Saved Queries

Carpinus betulus is an important ornamental landscape tree species with colorful foliage. It is widely used in landscaping due to its upright tree shape, significant seasonal changes, and good tolerance to pruning. Propagation methods for C. betulus include grafting, cutting, and seeding. However, the germination rate of seeding is low, and the rooting of cuttings is difficult; moreover, plant tissue culture techniques are complex, and the key technologies have not been disclosed. Grafting has therefore become the primary means of propagation. However, enabling the rapid reproduction of C. betulus through appropriate grafting methods and in appropriate environments remains an urgent issue to be addressed. In this study, Carpinus turczaninowii was used as a rootstock to graft C. betulus, and the effects of the grafting periods, technique, and environmental conditions on the survival rate of grafted C. betulus were discussed. The results showed that branch grafting (cleft graft and whip-and-tongue graft) performed in March to April and August to November resulted in the highest survival rates, whereas budding grafts (chip budding and patch budding) were more suitable in May and June. Increasing ambient humidity was a key measure for improving graft survival rates and germination rates. In terms of grafting survival rate, germination rate, and leaf growth, humidification and treatment with 60–70% light transmission had better results than treatment with natural humidity or 20–30% light transmission and full light treatment under humidification conditions. Under low-light conditions, increasing air humidity had a particularly pronounced effect on promoting the growth of grafted seedling branches. In the future, further research should be conducted on the molecular mechanism mediated by soil environment and temperature changes for the successful grafting of C. betulus, providing a theoretical basis for the propagation and cultivation of C. betulus. Full article

(This article belongs to the Section Plant Physiology and Metabolism)

►▼ Show Figures

Figure 1

21 pages, 651 KB

Open AccessArticle

Do Integrated CMD Management Practices Increase Cassava Yields? A Local Average Treatment Effect Analysis from Burkina Faso

by Agnès Ouédraogo, Eveline Sawadogo-Compaore, Ezechiel Bionimian Tibiri, Noël Thiombiano, Adama Sagnon, Seydou Sawadogo, Fidèle Tiendrébéogo and Justin Simon Pita

Agriculture 2026, 16(4), 441; https://doi.org/10.3390/agriculture16040441 - 13 Feb 2026

Abstract

Cassava mosaic disease (CMD) is a major constraint to cassava production in sub-Saharan Africa, particularly in Burkina Faso, where it poses a serious threat to rural food security. This study examined the impact of adopting innovative cassava mosaic disease management practices on cassava yields in the Guiriko and Nando regions of Burkina Faso. To address potential biases arising from differences in characteristics between adopters and non-adopters, an econometric approach based on the instrumental variables (IV) method within a counterfactual framework was employed to estimate the local average treatment effect (LATE). The data were drawn from a survey conducted in September 2023 among 511 cassava producers. The results indicate that the adoption of innovative cassava mosaic disease management practices had a positive and statistically significant effect on agricultural yields. Productivity gains were estimated at 29% in the Guiriko region and 41% in the Nando region, highlighting spatial heterogeneity in impacts. These findings suggest that promoting the diffusion of such practices can substantially improve cassava productivity and reduce the vulnerability of rural households. In addition, the analysis showed that socioeconomic and technical factors, including farmers’ age, membership in cassava producer organizations, household income levels, and the use of chemical fertilizers, also influence productivity outcomes. Overall, the study underscores the importance of strengthening agricultural extension services, supporting producer organizations, and promoting appropriate technologies to maximize the benefits of cassava mosaic disease management practices for food security and rural development. Full article

(This article belongs to the Special Issue Strategies for Resilient and Sustainable Agri-Food Systems—2nd Edition)

21 pages, 9135 KB

Open AccessArticle

Active Floodplain Sedimentation Dynamics in the Upper Tisza Region, Hungary, After River Regulation

by Róbert Vass, Azin Rooien, Péter Czomba, Dávid Balázs, Beáta Babka and György Szabó

