Search Results (1,286)

Somali refugee mothers navigating parenting in the United States face compounding challenges that extend well beyond the initial resettlement period. This study employed a multi-method qualitative design, including utilizing a focus group and follow-up key informant interviews with Somali refugee mothers. Thematic framework analysis identified three overarching domains of challenges and resilience. First, a pervasive deficit of functional literacy, defined as the practical capacity to navigate American institutional systems, emerged as the primary stressor, superseding material poverty as a barrier to daily functioning. Second, significant intergenerational tensions were documented, including role reversal between mothers and children, erosion of parental authority, and breakdown of the traditional expectations that adult children provide financial and social support to aging parents. Third, single motherhood amplified all other stressors, producing progressive role strain and mental health decline in the absence of extended family support. Despite these challenges, participants demonstrated substantial resilience through informal mutual aid networks, religious practice, and deliberate cultural and linguistic preservation. Findings have direct implications for the design of culturally responsive resettlement programming, family counseling services, and mental health interventions for Somali refugee populations. Full article

Objectives: The rise in the frequency of mental health concerns (depression and suicide) expressed on social media calls for reliable, explainable, and efficient computational methods for mental health surveillance. In this paper, we propose an interpretable framework for text-based detection of post- and community-level mental health risk on social media. Methods: The framework combines (i) Secretary Bird Optimization (SBO) for feature selection of informative linguistic and psychological features, (ii) a BERT (Bidirectional Encoder Representations from Transformers)—CNN (Convolutional Neural Network) model for post-level reasoning, and (iii) a Graph Neural Network (GraphSAGE) for community-level reasoning. The graph is estimated based on semantic similarity between posts and author relations, instead of social interactions (e.g., mentions, replies) between authors. We use SHAP and LIME for model interpretability, uncertainty, and calibration analysis to evaluate the trustworthiness of predictions. Results: The model delivers 93.1% accuracy, 0.91 F1-score, and 0.944 ROC-AUC on the eRisk and CLPsych datasets using a strict user-disjoint validation strategy. SBO lowers the number of features by about 38%, leading to better generalization. The graph-based model enables improved learning of post and user representations by capturing relational dependencies. Conclusions: Our approach offers an explainable and robust means of detecting mental health risk from text. Graph-based representations of semantic and authorship interactions enable community-level analyses, while interpretability and uncertainty estimation facilitate possible human-in-the-loop decision-making. This research does not explicitly consider a human-in-the-loop experiment. Full article

Automated depression assessment is critical for scalable mental healthcare but faces dual challenges: the lack of clinical interpretability in “black-box” deep learning models and the excessive computational cost of large-scale fusion architectures. To bridge this gap, we propose NS-Dep-KAN, a novel neuro-symbolic framework that harmonizes DSM-5-guided reasoning with Kolmogorov–Arnold Networks (KANs). Our approach leverages a Large Language Model (LLM) to extract symbolic symptom evidence aligned with diagnostic criteria, which then guides the aggregation of multimodal features from frozen pretrained encoders (WavLM and Qwen). Unlike traditional Multi-Layer Perceptrons, the proposed KAN prediction head employs learnable B-spline activation functions to capture complex nonlinear symptom–severity mappings with extreme parameter efficiency. Evaluations on the DAIC-WOZ benchmark demonstrate that NS-Dep-KAN achieves state-of-the-art performance among audio-text models (MAE 2.69, 13.5% improvement over the three-modality baseline MSGAF at MAE 3.11), with only ∼4.9 K trainable parameters. Moreover, the framework offers inherent interpretability, revealing granular symptom contribution profiles that align with clinical intuition. This work establishes a path toward explainable trustworthy AI for mental health screening. Full article

This article examines the context and aims of We Feed the UK: a multi-site series of arts projects and exhibitions, organised by the Gaia Foundation, that were exhibited at venues across the United Kingdom from February 2024 to June 2025. These aims were to celebrate and advocate for diverse regenerative food production businesses and community initiatives through poetry and photography. The featured enterprises combine food production with objectives such as biodiversity renewal, community development, mental health support and social justice, and the article proposes that this combination of environmental advocacy and affective social issues appeals to a wide and diverse audience. It supports this proposal through an examination of the first photography project in the series: Johannes Pretorius’s Intervention and Renewal, that engaged with a Cumbrian dairy farm that successfully combines biodiversity regeneration, organic agriculture and educational initiatives. Drawing upon Actor–Network Theory and notions of time as they pertain to the photograph, this examination reveals a project that offers both familiar imagery of British pastoral tropes, and the contemporary realities of the British food production system. As such it offers multiple points of engagement for audiences, and an effective entry point for the We Feed the UK programme. Full article

