https://doi.org/10.3390/app151910781 - 7 Oct 2025

Abstract

Traditional police interrogation processes remain largely time-consuming and reliant on substantial human effort for both analysis and documentation. Intuition Artificial Intelligence (INTU-AI) is a Windows application designed to digitalize the administrative workflow associated with police interrogations, while enhancing procedural efficiency through the integration [...] Read more.

Traditional police interrogation processes remain largely time-consuming and reliant on substantial human effort for both analysis and documentation. Intuition Artificial Intelligence (INTU-AI) is a Windows application designed to digitalize the administrative workflow associated with police interrogations, while enhancing procedural efficiency through the integration of AI-driven emotion recognition models. The system employs a multimodal approach that captures and analyzes emotional states using three primary vectors: Facial Expression Recognition (FER), Speech Emotion Recognition (SER), and Text-based Emotion Analysis (TEA). This triangulated methodology aims to identify emotional inconsistencies and detect potential suppression or concealment of affective responses by interviewees. INTU-AI serves as a decision-support tool rather than a replacement for human judgment. By automating bureaucratic tasks, it allows investigators to focus on critical aspects of the interrogation process. The system was validated in practical training sessions with inspectors and with a 12-question questionnaire. The results indicate a strong acceptance of the system in terms of its usability, existing functionalities, practical utility of the program, user experience, and open-ended qualitative responses. Full article

(This article belongs to the Special Issue Digital Transformation in Information Systems)

► Show Figures

Figure 1

21 pages, 1716 KB

Open AccessArticle

LAI-YOLO: Towards Lightweight and Accurate Insulator Anomaly Detection via Selective Weighted Feature Fusion

by Jianan Qu, Zhiliang Zhu, Ziang Jiang, Congjie Wen and Yijian Weng

Appl. Sci. 2025, 15(19), 10780; https://doi.org/10.3390/app151910780 - 7 Oct 2025

Abstract

While insulator integrity is critical for power grid stability, prevailing detection algorithms often rely on computationally intensive models incompatible with resource-constrained edge devices like unmanned aerial vehicles (UAVs). Key limitations—including redundant feature interference, inadequate sensitivity to small targets, rigid fusion weights, and sample [...] Read more.

While insulator integrity is critical for power grid stability, prevailing detection algorithms often rely on computationally intensive models incompatible with resource-constrained edge devices like unmanned aerial vehicles (UAVs). Key limitations—including redundant feature interference, inadequate sensitivity to small targets, rigid fusion weights, and sample imbalance—further restrict practical deployment. To address those problems, this study presents a lightweight insulator anomaly detection algorithm, LAI-YOLO. First, the SqueezeGate-C3k2 (SG-C3k2) module, equipped with an adaptive gating mechanism, is incorporated into the Backbone network to reduce redundant information during feature extraction. Secondly, we propose a High-level Screening–Feature Weighted Feature Pyramid Network (HS-WFPN) to replace FPN+PAN via selective weighted feature fusion, enabling dynamic cross-scale integration and enhanced small-target detection. Then, a reconstructed lightweight detection head coupled with Slide Weighted Focaler Loss (SWFocalerLoss) mitigates performance degradation from sample imbalance. Ultimately, the layer adaptation for the magnitude-based pruning (LAMP) technique slashes computational demands without sacrificing detection prowess. Experimental results on our insulator anomaly dataset demonstrate that the improved model achieves higher efficacy in identifying insulator anomalies, with mAP@0.5 increasing from 88.2% to 91.1%, while model parameters and FLOPs are diminished to 45.7% and 53.9% of the baseline, respectively. This efficiency facilitates the deployment of edge devices and highlights the method’s considerable application potential. Full article

(This article belongs to the Special Issue Advances in Wireless Networks and Mobile Communication)

► Show Figures

Figure 1

32 pages, 16950 KB

Open AccessArticle

Regression-Based Performance Prediction in Asphalt Mixture Design and Input Analysis with SHAP

by Kemal Muhammet Erten and Remzi Gürfidan

Appl. Sci. 2025, 15(19), 10779; https://doi.org/10.3390/app151910779 - 7 Oct 2025

Abstract

The primary aim of this study is to predict the Marshall stability and flow values of hot-mix asphalt samples prepared according to the Marshall design method using regression-based machine learning algorithms. To overcome the limited number of experimental observations, synthetic data generation was [...] Read more.

