Search Results (270)

Accurate upper-limb motor assessment is critical for post-stroke rehabilitation but relies on subjective clinical scales. This study proposes the Action-Aware Multimodal Wavelet Fusion Network (AMWFNet), integrating surface electromyography (sEMG) and robotic kinematics for automated Fugl-Meyer Assessment (FMA-UE)-aligned quantification. Continuous Wavelet Transform (CWT) converts heterogeneous signals into unified time-frequency scalograms. A learnable modality gating mechanism dynamically weights physiological and kinematic features, while action embeddings encode task contexts across 18 standardized reaching tasks. Validated on 40 participants (20 post-stroke, 20 healthy), AMWFNet achieved 94.68% accuracy in six-class classification, outperforming baselines by 9.17% (Random Forest: 85.51%, SVM: 85.30%, 1D-CNN: 91.21%). The lightweight architecture (1.27M parameters, 922ms inference) enables real-time assessment-training integration in rehabilitation robots, providing an objective, efficient solution. Full article

Chronic obstructive pulmonary disease (COPD) is a major health burden worldwide and in Taiwan, ranking as the third leading cause of death globally, and its prevalence in Taiwan continues to rise. Readmission within 14 days is a key indicator of disease instability and care efficiency, driven jointly by patient-level physiological vulnerability (such as reduced lung function and multiple comorbidities) and healthcare system-level deficiencies in transitional care. To mitigate the growing burden and improve quality of care, it is urgently necessary to develop an AI-based prediction model for 14-day readmission. Such a model could enable early identification of high-risk patients and trigger multidisciplinary interventions, such as pulmonary rehabilitation and remote monitoring, to effectively reduce avoidable early readmissions. However, medical data are commonly characterized by severe class imbalance, which limits the ability of conventional machine learning methods to identify minority-class cases. In this study, we used real-world clinical data from multiple hospitals in Kaohsiung City to construct a prediction framework that integrates data generation and ensemble learning to forecast readmission risk among patients with chronic obstructive pulmonary disease (COPD). CTGAN and kernel density estimation (KDE) were employed to augment the minority class, and the impact of these two generation approaches on model performance was compared across different augmentation ratios. We adopted a stacking architecture composed of six base models as the core framework and conducted systematic comparisons against the baseline models XGBoost, AdaBoost, Random Forest, and LightGBM across multiple recall thresholds, different feature configurations, and alternative data generation strategies. Overall, the results show that, under high-recall targets, KDE combined with stacking achieves the most stable and superior overall performance relative to the baseline models. We further performed ablation experiments by sequentially removing each base model to evaluate and analyze its contribution. The results indicate that removing KNN yields the greatest negative impact on the stacking classifier, particularly under high-recall settings where the declines in precision and F1-score are most pronounced, suggesting that KNN is most sensitive to the distributional changes introduced by KDE-generated data. This configuration simultaneously improves precision, F1-score, and specificity, and is therefore adopted as the final recommended model setting in this study. Full article

Sarcopenia is highly prevalent among patients with chronic kidney disease (CKD) and chronic heart failure (CHF), yet the underlying immunometabolic mechanisms remain insufficiently understood. Short-chain fatty acids (SCFAs), inflammatory cytokines, and body-composition alterations may jointly contribute to the development of muscle dysfunction in this population. In this cross-sectional study, 80 patients with CKD and CHF underwent comprehensive clinical, biochemical, bioimpedance, inflammatory, and SCFA profiling. Sarcopenia was diagnosed according to EWGSOP2 criteria. Multivariable logistic regression, LASSO feature selection, correlation analysis, PCA, and Random Forest modeling were used to identify key determinants of sarcopenia. Sarcopenia was present in 39 (49%) participants. Patients with sarcopenia exhibited significantly lower body fat percentage, reduced ASM, and slower gait speed. Hexanoic acid (C6) showed an independent positive association with sarcopenia (OR = 2.24, 95% CI: 1.08–5.37), while IL-8 showed an inverse association with sarcopenia (OR = 0.38, 95% CI: 0.13–0.94), indicating that lower IL-8 levels were more frequently observed in individuals with sarcopenia. Correlation heatmaps revealed distinct SCFA–cytokine coupling patterns depending on sarcopenia status, with stronger pro-inflammatory clustering in C6-associated networks. The final multivariable model integrating SCFAs, cytokines, and body-composition metrics achieved excellent discrimination (AUC = 0.911) and good calibration. Sarcopenia in CKD–CHF patients represents a systemic immunometabolic disorder characterized by altered body composition, chronic inflammation, and dysregulated SCFA signaling. Hexanoic acid (C6) and IL-8 may serve as informative biomarkers of muscle decline. These findings support the use of multidimensional assessment and highlight potential targets for personalized nutritional, microbiota-modulating, and rehabilitative interventions. Full article