Background/Objectives: Liuwei Dihuang (LWDH) is a classical plant-derived herbal formula widely used for cognitive decline. This study aimed to evaluate its efficacy and safety in cognitive disorders and to explore its potential pharmacological mechanisms using network pharmacology. Methods: We searched 11 databases through November 2024 for randomized controlled trials comparing LWDH plus conventional therapy with conventional therapy alone in cognitive disorders. Meta-analysis was performed for clinical outcomes, and herb–compound–target and disease-target datasets were integrated to identify core molecular modules. Results: Twelve randomized controlled trials involving 1137 participants were included. Adjunctive LWDH was associated with improvements in Mini-Mental State Examination scores (MD = 2.34, 95% CI 0.88–3.79), activities of daily living, and quality of life. However, substantial heterogeneity and methodological limitations, including unclear randomization and blinding, were observed across studies, indicating a potential risk of bias. Fewer adverse events were reported in the LWDH plus conventional treatment group, although reporting quality was limited. The overall risk of bias was judged as “some concerns”. Network pharmacology analysis identified a broad set of overlapping genes between LWDH-associated targets and cognitive disorder-related genes, which were further refined through filtering procedures. Subsequent analyses suggested associations with pathways related to neurodegeneration, apoptosis, and central nervous system function; however, these findings are exploratory and based on in silico predictions. Conclusions: LWDH may be associated with potential adjunctive benefits in cognitive disorders. However, given the methodological limitations and clinical heterogeneity of the included studies, the findings should be interpreted with caution. The proposed pharmacological mechanisms are exploratory and require further validation. Well-designed randomized controlled trials are needed to establish more robust evidence. Full article

People have become accustomed to studying or working with the guidance of artificial intelligence (AI) in recent years. Studies have begun investigating the risk factors of AI dependence, though most have used hypothesis-testing methods. The present study aimed to investigate predictors of AI dependence using machine-learning and network analysis, which are data-driven approaches. The included risk factors were Big Five personality traits, self-efficacy, depression, social anxiety, adverse childhood experiences, and AI use motivation, selected based on theories and empirical studies. Participants consisted of 1258 university students (942 females and 316 males) with a mean age of 22.11 years (SD = 2.69). Four machine-learning algorithms were tested, including Elastic Net, Random Forest, XGBoost, and LightGBM. Machine-learning results indicate that escape and social motivation for AI use, along with social anxiety, were the main predictors of AI dependence. Network analysis results show that escape and social motivation were the most central nodes, with the highest Expected Influence (EI) indices. This study indicates that when addressing mental health problems related to AI dependence, it is more effective to focus on emotional isolation and social interaction challenges rather than simply cutting down on AI use. Full article

This paper proposes an intelligent stress detection method based on convolutional neural networks and the DeepFace framework, addressing the challenges of increasingly prominent global mental health issues and the limitations of traditional psychological services in terms of early warning latency and coverage. A three-level cascaded strategy combining RetinaFace, MTCNN, and OpenCV is first employed for face detection and localization, and facial expression features are extracted via the DeepFace framework. By integrating Russell’s valence–arousal model with Lazarus’s cognitive appraisal theory, an emotion–stress mapping rule is constructed to convert seven-category emotion probability distributions into 1–5 scale stress values. The method employs a cloud–edge collaborative flow, with feature extraction performed at the edge and original images promptly destroyed to mitigate privacy risks. Experiments on public expression datasets indicate that the method achieves above 99% face detection accuracy, 84.99% emotion recognition accuracy, and 86.09% stress assessment consistency grounded in the emotion–stress mapping rule, with an average response time per frame of approximately 200 ms. Based on 233 multi-scenario surveys, some respondents show limited stress self-awareness, suggesting traditional self-reporting may have blind spots, and thus this method serves as a useful supplement. Full article

Heart rate variability (HRV) refers to variations in the time intervals between consecutive heart beats. Changes in HRV reflect changes in either sympathetic or decreased parasympathetic tone that can originate in the brain. This brain–heart connection has led to the proposal that HRV may have utility in the diagnosis of psychiatric conditions and/or be a predictor of the response to psychiatric medications. There have been attempts to improve the correlation between HRV and psychological and psychiatric conditions by using artificial intelligence or specific machine learning algorithms. The objective of this review is to synthesize data on the use of machine learning to improve accuracy in differentiating psychological conditions such as mental stress, as well as distinguishing persons with anxiety disorders, panic disorders, major depression disorders and schizophrenia from health subjects. Reported accuracies for the identification of mental stress vary from 42 to 94%, while accuracies for anxiety vary from 67 to 98%, panic disorders from 71 to 93% and depression from 71 to 95%. The ability of HRV to differentiate different psychological or psychiatric conditions from each other requires more investigation. The ‘best’ machine learning algorithm varied between studies, with some reporting the k-nearest neighbor algorithm, support vector machine, random forest, or neural networks to be the best. A number of studies combined HRV with other variables such as respiration, EEG, or electromyography to obtain a composite index, but in doing so obscured the independent contribution of HRV. In summary, HRV has shown promise in detecting abnormalities in a range of psychological and psychiatric conditions. The use of machine learning algorithms improves diagnostic accuracy. Full article

Background: Fibromyalgia is a chronic nociplastic pain condition often accompanied by mental health comorbidities, with anxiety and depression being the most prevalent. The objective of this umbrella review is to analyze the effects of physical exercise on anxiety and depression symptoms in individuals with fibromyalgia. Methods: Following Cochrane and PRIOR guidelines, a systematic search was conducted in PubMed, Web of Science, Scopus, and CINAHL Complete up to 28 August 2025. Systematic reviews with or without meta-analyses that evaluated physical exercise interventions in adults with fibromyalgia and reported anxiety or depressive symptom outcomes were included. Risk of bias was assessed with AMSTAR-2; overlap was evaluated using MOoR and CCA. Results: Fourteen reviews (eight meta-analyses, three systematic reviews, two meta-analyses treated as descriptive, and one network meta-analysis) were included, synthesizing 98 randomized controlled trials (RCTs) with 4325 participants (in the 12 reviews that provided data). The majority of the patients were women and people aged between 10 and 65. Regarding anxiety, five of seven reviews reported significant improvements. Aquatic exercise showed the greatest effect (SMD = −1.14). Regarding depression, eight of 11 reviews reported significant benefits. Aquatic exercise again stood out with the highest effect (SMD = −1.18). Adherence varied between 64% and 97%. Methodological quality according to AMSTAR-2 showed considerable heterogeneity. Conclusions: Physical exercise, especially aerobic and aquatic modalities, may support the reduction of symptoms of anxiety and depression in people with fibromyalgia. These findings support its inclusion in rehabilitation programs, although methodological and prescription variability suggests caution in interpreting optimal parameters. PROSPERO-ID: CRD42024590799. Full article

Background: Gender discrimination in sport remains a persistent global issue, reflected in women’s limited participation, leadership representation, media visibility, salary equity, and personal safety. These forms of discrimination also negatively affect athletes’ psychological well-being, mental health, and overall sports experience. Despite growing scholarly attention over the past three decades, a comprehensive quantitative synthesis of this research area has been lacking. Methodology: A bibliometric analysis of 397 peer-reviewed documents published between 1995 and 2026 was conducted using the Scopus database. Data were analysed through the Bibliometric R package 4.2.1 and Biblioshiny interface. Science-mapping techniques including keyword co-occurrence, thematic clustering, thematic evolution, and collaboration network analysis were combined with performance indicators such as annual publication output, leading sources, author productivity, and citation impact. Results: Scientific production increased markedly after the mid-2010s, involving 187 sources and 1106 authors, with rising collaboration and citation influence. Core research themes included gender inequality, leadership exclusion, media representation, harassment and abuse, and structural discrimination in sports systems. Importantly, many of these themes are directly linked to reduced athlete well-being, including increased stress, anxiety, and decreased participation. Recent thematic developments highlighted intersectionality, safeguarding, inclusion, governance, and athlete welfare. Conclusion: Research on discrimination against women in sport has evolved into a multidisciplinary, policy-relevant field. Addressing gender discrimination is essential not only to achieving equity but also to improving athletes’ subjective well-being and long-term participation in sport. However, significant gaps remain, particularly in Global South contexts and intervention-based studies, indicating the need for stronger evidence-driven strategies to advance gender equity, inclusion, and ethical governance in sport. Full article

This paper offers a comprehensive overview of academic research on sentiment analysis in urban built environments from 2000 to 2025. Based on data from the scientific database Scopus and drawing on bibliometric tools like Bibliometrix (R) and VOSviewer for performance analysis and scientific mapping, it identifies publication trends, key influential works, leading authors and institutions, funding sources, and thematic clusters. The final dataset comprises 1315 English-language documents authored by 3855 researchers across 160 sources, with a total of 14,058 citations worldwide. The academic production increased after 2009, peaking in 2025. Keyword and network analyses highlight central themes (and methodological approaches) to the study of sentiment analysis in urban built environments. These include social media platforms like Twitter/X, machine learning, smart cities, artificial intelligence, mental health, and urban planning. China, the USA, and India lead in publication output. Over the last twenty-five years, key publication outlets included Sustainability (Switzerland), Cities, and the International Journal of Environmental Research and Public Health, while the National Natural Science Foundation of China has been the main funder. The paper discusses how sentiment analysis can support urban planning and public health by linking environmental features to well-being and explores emerging methodological trends like deep learning, multimodal approaches, and context-aware models. Overall, it maps the field’s intellectual landscape and argues in future directions for human-centered, data-driven urban decision-making. Full article

