Refugee youth face elevated mental health risks yet evidence-based interventions remain limited. We systematically reviewed 67 studies of psychological interventions for refugee children and adolescents. Trauma-focused cognitive behavioral therapy showed strongest evidence for PTSD symptom reduction. School-based programs reached larger populations with moderate effectiveness for depression and anxiety. Cultural adaptation and community involvement improved intervention acceptability. Our synthesis identifies promising approaches while highlighting research gaps in humanitarian mental health programming.

Growing electric vehicle adoption creates unprecedented battery waste management challenges. We analyzed recycling technologies for lithium-ion batteries comparing pyrometallurgical and hydrometallurgical approaches. Direct recycling preserving cathode structures showed highest material recovery rates and lowest environmental impacts. Economic modeling identified scale thresholds for viable recycling operations. Extended producer responsibility policies and standardized battery designs could accelerate circular economy transitions. Our roadmap guides industry and policymakers toward sustainable EV battery lifecycles.

Museums seek innovative approaches to engage digitally-native visitors. We deployed augmented reality experiences in three Italian archaeological museums, overlaying reconstructed historical scenes on artifact displays. Eye-tracking and survey data from 840 visitors revealed significantly increased dwell times and information retention compared to traditional interpretive panels. Younger visitors showed strongest engagement while older visitors appreciated optional technology adoption. Our design framework balances technological enhancement with authentic museum experiences.

Rapid urbanization in Africa strains food supply systems serving growing city populations. We mapped food distribution networks in Kano, Nigeria examining market structures, transport logistics, and informal sector contributions. Urban food prices showed high volatility linked to rural production shocks and infrastructure constraints. Informal traders provided critical last-mile distribution in low-income neighborhoods. Policy interventions improving wholesale market facilities and transport connectivity could stabilize urban food access while supporting smallholder farmers.

Continuous glucose monitoring improves diabetes management but current sensors have durability limitations. We developed flexible biosensors using graphene-modified electrodes embedded in biocompatible hydrogel matrices. Novel enzyme immobilization techniques extended sensor operational lifetime to 21 days. Wireless data transmission enabled smartphone integration for real-time alerts. Clinical trials with 120 diabetic patients demonstrated accuracy comparable to commercial devices with superior comfort ratings. These advances support next-generation wearable health monitoring technologies.

Tourism development in indigenous communities risks cultural commodification while offering economic opportunities. We examined community-based tourism initiatives across 18 Aymara communities in the Bolivian highlands. Successful models maintained local decision-making authority and integrated tourism with traditional livelihood activities. Cultural preservation outcomes correlated with community ownership structures and benefit distribution mechanisms. Revenue reinvestment in language education and ceremonial practices strengthened cultural continuity. Our guidelines support culturally appropriate tourism development respecting indigenous self-determination.

Classical computing limitations constrain financial risk modeling for complex portfolios. We implemented quantum algorithms for Monte Carlo simulations on IBM quantum hardware, achieving speedups in value-at-risk calculations. Hybrid quantum-classical approaches outperformed purely classical methods for high-dimensional derivative pricing. Current hardware noise levels require error mitigation techniques for practical applications. Our framework identifies near-term opportunities for quantum advantage in financial services as hardware capabilities improve.

This study numerical analyses of unsteady magnetohydrodynamic (MHD) free convection flow of a viscous, incompressible, electrically conducting, and radiatively participating fluid past a vertical plate. The governing nonlinear partial differential equations are transformed into dimensionless form and solved numerically using an implicit Crank-Nicolson finite-difference scheme. The numerical stability of the solution method is established through von Neumann stability analysis. The vertical plate is impulsively started with an exponentially decaying velocity, and the wall temperature is suddenly raised (or lowered) while the wall concentration varies with time. The model considers a Darcy-type porous medium and accounts for the effects of species diffusion. The effects of dimensionless parameters like Hartmann number M, permeability parameter K, heat parameter Q, radiation parameter R, Prandtl number Pr, Schmidt number Sc, and Eckert number Ec, on the velocity and temperature fields are considered. The result shows that increasing magnetic field strength and radiation reduce velocity and temperature, while viscous dissipation enhances thermal energy. Higher permeability promotes flow speed, and buoyancy forces arising from both thermal and solutal gradients significantly enhance fluid speed, as depicted in graphs. The findings are relevant for engineering applications involving heat and mass transfer in porous, electrically conducting media such as magnetohydrodynamic (MHD) generators.

The recognition of emotions in text has become an essential component in the development of emotionally intelligent systems across various fields such as mental health, education, and conversational AI. Traditional methods, like TF-IDF and lexicon-based approaches, are interpretable but often lack the ability to capture contextual nuances and perform well across multiple languages. In contrast, deep learning techniques such as BiLSTM and XLM-R offer semantic richness but may lack transparency and psycholinguistic representation. To bridge the gap between interpretability and enriched emotion recognition informed by psycholinguistics, this paper proposes a human-centered, explainable hybrid approach. This approach combines statistical methods (TF-IDF), psycholinguistic tools (NRC, LIWC), and contextual deep learning embeddings (BERT, XLM-R) in a multi-model ensemble that includes XGBoost, BiLSTM, and XLM-R. The fusion of these models employs soft voting and feature concatenation. Moreover, explainability is enhanced through SHAP (SHapley Additive exPlanations) and attention mechanisms to improve model interpretability. Experiments conducted on multilingual emotion datasets demonstrate that our proposed system outperforms state-of-the-art benchmarks in terms of both performance and explainability. The system not only achieves superior emotion detection but also emphasizes interpretability, inclusivity, and psychological insight within the framework of cognitive and ethical AI.