Unmitigated climate change, coupled with habitat loss, has made the grassland and shrubland bird communities particularly vulnerable to extinction. Climate change-induced drought reduces net primary productivity, food availability, habitat quality, and alters vegetation structure. These factors collectively increase mortality in grassland and shrubland birds. However, limited data on habitat use by tropical birds hampers the development of effective management plans for drought-affected landscapes. We examined the foraging sites of 18 shrubland bird species, including two endemic and four declining species, across three shrubland forest sites in the Eastern Ghats of India during the dry season. We recorded microhabitat features within an 11 m radius of observed foraging points and compared them with random plots. Additionally, we examined the association between bird species and plant species where a bird was observed foraging. Foraging sites differed significantly from random plots, indicating active selection of microhabitats by shrubland birds. Using linear discriminant analysis, we found that the microhabitat features important for the bird species were presence of ground cover, shrub density, vegetational height, and vertical foliage stratification. Our results show that diet guild and foraging strata influence the foraging microhabitat selection of a species. Microhabitat attributes selected by shrubland specialist species differed from those of generalist shrubland users. Thirteen out of 18 focal species showed a significant association with at least one plant species. Birds were often associated with plants that were green during the dry season. Based on habitat selection and plant associations, we identified several habitat attributes that can be actively managed. Despite being classified as wastelands, the heavily degraded shrub forests can be rehabilitated through strategic and selective harvesting of forest products, targeting invasive species, and a spatially and temporally controlled livestock grazing regime. Full article

Despite the growing interest in physiological stress monitoring, an objective measure of stress is currently lacking, especially in clinical and rehabilitation contexts. With the emerging integration of artificial intelligence (AI) in data analytics, heart rate variability (HRV) has gained attention as an effective biomarker; however, the literature remains fragmented across disciplines, stress types, and methodological approaches. This scoping review aims to investigate how AI techniques are applied to HRV analysis for stress detection and prediction in adult populations. Although this review does not focus on a specific subtype of stress, its primary objective is to explore the current methodological state of the art as reported in the literature, without restrictions on stress typology. Following PRISMA-ScR guidelines, a systematic search was conducted across PubMed, Scopus, and Google Scholar for studies published between 2005 and 2025, using MeSH terms including “HRV”, “Rehabilitation”, “SCI” (for Spinal Cord Injury), “Stress”, “Sympathetic”, “Parasympathetic”, “Non-linear”, “Gamification”, “AI” and “Machine Learning”. Inclusion criteria targeted adult human populations and studies employing HRV features as input for AI and machine learning techniques for psychophysical stress assessment. Of the 566 records identified, 15 studies met the eligibility criteria. The reviewed studies exhibit substantial heterogeneity in terms of settings, populations, sensors, and algorithms with most employing supervised methods (e.g., random forest, support vector machine), alongside several applications of deep learning and explainable AI. Only one study focused specifically on physiological stress, none focused on SCI populations, and rehabilitation-related research was scarce, thereby underscoring important gaps in the current literature. Overall, HR variability analysis, especially when combined with artificial intelligence techniques, represents a promising approach for stress assessment; however, the field is methodologically fragmented and clinically underdeveloped in critical areas, underscoring the need for a multidisciplinary methodological framework. Full article