The primary aim of this study is to predict the Marshall stability and flow values of hot-mix asphalt samples prepared according to the Marshall design method using regression-based machine learning algorithms. To overcome the limited number of experimental observations, synthetic data generation was applied using the Conditional Tabular Generative Adversarial Network (CTGAN), while the structural consistency of the generated data was validated through Principal Component Analysis (PCA). Two datasets containing 17 physical and mechanical input variables were analyzed, and multiple regression models were compared, including Extra Trees, Random Forest, Gradient Boosting, AdaBoost, and K-Nearest Neighbors. Among these, the Extra Trees Regressor consistently achieved the best results with near-perfect accuracy in flow predictions (MAE ≈ 4.06 × 10⁻¹⁵, RMSE ≈ 4.97 × 10⁻¹⁵, Accuracy ≈ 99.99%) and high performance in stability predictions (MAE = 109.52, RMSE = 150.67, accuracy = 90.45%). Furthermore, model interpretability was ensured by applying SHapley Additive Explanations (SHAP), which revealed that parameters such as softening point, VMA, penetration, and void ratios were the most influential features. These findings demonstrate that regression-based ensemble models, combined with synthetic data augmentation and explainable AI methods, can serve as reliable and interpretable tools in asphalt mixture design. Full article

► Show Figures

Figure 1

21 pages, 1768 KB

Open AccessReview

Evolution of Deep Learning Approaches in UAV-Based Crop Leaf Disease Detection: A Web of Science Review

by Dorijan Radočaj, Petra Radočaj, Ivan Plaščak and Mladen Jurišić

Appl. Sci. 2025, 15(19), 10778; https://doi.org/10.3390/app151910778 - 7 Oct 2025

Abstract

The integration of unmanned aerial vehicles (UAVs) and deep learning (DL) has significantly advanced crop disease detection by enabling scalable, high-resolution, and near real-time monitoring within precision agriculture. This systematic review analyzes peer-reviewed literature indexed in the Web of Science Core Collection as [...] Read more.

The integration of unmanned aerial vehicles (UAVs) and deep learning (DL) has significantly advanced crop disease detection by enabling scalable, high-resolution, and near real-time monitoring within precision agriculture. This systematic review analyzes peer-reviewed literature indexed in the Web of Science Core Collection as articles or proceeding papers through 2024. The main selection criterion was combining “unmanned aerial vehicle*” OR “UAV” OR “drone” with “deep learning”, “agriculture” and “leaf disease” OR “crop disease”. Results show a marked surge in publications after 2019, with China, the United States, and India leading research contributions. Multirotor UAVs equipped with RGB sensors are predominantly used due to their affordability and spatial resolution, while hyperspectral imaging is gaining traction for its enhanced spectral diagnostic capability. Convolutional neural networks (CNNs), along with emerging transformer-based and hybrid models, demonstrate high detection performance, often achieving F1-scores above 95%. However, critical challenges persist, including limited annotated datasets for rare diseases, high computational costs of hyperspectral data processing, and the absence of standardized evaluation frameworks. Addressing these issues will require the development of lightweight DL architectures optimized for edge computing, improved multimodal data fusion techniques, and the creation of publicly available, annotated benchmark datasets. Advancements in these areas are vital for translating current research into practical, scalable solutions that support sustainable and data-driven agricultural practices worldwide. Full article

► Show Figures

Figure 1

13 pages, 256 KB

Open AccessArticle

Effect of Artificial Diet Modification with Dextrose on the Growth and Fatty Acid Composition of Tenebrio molitor Larvae for Biodiesel Production

by Miguel Alejandro Flores-Sánchez, Amanda Kim Rico-Chávez, Marco Martín González-Chávez, Rodolfo Figueroa-Brito, Juan Campos-Guillen, Carlos Eduardo Zavala-Gómez, Aldo Amaro-Reyes, Joel de Jesús Barba-Franco and Miguel Angel Ramos-López

Appl. Sci. 2025, 15(19), 10777; https://doi.org/10.3390/app151910777 - 7 Oct 2025

Abstract

Tenebrio molitor larvae are a promising source for the next generation of liquid biofuels. However, the conditions and processes required for rearing this insect for biodiesel production need to be investigated. In this study, the effect of dextrose addition to an artificial diet [...] Read more.

Tenebrio molitor larvae are a promising source for the next generation of liquid biofuels. However, the conditions and processes required for rearing this insect for biodiesel production need to be investigated. In this study, the effect of dextrose addition to an artificial diet in the oil and biodiesel yield was evaluated. Larvae were fed artificial diets modified with 3, 6, 9, and 15% dextrose. Survival rate, mean dry weight, and oil yield were registered. The 15% dextrose addition resulted in 75% survival, 25 mg individual dry weight, and 29% oil yield. The main components of this oil were palmitic acid (14%), oleic acid (37%), and linoleic acid (20%). With the addition of dextrose, the total saturated fatty acids increased 13% and the polyunsaturated fatty acids decreased 16% compared to the control. However, with the transesterification reaction, the fatty acid methyl esters remained similar for both treatments, with methyl oleate, methyl linoleate, and methyl linolenate as major components. This suggest that the transesterification reaction was incomplete, due to the oil/methanol ratio or the catalyst. Thus, even though a more balanced saturated/unsaturated fatty acid profile can be achieved through dextrose addition, another transesterification method should be tested to obtain a complete reaction. Full article

29 pages, 3544 KB

Open AccessReview

Modern Trends in the Application of Electronic Nose Systems: A Review

by Stefan Ivanov, Jacek Łukasz Wilk-Jakubowski, Leszek Ciopiński, Łukasz Pawlik, Grzegorz Wilk-Jakubowski and Georgi Mihalev

Appl. Sci. 2025, 15(19), 10776; https://doi.org/10.3390/app151910776 - 7 Oct 2025

Abstract

Electronic nose (e-nose) systems have emerged as transformative tools for odor and gas analysis, leveraging advances in nanomaterials, sensor arrays, and machine learning (ML) to mimic biological olfaction. This review synthesizes recent developments in e-nose technology, focusing on innovations in sensor design (e.g., [...] Read more.

Electronic nose (e-nose) systems have emerged as transformative tools for odor and gas analysis, leveraging advances in nanomaterials, sensor arrays, and machine learning (ML) to mimic biological olfaction. This review synthesizes recent developments in e-nose technology, focusing on innovations in sensor design (e.g., graphene-based nanomaterials, MEMS, and optical sensors), drift compensation techniques, and AI-driven data processing. We highlight key applications across healthcare (e.g., non-invasive disease diagnostics via breath analysis), food quality monitoring (e.g., spoilage detection and authenticity verification), and environmental management (e.g., pollution tracking and wastewater treatment). Despite progress, challenges such as sensor selectivity, long-term stability, and standardization persist. The paper underscores the potential of e-noses to replace conventional analytical methods, offering portability, real-time operation, and cost-effectiveness. Future directions include scalable fabrication, robust ML models, and IoT integration to expand their practical adoption. Full article

(This article belongs to the Special Issue Gas Sensors: Optimization and Applications)

► Show Figures

Figure 1

17 pages, 2676 KB

Open AccessArticle

Stiffness and Strength of Scots Pine Wood Under Compression Perpendicular to the Grain and Rolling Shear Loading

by Shaheda T. Akter, Anders Olsson and Thomas K. Bader

Appl. Sci. 2025, 15(19), 10775; https://doi.org/10.3390/app151910775 - 7 Oct 2025

Abstract

To increase and optimize the use of wood in structural elements, a deep understanding of its mechanical behavior is necessary. The transverse material properties of wood are particularly important for mass timber construction and for utilizing wood as a strengthening material in timber [...] Read more.

To increase and optimize the use of wood in structural elements, a deep understanding of its mechanical behavior is necessary. The transverse material properties of wood are particularly important for mass timber construction and for utilizing wood as a strengthening material in timber connections. This study experimentally determined the stiffness and strength of Scots pine wood under compression perpendicular to the grain and rolling shear loading, as well as their dependence on the annual ring structure. A previously established biaxial test configuration was employed for this purpose. The modulus of elasticity in the radial direction was found to be about twice that in the tangential direction (687 vs. 372 N/mm²), although the strength in the tangential direction (5.19 N/mm²) was comparatively higher than that in the radial direction (4.70 N/mm²). For rolling shear, especially for the rolling shear modulus, a large variation was found, and its relationship with annual ring structure was assessed. The obtained RS modulus ranged from 50 to 254 N/mm², while RS strength was found to be between 2.14 and 4.61 N/mm². The results aligned well with previous findings. Full article

(This article belongs to the Special Issue International Conference Wood Science and Engineering in the Third Millennium–ICWSE 2025)

► Show Figures

Figure 1

Journal Description

Applied Sciences

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI