Search Results (774)

Healthcare-associated infections remain an ongoing concern across medical and dental practice, prompting continuous evaluation of infection prevention measures. In dental extractions, the necessity of sterile gloves is debated, as the oral cavity represents an inherently contaminated environment. This systematic review and meta-analysis evaluated whether the use of sterile gloves reduces postoperative socket infections compared with non-sterile gloves. A search of MEDLINE, Embase, Web of Science, Cochrane CENTRAL, and Google Scholar identified randomized controlled trials, clinically controlled trials, and observational trials directly comparing sterile versus non-sterile glove use during dental extractions. The primary outcome of this study was extraction socket infection at day 7 post-surgery. A meta-analysis using relative risk (RR) was performed for dichotomous data. Of the initial 7170 publications found, seven articles met inclusion criteria. Infection rates ranged from 0% to 3.9%, with an overall infection rate of 0.3% in the sterile glove group (672 patients) and 1.3% in the non-sterile glove group (758 patients). Three studies qualified for meta-analysis, resulting in an RR of 0.30 (95% CI 0.07–1.24), indicating no significant difference in postoperative infections between sterile and non-sterile glove usage. Given the limitations of small sample sizes, low event rates, incomplete reporting, and lack of subgroup data for surgical versus non-surgical extractions, no difference in postoperative infection was found between sterile and non-sterile glove use. Additional research is needed to determine whether glove sterility influences infection risk, particularly in surgical procedures. Full article

Pneumatic soft robotic devices are emerging as promising tools for assisting hand rehabilitation in individuals with post-stroke motor impairment. However, evidence regarding their immediate functional impact remains limited, particularly across different impairment levels. This study presents a pilot validation of the YAD_V2 pneumatic finger rehabilitation robot and evaluates acute changes in finger range of motion (ROM) and task performance during a single intervention session. Twenty stroke participants were categorized into two groups based on the Fugl-Mayer Hand sub score: severe impairment (FMA-Hand < 10) and mild-to-moderate impairment (FMA-Hand ≥ 10). ROM was measured using integrated bending sensors during voluntary flexion–extension before, during, and after a 10-min pneumatic actuation session. A mixed 2 × 3 repeated-measure ANOVA revealed a significant Group × Time interaction (F(2, 36) = 4.628, p = 0.016, partial η² = 0.205). In the severe group, ROM increased from 8.53° to 28.46° during actuation (p = 0.002), and partially returned to baseline afterward. In the mild–moderate group, no significant ROM changes were observed; however, cube-transfer time improved significantly (mean improvement: 0.88 s, p = 0.039). These findings indicate that pneumatic assistance induces distinct acute effects depending on impairment severity. This study provides preliminary evidence supporting the feasibility of the YAD_V2 robotic system and highlights the need for multi-session clinical trials to determine therapeutic efficacy. Full article

Osteoarthritis (OA) is a progressive musculoskeletal disorder that affects not only older adults but also younger populations, often leading to chronic pain, joint stiffness, functional impairment, and a decline in quality of life. Non-invasive physical rehabilitation plays a critical role in slowing disease progression, alleviating symptoms, and maintaining joint mobility. However, rehabilitation tools such as compression gloves and manual exercise aids are typically passive and provide minimal real-time feedback to patients or clinicians. Others, such as exoskeletons and soft-actuated devices, can be costly or complex to use. This study presents the design and development of an electrically actuated glove integrated with force and flex sensors, intended to assist individuals diagnosed with Stage 2 OA in performing guided finger exercises. The system integrates a digital front-end application that offers real-time feedback and data visualization, enabling more personalized and trackable therapy sessions for both patients and healthcare providers. Preliminary results from an initial human trial with healthy participants demonstrate that the glove enables naturalistic movement without imposing excessive restriction or augmentation of motion. These findings support the glove’s potential in preserving hand coordination and dexterity, key objectives in early-stage OA intervention, and suggest its suitability for integration into home-based or clinical rehabilitation protocols. Full article