Across the world, asthma is a prominent and widespread respiratory disorder that has a substantial clinical and socioeconomic influence. The classification of asthma subtypes should be performed precisely and effectively, with objectives such as personalized treatments, improved rehabilitation outcomes, and preventing tragic exacerbations. Typical screening approaches are primarily based on spirometry measures, immunologic assessments, and individual clinical diagnoses, and they are commonly affected by limitations such as uncertainty, crossover disparities, and restricted generalizability among various groups of patients. This study utilizes machine learning (ML) methodologies as a Data-Driven Approach (DDA)-based framework for asthma classification to overcome the mentioned challenges. Methodically constructed and evaluated classifiers, such as Random Forest and XGBoost, use the Asthma Disease Dataset from Kaggle, which consists of demographic data, lung function metrics (FEV1, FVC, FEV1/FVC ratio, and PEFR), and immunoglobulin E (IgE) biomarkers. A wide range of metrics such as accuracy, precision, recall, F1-score, receiver operating characteristic area under the curve (ROC-AUC), and average precision (AP) are used exhaustively to assess the performance of the model. The results indicate that though each model exhibits outstanding forecasting abilities, XGBoost has an enhanced classification capability, especially in recall and AP, which minimizes the proportion of false negatives, resulting in a clinically noteworthy result. The significance of the FEV1/FVC ratio, IgE levels, and PEFR as key indicators is recognized by feature interpretability analysis. These results emphasize the ability of ML-powered evaluation in advancing personalized healthcare and revolutionizing the clinical management of asthma. Full article

With the rapid expansion and aging of global road networks, cold recycled mix asphalt (CRMA) has gained significant attention as a sustainable pavement rehabilitation technology. However, the road performance of CRMA is highly sensitive to material composition and curing conditions, making accurate performance prediction challenging. This study develops machine learning (ML) models to predict two critical performance indicators: dynamic stability (DS) for high-temperature stability and indirect tensile strength (ITS) for low-temperature crack resistance. Four ML algorithms, Artificial Neural Network (ANN), Extreme Gradient Boosting (XGBoost), Random Forest (RF), and Support Vector Regression (SVR), were trained on a comprehensive dataset of 436 samples. A genetic algorithm (GA) was employed to optimize model hyperparameters, significantly enhancing prediction accuracy and robustness. The SHAP method was further applied to interpret model outputs and identify key influencing factors. Results demonstrate that GA-optimized XGBoost achieved the highest predictive performance for both dynamic stability (DS) and indirect tensile strength (ITS), with R² values of 0.9793 and 0.9694, respectively. Curing temperature, RAP content, and curing time were identified as the most influential factors. This study provides an accurate and interpretable ML-based framework for CRMA performance prediction, facilitating optimized mix design for pavement construction and maintenance. Full article

For the development of a green future, managerial decision making at the local level plays an important role. The study is dedicated to the analysis of the current state of green areas, and development and rehabilitation of green areas in the territorial communities of Ukraine. The goal of the study is the development of a set of measures to create a sustainable green infrastructure at the local level in Ukraine. The main trends of green land policies by territorial communities were substantiated: keeping the natural afforestation of agricultural land; the development and rehabilitation of water conservation zones, windbreak belts, anti-erosion forests, green belts of inhabited areas, and nature conservation or recreation areas; and promoting gardening. A land reallotment methodology, which allows for the expansion of a spatial environment for the development and rehabilitation of green areas was suggested. The methods and approaches presented were tested in the Petrivska Territorial Community of Kyiv Region. The presented measures allow for an increase the green area of a territorial community by 1,084,352 m². The approach allows for the minimization of the condemnation of land from landowners, creates a more comfortable environment for the population, facilitates the effectiveness of agriculture due to containing the erosion, and conservation of natural landscapes. The research findings approved that the main challenges for the implementation of green policies are the acquisition of land for green areas in the environment of the historically established land use, and controlling the sustainable use of green areas and their surroundings responsibly to prevent their violation. Full article

Brain–computer interfaces (BCIs) have successfully been used for stroke rehabilitation by pairing movement intentions with, e.g., functional electrical stimulation. It has also been proposed that BCI training is beneficial for people with cerebral palsy (CP). To develop BCI training for CP patients, movement intentions must be detected from single-trial EEG. The study aim was to detect movement intentions in CP patients and able-bodied participants using different classification scenarios to show the technical feasibility of BCI training in CP patients. Five CP patients and fifteen able-bodied participants performed wrist extensions and ankle dorsiflexions while EEG was recorded. All but one participant repeated the experiment on 1–2 additional days. The EEG was divided into movement intention and idle epochs that were classified with a random forest classifier using temporal, spectral, and template matching features to estimate movement intention detection performance. When calibrating the classifier on data from the same day and participant, 75% and 85% classification accuracies were obtained for CP- and able-bodied participants, respectively. The performance dropped by 5–15 percentage points when training the classifier on data from other days and other participants. In conclusion, movement intentions can be detected from single-trial EEG, indicating the technical feasibility of using BCIs for motor training in people with CP. Full article