Land 2026, 15(2), 322; https://doi.org/10.3390/land15020322 - 13 Feb 2026

Abstract

River regulation and embankment construction have fundamentally altered the hydrological relationships and sediment accumulation dynamics of floodplains worldwide. This study examines the accumulation conditions in the Upper Tisza (Hungary) floodplain, focusing on the different surface development conditions of oxbow lakes and fossil natural levees following human intervention. During the study, we integrated high resolution LIDAR terrain models with detailed sedimentological analyses (grain size composition, pH, EC, OC, CaCO₃). We used multivariate statistical methods (principal component and cluster analysis) to separate soil formation processes and sediment accumulation. Based on our results, we identified sharp sedimentological boundaries indicating artificial meander cutting (1852). We demonstrated that the cut meanders function as sediment traps, where the accumulation of fine grained sediments is significantly faster (>0.34 cm/year) than on the higher elevation natural levees (0.1 cm/year). Statistical analysis identified five distinct sedimentation environments, successfully separating recent soil levels from river sediments. These results provide an important basis for the complex management of floodplains, such as flood protection, water retention, and habitat management planning. Full article

(This article belongs to the Section Land, Soil and Water)

►▼ Show Figures

Figure 1

14 pages, 633 KB

Open AccessArticle

Association Between Sub-National Regional Socioeconomic Status and Childhood Obesity in Five South-East European Countries: The WHO European Childhood Obesity Surveillance Initiative—COSI (2019)

by Sanja Musić Milanović, Helena Križan, Nika Šlaus, Emanuel Brađašević, Maja Lang Morović, Visnja Djordjic, Enisa Kujundžić, Sergej M. Ostojic, Igor Spiroski and Gregor Starc

Children 2026, 13(2), 267; https://doi.org/10.3390/children13020267 - 13 Feb 2026

Abstract

Background/Objectives: This study focused on the sub-national regional heterogeneity in childhood obesity prevalence across five countries in south-east Europe and the correlation between this heterogeneity and socioeconomic differences. Previous studies have mainly observed national or cross-national data but this study used a sub-national regional approach that may be beneficial in the further investigation of childhood obesity. Methods: Nationally representative samples of children from Croatia, Montenegro, North Macedonia, Serbia and Slovenia were selected using the COSI methodology and used to estimate regional childhood obesity prevalence values. The Sub-national Human Development Database provided data on the Sub-national Human Development Index (SHDI). The spatial autocorrelation analysis of childhood obesity prevalence in sub-national regions was performed and its association with sub-national human development was tested with an ordinary least squares regression model. Results: This study found statistically significant differences in childhood obesity prevalence across sub-national regions in Croatia, Slovenia and Serbia, while no such differences were observed in North Macedonia and Montenegro. There was moderate clustering in childhood obesity rates (Moran’s I = 0.337). The results indicated a significant negative association between SHDI and childhood obesity prevalence across the 48 regions (β = −66.63, p < 0.001). Conclusions: Future public health efforts should take into consideration regional differences in childhood obesity prevalence, and more targeted research is essential for understanding the mechanisms of resilience and vulnerability on a sub-national level. Full article

(This article belongs to the Special Issue Identification of Factors Associated with Obesity in Children: Present and Future Challenges)

18 pages, 1410 KB

Open AccessArticle

Visitor Characteristics and Museum Fatigue: A Case Study at the ETRU Museum in Rome

by Claudio Zavattaro, Emanuele Cirillo, Hilary Serra, Gianluca D’Agostino, Paolo Dabove, Michela Benente, Valeria Minucciani, Anna Berti and Raffaella Ricci

Brain Sci. 2026, 16(2), 225; https://doi.org/10.3390/brainsci16020225 - 13 Feb 2026

Abstract

Background/Objectives: Museum fatigue decreases visitors’ interest due to environmental, social, and personal factors. However, it remains unclear whether physiological parameters can capture museum fatigue, and whether personal factors contribute to psychophysiological changes associated with museum fatigue. Methods: To fill these knowledge gaps, 61 participants visited the ETRU museum in Rome while their position and heart rate (HR) values were continuously recorded. Emotional state was rated after the visit. Time-series analyses assessed trends in viewing time and HR across the full sample and in three clusters defined by personal factors, with correlations examining associations among visit time, HR, and emotional states. Results: Overall, viewing time decreased, while HR increased during the visit. Emotional state correlated positively with visit time, but negatively with HR. The viewing time decrease was consistent across clusters, while HR trends and correlations differed. Conclusions: These findings confirmed that environmental characteristics induce museum fatigue in the visitors and showed that heart rate may be employed as an implicit measure of museum fatigue. In addition, this study revealed that personal factors can modulate the emergence of this phenomenon. Full article

(This article belongs to the Section Behavioral Neuroscience)

21 pages, 3692 KB

Open AccessArticle

Triple-Voltage Gain and Self-Balancing in a New Switched-Capacitor Seven-Level Inverter for Microgrid Integration

by Mohamed Salem, Mahmood Swadi, Anna Richelli, Yevgeniy Muralev and Faisal A. Mohamed

Energies 2026, 19(4), 1001; https://doi.org/10.3390/en19041001 - 13 Feb 2026

Abstract

In the context of power electronic interfaces in photovoltaic (PV), fuel cell, battery, and microgrid applications, the low output voltage of the DC source necessitates a voltage-boosting inverter. This paper proposes a single-source seven-level switched-capacitor boost inverter, particularly for low-voltage applications. The proposed inverter has the capability to produce seven different output voltage levels, i.e., intermediate boosted levels, with a total gain of three times the input voltage. The inverter has the advantage of a reduced number of power switches, diodes, and a switched-capacitor unit, which allows for single-stage operation without the need for a second DC-DC converter. The operating principle of the proposed inverter is explained in detail with a complete switching state analysis, conduction path analysis, and output voltage generation. The capacitor size is calculated using a charge balance-based equation. The self-balancing capability is validated for mismatched initial voltages with a bounded steady-state ripple. To evaluate the performance of the proposed inverter in a more realistic scenario, the effects of non-ideal device characteristics are considered, and the efficiency of the inverter is estimated using a loss model. A predictive current control technique is applied to control the output current under inductive load conditions. The simulation results obtained in MATLAB/Simulink software validate the proper seven-level operation of the inverter, the self-balancing capability of the capacitors, improved output waveform quality, and current control. The proposed inverter can be extended to grid-connected applications, where conventional output filters can be applied to meet the harmonic standards. Full article

(This article belongs to the Special Issue Advances in Power Converters and Inverters)

►▼ Show Figures

Figure 1

19 pages, 1244 KB

Open AccessFeature PaperArticle

Assessing the Uptake of Toxic Elements by Brassica rapa and Associated Health Risks in Soils with Different Natural Background Levels

by Maurizio Ambrosino, Eleonora Di Salvo, Vincenzo Nava, Shashank Sagar Saini, Claudia Genovese, Nicola Cicero, Giuseppe Diego Puglia and Domenico Cicchella

Environments 2026, 13(2), 106; https://doi.org/10.3390/environments13020106 - 13 Feb 2026

Abstract

This research investigates the uptake of potentially toxic elements (PTEs) by Brassica rapa L. grown in volcanic and clay soils with high natural background levels of these elements, and assesses related human health risks. The study was conducted in two Italian regions that produce B. rapa L. for food use (Campania and Sicily). The results of this exploratory research indicate that the naturally elevated concentrations of PTEs in soils lead to correspondingly high levels of these elements in B. rapa L. The investigated soils exhibited marked chemical differences. Volcanic soils had higher Total Organic Carbon (TOC) and PTEs concentrations alongside lower pH and Cation Exchange Capacity (CEC) than clayey soils. In the investigated plants, PTEs accumulated mainly in roots and stems, with notable Hg levels in leaves. While As exceeded safety limits in only one edible sample from volcanic soil, Cd, Hg, and Pb frequently surpassed them. Health risk assessments revealed significant carcinogenic and non-carcinogenic risks from plants grown on volcanic soils, with levels that remain unacceptable even at low consumption rates. In contrast, lower risk levels are associated with the consumption of Brassica rapa grown in clay soils, with values that are generally considered tolerable at low consumption rates. The preliminary findings of this study highlight that natural soil enrichment can cause PTE levels in B. rapa L. that often exceed safe consumption thresholds. These results provide a foundation for future research aimed at more thoroughly investigating the mechanisms of metal uptake by edible plants in areas naturally enriched with PTEs in order to enhance the safety and sustainability of our food. Full article

(This article belongs to the Special Issue New Insights into the Occurrence, Mobility, Transport, and Fate of Contaminants in Surface and Subsurface Environments)

25 pages, 3426 KB

Open AccessArticle

PPAR-Delta Agonist Therapies Did Not Rescue Hallmark Disease Phenotypes in Two Sets of Preclinical Trials in ALS TDP-43 and C9orf72 Model Mice

by David T. Luong, Chenchen Niu, Eunice Kim, Nolan Tanji, Ivy Duong, Brandon Galero, Yong-Jie Zhang, Craig L. Bennett and Albert R. La Spada

Int. J. Mol. Sci. 2026, 27(4), 1820; https://doi.org/10.3390/ijms27041820 - 13 Feb 2026

Abstract

Peroxisome-proliferator–activated receptor delta (PPARδ) regulates metabolic, mitochondrial, and inflammatory pathways implicated in neurodegeneration, making it an attractive therapeutic target for amyotrophic lateral sclerosis (ALS). In this study, we evaluated two PPARδ agonists, KD3010 and T3D-959, in two established ALS/FTD mouse models: an AAV-mediated C9orf72 G4C2-repeat expansion model (C9-149R) and the TDP-43^Q331K transgenic model. Drug treatment was initiated prior to the emergence of key disease features and continued for 9–10 months. Comprehensive behavioral, neuropathological, and biomarker analyses revealed marked differences between the two models. C9-149R mice exhibited reduced body weight and subtle behavioral alterations without robust motor deficits, whereas TDP-43^Q331K mice developed pronounced, progressive motor and cognitive impairments accompanied by a ~7-fold elevation in plasma neurofilament light chain (NfL). Despite effective target engagement—particularly for T3D-959—neither PPARδ agonist improved motor performance, cognitive behavior, neuroanatomical measures, plasma NfL levels, or disease-associated molecular phenotypes in either model. Prolonged KD3010 treatment resulted in loss of target engagement, consistent with drug tolerance, while T3D-959 sustained PPARδ activation without therapeutic benefit. Together, these findings demonstrate that PPARδ agonism is insufficient to modify disease progression in these ALS/FTD mouse models and underscore the importance of publishing well-powered negative preclinical studies to refine therapeutic strategies for ALS. Full article

(This article belongs to the Special Issue Neurodegenerative Disorders: Advances in Mechanisms, Models, and Therapeutics)

37 pages, 6059 KB

Open AccessArticle

A Machine Learning-Based Early Design Energy Prediction Framework for School Buildings Across Multiple Climatic Regions of Türkiye

by Aslihan Senel Solmaz

Buildings 2026, 16(4), 779; https://doi.org/10.3390/buildings16040779 - 13 Feb 2026

Abstract

School buildings are important in terms of energy performance, and their energy demand varies significantly across different climates. Early design decisions strongly influence this demand; however, building energy simulations are computationally intensive and limit rapid evaluation of alternative design options at scale. This study proposes a machine learning-based surrogate modeling framework to support early design energy assessment of school buildings across Türkiye’s six TS 825 climatic regions. A comprehensive design space is defined by varying key parameters, including building shape, orientation, window-to-wall ratio, shading, glazing systems, and insulation alternatives. Representative design configurations are generated using stratified random sampling, and then simulated in EnergyPlus, resulting in a dataset of 30,000 samples. Random Forest, Support Vector Regression, and Multilayer Perceptron models are developed within a multi-output regression framework to predict annual heating and cooling energy demand across climatic regions. The models achieve high predictive accuracy and consistent generalization, with test R² values exceeding 0.93, while exhibiting performance differences among the evaluated algorithms. Feature importance analysis identifies window-to-wall ratio and glazing-related parameters as the most influential early design variables. Overall, the results demonstrate that machine learning-based surrogate models can substantially reduce computational effort while providing reliable, climate-responsive support for early design decision-making. Full article

(This article belongs to the Section Building Energy, Physics, Environment, and Systems)

32 pages, 5150 KB

Open AccessFeature PaperArticle

Model Validation for Multivariate Functional Responses via Autoencoder-Based Dual-Layer Feature Extraction

by Dengyu Wu, Xiaodong Zhang, Daobo Sun, Haidong Lin, Jinhui Li and Baoqiang Zhang

Mathematics 2026, 14(4), 674; https://doi.org/10.3390/math14040674 - 13 Feb 2026

Abstract

Model validation for complex simulation models with multivariate functional responses poses significant challenges, as it involves the dual coupling of physical correlations among variables and field correlations in time-series data. A novel Autoencoder-based Dual-Layer Feature Extraction (AE-DLFE) method is proposed. The first layer uses joint principal component analysis to decouple physical correlations, while the second layer develops an Autoencoder-improved Feature Selective Validation (AE-FSV) method that adaptively extracts features of time-series data and measures feature discrepancies via deep representation learning. On this basis, a new validation metric named U-PCDM (Uncertainty Principal Component Difference Measure) is developed to quantify the discrepancies between simulation and experiment under uncertainty. Theoretical analysis confirms the boundedness and unique temporal permutation sensitivity of the proposed metric. Case study results demonstrate that the proposed AE-FSV enhances the evaluation accuracy of traditional FSV on transient data. Furthermore, compared to benchmark methods such as MD-pooling, the U-PCDM metric significantly improves computational efficiency—especially in high-dimensional scenarios—while maintaining consistent model rankings. This work effectively addresses the heterogeneous correlation coupling issue, offering a robust quantitative tool for model validation. Full article

(This article belongs to the Special Issue Advanced Intelligent Algorithms for Decision Making Under Uncertainty)

17 pages, 944 KB

Open AccessArticle

Quantifying the Spread and Economic Consequences of the Codling Moth (Cydia pomonella) in China Using Biomod2 and Monte Carlo Synergy

by Shengkang Zou, Zhongxiang Sun, Hongkun Huang, Xiaoqing Xian and Guifen Zhang

Agriculture 2026, 16(4), 439; https://doi.org/10.3390/agriculture16040439 - 13 Feb 2026

Abstract

The codling moth, Cydia pomonella (Linnaeus, 1758) (Lepidoptera: Tortricidae), was first detected in Xinjiang, China, in 1953 and has since spread to nine provinces, with its distribution continuing to expand into other apple- and pear-producing regions. In this study, we combined the Biomod2 model with Monte Carlo simulations to perform a spatially explicit, pixel-level assessment of the moth’s potential habitat suitability and associated economic impacts in China’s major fruit-producing areas. Results indicate that temperature is the primary factor limiting its distribution, followed by human activities, while topography plays a regulatory role at local scales. The Loess Plateau and Bohai Rim regions were identified as core suitable areas, with moderate suitability in the Northern Cold region and Xinjiang and lower suitability in the Southwest and Yangtze River Basin. Pearson correlation analysis revealed weak spatial coupling between suitable habitats and fruit yields. Monte Carlo simulations showed that potential economic losses vary spatially across regions and crop types. These findings suggest that the codling moth’s suitability differs among regions; high-yield areas do not necessarily face higher invasion risk, but once an invasion occurs, economic losses tend to be concentrated and severe. Accordingly, early warning and region-specific, differentiated management should be prioritized in key areas to mitigate damage. Full article