The digital transformation of mental health practice increasingly requires psychologists to integrate technological literacy with emotional and cognitive skills. This study presents a pilot project combining Visual Education and Artificial Intelligence (AI) through a Computer Vision model for emotional recognition. A convolutional neural network (CNN), based on MobileNetV2, was trained to identify facial emotions and tested for educational use within a serious game for psychologists’ professional development. Using transfer learning, the model achieved an accuracy of about 75% under controlled conditions but only 15.54% on a biased dataset. These results reveal both the potential and limitations of AI in emotional learning. The findings are discussed in relation to visual literacy, digital mental health, and AI ethics, illustrating how computational bias can become a pedagogical tool for psychology professionals. Full article

Artificial intelligence applications in mental health diagnosis face persistent challenges related to interpretability, trust, and the integrity of results. This study presents a trust-aware explainable deep learning framework that combines systematic benchmarking, SHAP-based interpretability, and permissioned blockchain verification to achieve secure mental health classification. The Depression & Mental Health Classification Dataset was used, which contains 1999 records, 21 features, and 12 classes. Data preprocessing included categorical encoding and Z-score normalization for continuous variables. To ensure robust evaluation, a stratified train–test split was applied, and class imbalance was addressed using the Synthetic Minority Over-sampling Technique (SMOTE). Eight machine learning and deep learning models were assessed under identical preprocessing and validation settings. In addition, two models were proposed: Feature Attention XGBoost (FA-XGBoost) and Feature Attention Feedforward Neural Network (FA-FNN). The FA-FNN model achieved the best performance, attaining an accuracy of 96.00%, precision of 98.31%, recall of 97.31%, and F1-score of 98.04%. To address deep learning’s black-box limitation, SHapley Additive ExPlanations (SHAPs) were used to provide both global feature importance and instance-level explanations, enabling transparent identification of the most influential mental health markers. Beyond interpretability, a permissioned blockchain layer was added to provide tamper-proof logging and traceable verification of AI results. The framework securely stores cryptographic hashes of model versions, prediction results, and generated SHAP artifacts, including visualization images, without exposing sensitive medical data. By integrating explainable decision-making, high-performance classification, and blockchain-based trust enforcement, the proposed framework creates a transparent and secure pipeline suitable for real-world mental healthcare systems. Controlled experiments on a permissioned Ethereum-InterPlanetary File System (IPFS) network demonstrated predictable latency, stable throughput (≈28–30 transactions/s), and lower operational costs, proving the framework’s suitability for enterprise and healthcare deployments. Full article

Black children and youth in Canada often hold high educational aspirations; however, systemic biases, deficit-based perceptions, and structural barriers limit their opportunities. These challenges, linked to anti-Black racism, migration-related disruptions, and socioeconomic inequities, contribute to lower engagement, underrepresentation, and reduced access to equitable educational resources. This scoping review examines these intersecting factors to inform equity-focused policy and practice. Following the Arksey and O’Malley framework and reported according to PRISMA-ScR guidelines, this review analyzed 96 studies published from database inception to May 2024, including 55 qualitative, 37 quantitative, and 4 mixed-methods studies. Bibliometric analysis was used to summarize study characteristics, while a thematic synthesis guided by intersectionality identified patterns in barriers, experiences, and interventions. Findings indicate that Black children and youth face persistent barriers, including systemic racism, disproportionate disciplinary practices, and Eurocentric curricula, with inequities further shaped by intersections of race, immigration status, and socioeconomic position. At the same time, mentorship, sponsorship, and community networks support academic resilience. Reported interventions include anti-racism training for educators and school stakeholders, culturally responsive curricula, mentorship initiatives, mental health supports, and financial aid. Advancing equity for Black children and youth in Canada requires systemic reform, culturally responsive pedagogy, and intersectionality-informed policies. Future research should prioritize participatory and longitudinal designs to generate evidence on effective and scalable strategies that foster educational opportunity and well-being. Full article

Major depressive disorder (MDD) is a prevalent and severe mental disorder, and EEG-based automated detection has become a promising approach for auxiliary screening diagnosis. In this work, we propose a novel multiscale spatiotemporal graph neural network for MDD detection from multichannel EEG signals. Specifically, a left–right hemispheric partitioning prior is used to encode brain functional organization. Based on this partitioning, adaptive graphs are then constructed and graph message passing is performed to model intra-hemispheric interactions. The approach not only incorporates brain functional organization into the learning process but also enhances the extraction of discriminative features related to depressive brain dynamics. The proposed method was validated in a cross-subject scenario on a private resting-state EEG dataset including 54 adult participants (27 MDD patients and 27 healthy controls; age range: 27–48 years). Experimental results on the dataset achieve an accuracy of 92.21%, surpassing the baseline models. Meanwhile, ablation experiments demonstrate the effectiveness of our proposed method. Full article