Post-harvest forest residue management and liming practices can significantly affect soil quality. This study evaluated the impacts of burnt pine harvest residues and lime application methods (surface-applied vs. incorporated) on the chemical and physical properties of a Dystric Cambisol in Southern Brazil. Soil samples were collected at two depths (0–10 cm and 10–20 cm) and analyzed for pH, exchangeable acidity, organic carbon, cation exchange capacity, macroporosity, microporosity, and bulk density. The results showed that changes were more pronounced in the 0–10 cm layer and mainly affected chemical attributes. Incorporated lime increased pH from 4.7 to 5.1, increased base saturation from 17% to 36%, and reduced Al saturation from 45% to 13% in the 0–10 cm layer. Burnt residues alone did not significantly alter soil properties, whereas lime incorporation led to improved chemical conditions and enhanced soil structure, especially in the surface layer. The treatments that maintained pine residues on the surface favored biological processes in the topsoil, while the burning of these residues had variable impacts on soil structure and nutrient availability. These findings highlight the importance of incorporating lime to optimize soil rehabilitation following pine harvesting in subtropical forest systems. Full article

The Timed Up and Go (TUG) test is a widely used clinical tool for assessing mobility and fall risk in older adults and individuals with neurological or musculoskeletal conditions. While it provides a quick measure of functional independence, traditional stopwatch-based timing offers only a single completion time and fails to reveal which movement phases contribute to impairment. This study presents a smartphone-based system that automatically segments the TUG test into distinct phases, delivering objective and low-cost biomarkers of lower-limb performance. This approach enables clinicians to identify phase-specific impairments in populations such as individuals with Parkinson’s disease, and older adults, supporting precise diagnosis, personalized rehabilitation, and continuous monitoring of mobility decline and neuroplastic recovery. Our method combines adaptive preprocessing of accelerometer and gyroscope signals with supervised learning models (Random Forest, Support Vector Machine (SVM), and XGBoost) using statistical features to achieve continuous phase detection and maintain robustness against slow or irregular gait, accommodating individual variability. A threshold-based turn detection strategy captures both sharp and gradual rotations. Validation against video ground truth using group K-fold cross-validation demonstrated strong and consistent performance: start and end points were detected in 100% of trials. The mean absolute error for total time was 0.42 s (95% CI: 0.36–0.48 s). The average error across phases (stand, walk, turn) was less than 0.35 s, and macro F₁ scores exceeded 0.85 for all models, with the SVM achieving the highest score of 0.882. Combining accelerometer and gyroscope features improved macro F1 by up to 12%. Statistical tests (McNemar, Bowker) confirmed significant differences between models, and calibration metrics indicated reliable probabilistic outputs (ROC-AUC > 0.96, Brier score < 0.08). These findings show that a single smartphone can deliver accurate, interpretable, and phase-aware TUG analysis without complex multi-sensor setups, enabling practical and scalable mobility assessment for clinical use. Full article

Accurate, timely maps of post-fire burn severity are vital for rehabilitation, hydrologic hazard assessment, and ecosystem recovery in the western United States, where large, frequent wildfires and steep environmental gradients challenge model generalization. Machine learning models, particularly tree-based regressors, are increasingly used to relate satellite-derived spectral features to ground-based severity metrics such as the Composite Burn Index (CBI). However, model generalization across spatial domains, both within and between wildfires, remains poorly characterized. In this study, we benchmarked six tree-based regression models (Decision Tree-DT, Random Forest-RF, Extra Trees-ET, Bagging, Gradient Boosting-GB, and AdaBoost-AB) for predicting wildfire severity from Landsat surface reflectance data across ten U.S. fire events. Two spatial validation strategies were applied: (i) within-fire spatial generalization via Leave-One-Cluster-Out (LOCO) and (ii) cross-fire transfer via Leave-One-Fire-Out (LOFO). Performance is assessed with R², RMSE, and MAE under identical predictors and default hyperparameters. Results indicate that, under LOCO, variance-reduction ensembles lead: RF attains R² = 0.679, MAE = 0.397, RMSE = 0.516, with ET statistically comparable (R² = 0.673, MAE = 0.393, RMSE = 0.518), and Bagging close behind (R² = 0.668, MAE = 0.402, RMSE = 0.525). Under LOFO, ET transfers best (R² = 0.616, MAE = 0.450, RMSE = 0.571), followed by GB (R² = 0.564, MAE = 0.479, RMSE = 0.606) and RF (R² = 0.543, MAE = 0.490, RMSE = 0.621). These results indicate that tree ensembles, especially ET and RF, are competitive under minimal tuning for rapid severity mapping; in practice, RF is a strong choice for an individual fire with local calibration, whereas ET is preferred when model transferability to unseen fires is paramount. Full article

Introduction: Cardiovascular disease (CVD) remains a primary cause of mortality worldwide, with persistent geographic disparities driven by a complex interplay of risk factors. Continual updates of localized variations in CVD mortality are essential to develop targeted interventions for optimizing disease and healthcare management. Methods: This study investigated associations between CVD mortality and a comprehensive set of biological, environmental, behavioral, and socioeconomic factors across all U.S. counties, employing correlation, geospatial visualization, stepwise multiple regression, and machine learning models to evaluate the importance of risk associations. Results: Significant disparities in CVD mortality trend were observed across race, age, sex, and region, with elevated rates among older adults, men, and Blacks, particularly in southeastern states exhibiting severe social vulnerability. Correlation analysis identified disease management (e.g., COPD, hypertension, medication non-adherence), environmental factors (PM2.5), lifestyle behaviors (e.g., smoking, sleep duration), and socioeconomic status (e.g., poverty, single-parent households, education) as important contributors to CVD mortality. Conversely, higher household income, physical activity, and cardiac rehabilitation participation were strong protectors. Multiple regression explained 66.9% variance in CVD mortality, recognizing PM2.5, smoking, and medication non-adherence as top associated factors. Random Forest models underscored COPD’s predictive dominance, followed by medication non-adherence, smoking, and sleep duration. Conclusions: The findings highlight the geospatial connection of risk factors to CVD mortality disparities across U.S. counties. They emphasize the critical importance of data-driven strategies targeting air quality, tobacco control, social inequities, and chronic disease management to mitigate CVD burden and promote health equity. Full article

The Asia-Pacific Network for Sustainable Forest Management and Rehabilitation (APFNet) Fellowship Program, established in 2008, aims to develop future leaders in sustainable forest management (SFM) within the Asia-Pacific region. This study represents the first systematic evaluation of the APFNet Fellowship Program, focusing on its effectiveness in building capacity for forest conservation and management. A mixed-methods approach was employed, combining quantitative pre- and post-program surveys with qualitative interviews and case studies of fellows. Quantitative analysis of survey data from 57 fellows revealed significant improvements in knowledge and skills related to forest conservation and sustainable development. Paired-sample t-tests showed statistically significant increases in the knowledge and abilities of participants, with an average improvement of 23% across key survey domains (t = 5.24, p < 0.05). The analysis also indicated that 95% of participants perceived the program to be highly relevant to their career goals and sustainable development objectives, with 87% reporting strong satisfaction with the quality of learning opportunities. Qualitative data from semi-structured interviews and focus groups revealed that while fellows appreciated the academic rigor and practical training, challenges such as financial limitations, language barriers, and institutional support issues were prevalent. Despite these challenges, fellows expressed strong commitment to applying their training to sustainable forestry practices in their home countries. This study highlights the critical role of the APFNet program in nurturing the next generation of forestry leaders in the Asia-Pacific region. The findings provide a foundation for future evaluations, highlighting the need for continued program refinement to address the identified challenges and maximize the long-term impact on forest conservation. Full article

Monitoring muscle fatigue is essential for ensuring safety and maximizing the effectiveness of resistance training. Conventional methods such as electromyography (EMG), inertial measurement units (IMU), and ratings of perceived exertion (RPE) involve complex procedures and have limited applicability, particularly in unsupervised or robotic exercise environments. This study proposes a machine learning-based approach to directly predict RPE from force–time data collected during repeated isokinetic bench press sets. Thirty-two male participants (64 limb datasets) performed seven sets at a standardized 7RM load, with load cell data and RPE scores recorded. Biomechanical features representing magnitude, variability, energy, and temporal dynamics were extracted, along with engineered features reflecting relative changes and inter-set variations. The findings indicate that RPE is more closely related to relative fatigue progression than to absolute biomechanical output. Incorporating engineered features substantially improved predictive performance, with the Random Forest model achieving the highest accuracy and more than 93% of predictions falling within ±1 RPE unit of the reported values. The proposed approach can be seamlessly integrated into intelligent resistance machines, enabling automated load adjustment and providing substantial potential for applications in both athletic training and rehabilitation contexts. Full article