(This article belongs to the Section Crop Protection, Diseases, Pests and Weeds)

28 pages, 4945 KB

Open AccessArticle

Research on the Coupling Coordination Between Economic Resilience and Ecological Resilience in China’s Coastal Cities from the Perspective of Evolutionary Ecological Economics

by Chongyang Wu, Mingjing Wu, Pengzhou Yan and Dongjian Ci

Sustainability 2026, 18(4), 1963; https://doi.org/10.3390/su18041963 - 13 Feb 2026

Abstract

The conflict between the economy and the ecological environment is prominent in China’s coastal cities, and these cities contend with heightened uncertainty. Therefore, this study uses the econometric model to analyze the spatial–temporal pattern characteristics and affecting factors of the coupling coordination level between urban economic resilience (ER) and urban ecological resilience (EcR) in China’s coastal cities based on improvement of the evaluation index system, thus advancing policy suggestions. The main conclusions are as follows: (1) The coupling coordination degree (CCD) between ER and EcR across different types of coastal cities strongly correlates with their spatial distribution patterns of economic development. From the East China Sea to the South China Sea and Yellow and Bohai Sea Coast cities and from central cities to industrial cities, other types of cities, and resource-based cities, CCD exhibits an overall declining trajectory. (2) The gap in CCD in China’s coastal cities generally shows an expanding trend. (3) The spatial distribution pattern of the centrality of CCD in China’s coastal cities has a relatively high consistency. Urban spillover roles are highly consistent with levels of economic development. (4) The number and diversity of dominant influencing factors have steadily increased. Full article

(This article belongs to the Section Social Ecology and Sustainability)

►▼ Show Figures

Figure 1

22 pages, 46711 KB

Open AccessArticle

CMNet: Global–Local Feature Fusion CNN-Mamba Network for Remote Sensing Object Detection

by Jin Liu, Liangliang Li, Xiaobin Zhao, Ming Lv, Zhenhong Jia, Xueyu Zhang, Gemine Vivone and Hongbing Ma

Remote Sens. 2026, 18(4), 591; https://doi.org/10.3390/rs18040591 - 13 Feb 2026

Abstract

In the field of remote sensing object detection (RSOD), significant challenges remain, including the vast field of view in remote sensing images, the diverse array of target categories, and complex backgrounds. Traditional methods for processing remote sensing images face limitations in this context. While convolutional neural networks (CNNs) can expand the receptive field by utilizing kernels of different sizes, larger kernels increase the number of parameters and introduce noise. Vision Transformers (ViT) achieve global receptive fields through their global attention mechanism. However, their quadratic computational complexity struggles with high-resolution images. Recently, Mamba has gained prominence in image processing. Its unique four-directional scanning mechanism allows focusing on regions of interest from multiple angles while maintaining linear model complexity and achieving global receptive fields. In this work, we propose a new CNN–Mamba network (CMNet) that synergistically exploits the advantages of both architectures. Specifically, we employ VMamba(VM) to extract global semantic features from images. Moreover, we design a multi-scale local feature extraction (MLFE) module, which captures local texture information and edge details through the local feature extraction (LFE) and the global attention module (GAM). The synergy between VMamba and MLFE creates complementary global–local features. To address the representational differences between these two kinds of features, we further design a feature cross-complementary (FCC) module. This module achieves cross-complementarity of features, solving feature disparity issues. Our CMNet achieves 79.38% mAP50 on the DOTA v1.0 dataset and 90.60% mAP50 on the HRSC dataset, outperforming existing state-of-the-art approaches. Full article

(This article belongs to the Special Issue Advances in Detection-Oriented Multi-Sensor Fusion Beyond the Visible Spectrum)