Millions of healthcare workers experience percutaneous exposure to bloodborne communicable infectious disease pathogens annually, with the risk of contracting occupationally acquired infections. In this study, we aimed to assess the status of occupational safety and outbreak preparedness in Congolese nurses and laboratory technicians in Kongo central and the Katanga area, amidst multiple ongoing public health emergencies in the Democratic Republic of the Congo (DRC). This was a multicenter analytical cross-sectional study conducted in five referral hospitals located in Kongo central province and the Katanga area between 2019 and 2020 amidst Ebola, Yellow fever, Cholera and Chikungunya outbreaks. Participants were adult A0 grade nurses, A1 nurses, A2 nurses and medical laboratory technicians (N = 493). They answered a structured, self-administered questionnaire related to hospital hygiene and standard precautions for occupational infection prevention. The majority of the respondents were females (53.6%), and 30.1% of them have never participated in a training session on hospital infection prevention during their career. The proportions of those who have been immunized against hepatitis B virus (HBV) was markedly low, at 16.5%. Of the respondents, 75.3% have been using safety-engineered medical devices (SEDs), whereas 93.5% consistently disinfected medical devices after use. Moreover, 78% of the respondents used gloves during medical procedures and 92.2% wore masks consistently. A large majority of the respondents, 82.9%, have been recapping the needles after use. Regarding participation in outbreak response, 24.5% and 12.2% of the respondents were Chikungunya and Cholera epidemic responders, respectively; 1.8% have served in Ebola outbreak sites. The proportion of the respondents who sustained at least one percutaneous injury by needlestick or sharp device, blood/body fluid splash or both in the previous 12-month period was high, 89.3% (41.8% for injury, 59.2% for BBF event), and most of them (73%) reported over 11 events. Compared to laboratory technicians, nurses had higher odds for sustaining percutaneous injury and BBF events [OR = 1.38 (0.16); p < 0.01], whereas respondents with longer working experience were less likely to sustain those events [OR = 0.47 (0.11); p < 0.001]. Findings from this study suggest that Congolese nurses and laboratory technicians experience a high frequency of injury and BBF events at work, and remain at high risk for occupationally acquired infection. There is a need for periodic capacity-building training for the healthcare workforce to improve infection prevention in health settings, the provision of sufficient and appropriate PPE and SEDs, post-exposure follow-up and keeping records of occupational injuries in hospitals in Congolese healthcare settings. Full article

Soft pneumatic glove actuators for hand rehabilitation require compact, accurate models that can be evaluated in real time. At the same time, high-fidelity finite element (FE) simulations are too slow for iterative design and control. We develop a finite element-based calibration pipeline that combines a dependency-constrained human finger kinematic model with a segmented pseudo-rigid-body (PRB) description of ribbed-bellow soft pneumatic actuators sized to individual fingers. FE models with symmetry and contact generate pressure–pose data for the MCP, PIP, and DIP spans, from which we extract per-segment bending angles and axial elongations, fit simple pressure–kinematics relations, and identify PRB parameters using basin-hopping global optimization. The calibrated PRB reproduces FE flexion–extension trajectories for index and little finger actuators with millimetric accuracy (mean segment positioning errors of approximately 2.3 mm and 0.7 mm), preserves finger-like bending localized in the bellows, and maintains negligible compression of inter-joint links (below 1.2%). The pressure–bend and pressure–elongation maps achieve near-unity adjusted $R^2$, and the PRB forward kinematics evaluates complete pressure trajectories in less than half a millisecond, compared with several hours for the corresponding FE simulations. This pipeline provides a practical route from detailed FE models to controller-ready reduced-order surrogates for design-space exploration and patient-specific control of soft rehabilitation actuators. Full article

Advances in robotic technology for hand rehabilitation, particularly soft robotic gloves, have significant potential to improve patient outcomes. While vision-based algorithms pave the way for fast and convenient hand pose estimation, most current models struggle to accurately track hand movements when soft robotic gloves are used, primarily due to severe occlusion. This limitation reduces the applicability of soft robotic gloves in digital and remote rehabilitation assessment. Furthermore, traditional clinical assessments like the Fugl-Meyer Assessment (FMA) rely on manual measurements and subjective scoring scales, lacking the efficiency and quantitative accuracy needed to monitor hand function recovery in data-driven personalised rehabilitation. Consequently, few integrated evaluation systems provide reliable quantitative assessments. In this work, we propose an RGB-based evaluation system for soft robotic glove applications, which is aimed at bridging these gaps in assessing hand function. By incorporating the Hand Mesh Reconstruction (HaMeR) model fine-tuned with motion capture data, our hand estimation framework overcomes occlusion and enables accurate continuous tracking of hand movements with reduced errors. The resulting functional metrics include conventional clinical benchmarks such as the mean per joint angle error (MPJAE) and range of motion (ROM), providing quantitative, consistent measures of rehabilitation progress and achieving tracking errors lower than 10°. In addition, we introduce adapted benchmarks such as the angle percentage of correct keypoints (APCK), mean per joint angular velocity error (MPJAVE) and angular spectral arc length (SPARC) error to characterise movement stability and smoothness. This extensible and adaptable solution demonstrates the potential of vision-based systems for future clinical and home-based rehabilitation assessment. Full article

This paper investigates the effect of tactile feedback on the power efficiency and timing of controlling a remote robotic arm using a custom-built haptic glove. The glove integrates flex sensors to monitor finger movements and vibration motors to provide tactile feedback to the user. Communication with the robotic arm is established via the ESP-NOW protocol using an Arduino Nano ESP32 microcontroller (Arduino, Turin, Italy). This study examines the impact of tactile feedback on task performance by comparing precision, completion time, and power efficiency in object manipulation tasks with and without feedback. Experimental results demonstrate that tactile feedback significantly enhances the user’s control accuracy, reduces task execution time, and enables the user to control hand movement during object grasping scenarios precisely. It also highlights its importance in teleoperation systems. These findings have implications for improving human–robot interaction in remote manipulation scenarios, such as assistive robotics, remote surgery, and hazardous environment operations. Full article

Foundation models excel on general benchmarks but often underperform in clinical settings due to domain shift between internet-scale pretraining data and medical data. Multimodal deep learning, which jointly leverages medical images and clinical text, is promising for diagnosis, yet it remains unclear whether domain adaptation is better achieved by fine-tuning large vision–language models or by training lighter, task-specific architectures. We address this question by introducing PairDx, a balanced dataset of 22,665 image–caption pairs spanning six medical document classes, curated to reduce class imbalance and support fair, reproducible comparisons. Using PairDx, we develop and evaluate two approaches: (i) PairDxCLIP, a fine-tuned CLIP (ViT-B/32), and (ii) PairDxFusion, a custom hybrid model that combines ResNet-18 visual features and GloVe text embeddings with attention-based fusion. Both adapted models substantially outperform a zero-shot CLIP baseline (61.18% accuracy) and a specialized model, BiomedCLIP, which serves as an additional baseline and achieves 66.3% accuracy. Our fine-tuned CLIP (PairDxCLIP) attains 93% accuracy and our custom fusion model (PairDxFusion) reaches 94% accuracy on a held-out test set. Notably, PairDxFusion achieves this high accuracy with 17 min, 55 s of training time, nearly four times faster than PairDxCLIP (65 min, 52 s), highlighting a practical efficiency–performance trade-off for clinical deployment. The testing time also outperforms the specialized model—BiomedCLIP (0.387 s/image). Our results demonstrate that carefully constructed domain-specific datasets and lightweight multimodal fusion can close the domain gap while reducing computational cost in healthcare decision support. Full article

This study investigates the anti-cut properties of a composite based on basalt fabrics with varied structural characteristics, including weave and thread density, enhanced with reduced graphene oxide (rGO). The primary aim is to evaluate the potential of integrating rGO into a basalt matrix to improve its resistance to cutting and mechanical damage. The results indicate that adding rGO significantly increases the cutting resistance of the composite. Molecular simulations demonstrate that the composite, which combines a cross-linked LG 700 resin, rGO, and basalt, is one of the most thermodynamically stable configurations due to strong electrostatic interactions between its components. These interactions and the formation of physical bonds at the interfaces stiffen the material, while also allowing for a unique crack-toughening effect. This resilience, which enables the reformation of physical interactions after stress, directly contributes to the composite’s enhanced resistance to catastrophic failure and its observed performance in cutting tests. These findings suggest that basalt–resin with rGO composites hold great potential for applications requiring high mechanical strength and durability, such as protective clothing (e.g., gloves) and anti-vandalism materials. The study concludes that the developed composite represents a promising advancement for materials exposed to cutting forces. Full article

Assessing grip strength is fundamental to ergonomics, medicine, sports, and rehabilitation, as it reflects the functional capabilities of the upper limb. The main goal of this paper is to design and create a device that measures hand force to help keep workers’ physical workload safe and sustainable in factories. The paper looks at the long-term health and safety of employees, explaining what sensors and methods can be used to measure hand force and pressure. It describes a new device for measuring hand grip, which was tested and adjusted in real factory conditions. This device could be used to measure force when using tools and to check for risks of heavy physical strain at work. The system helps find possible injury risks in any type of workplace, especially where people need to grip with their hands many times. It checks if workers use too much force, especially in jobs with repeated tasks where the same hand movement happens again and again. The text suggests that it is important to track hand forces during any job that repeats more than 40 times in a shift, no matter how long the shift is. The system is made up of a smart glove and software for analysis. Its purpose is to estimate how much force is needed to complete a task safely, not the maximum force a person can make. The goal is to measure the required force for the job so that workers’ workload remains safe and healthy over the long term. Full article

People with Parkinson’s disease often show changes in their movement abilities during the day, especially around the time they take medication. Being able to record these variations in an objective way can help doctors adapt treatment and follow disease changes more closely. A methodology for quantitative motor assessment is proposed in this work. It employs data from a custom SmartGlove equipped with inertial sensors. A multi-method feature selection scheme is developed, integrating statistical significance, model-based importance, and variance contribution. The most significant features were retained, and higher-level artificial features were generated using Grammatical Evolution (GE). The framework combines multi-criteria feature selection with evolutionary feature construction, providing a compact and interpretable representation of motor behavior. Additionally, the framework highlights nonlinear and composite features as potential digital biomarkers for Parkinson’s monitoring. The method was validated on recordings collected from Parkinson’s patients before and after medication intake. The recordings have been retrieved during four standardized hand motor tasks targeting tremor, bradykinesia, rigidity, and general movement anomalies. The proposed method was compared with five existing machine learning models based on artificial neural networks. GE-based features reduced classification errors to 10–19%, outperforming baseline models. Furthermore, the proposed methodology performs prediction and recall 80–88%. Full article

This study presents the design and evaluation of a multi-model artificial intelligence (AI) framework for proactive quality risk management in projects. A dataset comprising 2000 risk records was developed, containing four columns: Risk Description (input), Risk Category, Trigger, and Impact (outputs). Each output variable was modeled using three independent classifiers, forming a multi-step decision-making pipeline where one input is processed by multiple specialized models. Two feature extraction techniques, Term Frequency–Inverse Document Frequency (TF-IDF) and GloVe100 Word Embeddings, were compared in combination with several machine learning algorithms, including Logistic Regression, Support Vector Machines (SVMs), Random Forest, Multinomial Naive Bayes, and XGBoost. Results showed that model performance varied with task complexity and the number of output classes. Trigger prediction (28 classes), Logistic Regression, and SVM achieved the best performance, with a macro-average F1-score of 0.75, while XGBoost with TF-IDF features produced the highest accuracy for Risk Category classification (five classes). In Impact prediction (15 classes), SVM with Word Embeddings demonstrated superior results. The implementation, conducted in Python (v3.9.12, Anaconda), utilized Scikit-learn, XGBoost, SHAP, and Gensim libraries. SHAP visualizations and confusion matrices enhanced model interpretability. The proposed framework contributes to scalable, text-based, predictive, quality risk management, supporting real-time project decision-making. Full article

Glove fit is crucial for both wear comfort and safety. A well-fitting glove can be realized by combining anthropometric hand measurements from three-dimensional (3D) scanning with the finite element method (FEM). This study reviews how accurate hand measurements and model interactions can be achieved to improve design and enhance protection. A total of 26 articles were selected for an integrated analysis and evaluation. The results indicate an increase in accuracy in 3D scanning with greater resolution, in which the optimum value has not been discovered. While the numbers of landmarks (ranging from 14 to 50) depend on the specific purpose, they do not directly correlate with precision. On the other hand, the authenticity of the FEM is closely related to the number and size of the finite elements, with simulation error decreasing as the applied force increases (R² > 0.78). It is also noteworthy that the image-based approach, motion state, and model used in the FEM do not significantly affect precision. Both technologies provide a comprehensive approach for glove design, as they combine accurate anatomical data with predictive modeling of mechanical performance and fit. Yet, challenges are identified, such as divergent standards in data retrieval and low accessibility, which inhibit the application of these two techniques in glove construction. Future studies should address the issues of improving scanning coverage, standardizing the data collection models, expanding glove application in different fields, and adopting artificial intelligence to improve the design, construction, or manufacture of gloves. Full article

Sentiment analysis is essential for understanding consumer opinions, yet selecting the optimal models and embedding methods remains challenging, especially when handling ambiguous expressions, slang, or mismatched sentiment–rating pairs. This study provides a comprehensive comparative evaluation of sentiment classification models across three paradigms: traditional machine learning, pre-transformer deep learning, and transformer-based models. Using the Amazon Magazine Subscriptions 2023 dataset, we evaluate a range of embedding techniques, including static embeddings (GloVe, FastText) and contextual transformer embeddings (BERT, DistilBERT, etc.). To capture predictive confidence and model uncertainty, we include categorical cross-entropy as a key evaluation metric alongside accuracy, precision, recall, and F1-score. In addition to detailed quantitative comparisons, we conduct a systematic qualitative analysis of misclassified samples to reveal model-specific patterns of uncertainty. Our findings show that FastText consistently outperforms GloVe in both traditional and LSTM-based models, particularly in recall, due to its subword-level semantic richness. Transformer-based models demonstrate superior contextual understanding and achieve the highest accuracy (92%) and lowest cross-entropy loss (0.25) with DistilBERT, indicating well-calibrated predictions. To validate the generalisability of our results, we replicated our experiments on the Amazon Gift Card Reviews dataset, where similar trends were observed. We also adopt a resource-aware approach by reducing the dataset size from 25 K to 20 K to reflect real-world hardware constraints. This study contributes to both sentiment analysis and sustainable AI by offering a scalable, entropy-aware evaluation framework that supports informed, context-sensitive model selection for practical applications. Full article

To investigate the characteristic pressure distribution patterns when gripping ski poles during skiing, this study addresses the challenges of measuring grip force on the complex curved surfaces of ski poles. A dataset of experimental samples was established, and grip force data were extracted using deep neural network (DNN) training. To reduce errors caused by dynamic force distribution and domain shifts due to varying hand postures, a hybrid method combining deep neural networks with the bio-inspired Gray Wolf Optimization (GWO) algorithm was proposed. This approach enables the fusion of hand-related feature data, facilitating the development of a high-precision grip force prediction model for skiing. A multi-point flexible array sensor was selected to detect force at key contact points. Through system calibration, grip force data were collected and used to construct a comprehensive database. A backpropagation (BP) neural network was then developed to process the sensor data at these characteristic points using deep learning techniques. The data fusion model was trained and further optimized through the GWO-BPNN (Gray Wolf Optimizer–backpropagation neural network) algorithm, which focuses on correcting and classifying force data based on dominant force-bearing units. Experimental results show that the optimized model achieves a relative error of less than 2% compared to calibration experiments, significantly improving the accuracy of flexible sensor applications. This model has been successfully applied to the development of intelligent skiing gloves, offering a scientific foundation for performance guidance and evaluation in skiing sports. Full article