16 pages, 5178 KB

Open AccessArticle

Long-Term Associations of Early-Life Human Milk Oligosaccharide Intake with Allergic Disease Development and Gut Microbiota Profiles in 5-Year-Old Children

by Ruixin Kou, Che Pan, Xiaolong Xing, Jin Wang, Sinéad T. Morrin, Rachael H. Buck, Xiang Li, Yingyi Mao and Shuo Wang

Nutrients 2026, 18(4), 624; https://doi.org/10.3390/nu18040624 - 13 Feb 2026

Abstract

Background: Based on our extensive cohort study, the Maternal Nutrition and Infant Investigation (MUAI), this research investigated the associations between human milk oligosaccharide (HMO) intake during the postnatal period and allergic disease development and gut microbiome composition in early childhood through long-term follow-up. Methods: Human breast milk (HBM) samples at five lactation stages and fecal samples of infants and young children were collected. Children aged 5 years included in this study were categorized into allergic and non-allergic groups via standardized allergen testing. Results: The findings indicated that higher HMO intake levels across five distinct lactation periods may be linked to a reduced incidence of allergies in children. The consumption of six major structurally representative HMOs was significantly associated with alterations in the gut microbiota profiles of young children. Moreover, there were notable differences in gut microbiota composition between allergic and non-allergic children. Specifically, beneficial bacteria such as Bifidobacterium, Akkermansia, and Ruminococcus were significantly enriched, in addition to the levels of metabolite propionic acid, a beneficial short-chain fatty acid, which were notably higher in the non-allergic group. To further validate the relationship between Bifidobacterium abundance and early HMO intake, the analysis revealed that a differential strain biomarker, Bifidobacterium adolescentis (B. adolescentis), exhibited significant correlations with specific HMOs at different lactation stages, particularly showing a strong positive correlation with 2′-fucosyllactose (2′-FL) content. Conclusions: These findings suggest that early-life HMO intake is associated with long-term differences in allergic outcomes, potentially through modulation of gut microbiota composition, particularly the enrichment of B. adolescentis. Full article

(This article belongs to the Section Pediatric Nutrition)

►▼ Show Figures

Figure 1

28 pages, 5322 KB

Open AccessArticle

Facial Expression Annotation and Analytics for Dysarthria Severity Classification

by Shufei Duan, Yuxin Guo, Longhao Fu, Fujiang Li, Xinran Dong, Huizhi Liang and Wei Zhang

Sensors 2026, 26(4), 1239; https://doi.org/10.3390/s26041239 - 13 Feb 2026

Abstract

Dysarthria in patients post-stroke is often accompanied by central facial paralysis, which impairs facial motor control and emotional expression. Current assessments rely on acoustic modalities, overlooking facial pathological cues and their correlation with emotional expression, which hinders comprehensive disease assessment. To address this issue, we propose a multimodal severity classification framework that integrates facial and acoustic features. Firstly, a multi-level annotation algorithm based on a pre-trained model and motion amplitude was designed to overcome the problem of data scarcity. Secondly, facial topology was modeled using Delaunay triangulation, with spatial relationships captured via graph convolutional networks (GCNs), while abnormal muscle coordination is quantified using facial action units (AUs). Finally, we proposed a multimodal feature set fusion technology framework to achieve the compensation of facial visual features for acoustic modalities and the analysis of disease classification. Our experimental results using the THE-POSSD dataset demonstrate an accuracy of 92.0% and an F1 score of 91.6%, significantly outperforming single-modality baselines. This study reveals the changes in facial movements and sensitive areas of patients under different emotional states, verifies the compensatory ability of visual patterns for auditory patterns, and demonstrates the potential of this multimodal framework for objective assessment and future clinical applications in speech disorders. Full article

(This article belongs to the Section Sensing and Imaging)

Show export options Show export options

Select all

Export citation of selected articles as:

Error

Oops... you haven't selected anything for export.

Displaying article 1-50 on page 1 of 2000.

Go to page 1 2 3 4 5

Search Results (579,521)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI