The aim of the present study was to map the scientific landscape and to provide a quantitative and network analysis of global research on the impact of psychological stress on chronic diseases from 1950 to 2025. This descriptive and analytical study used bibliometric and scientometric methods. Data were retrieved from three international databases, PubMed, Scopus, and Web of Science, on 20 March 2025. After removing duplicates and applying inclusion and exclusion criteria, 2134 original research articles were selected for final analysis. Data processing was performed in Python and analyses were conducted with VOSviewer (version 1.6.20). Descriptive indicators, author co-authorship networks, keyword co-occurrence, and source co-citation were examined. Of the 2134 articles published between 1950 and 2025, publication output demonstrated a consistent upward trajectory. The largest shares of publications were contributed by the United States (21.4

Mental health has emerged as a critical public health concern across the Middle East, influenced by conflict, displacement, sociocultural factors, and limited health infrastructure. Despite growing research activity, a comprehensive mapping of regional scientific output has been lacking. This bibliometric and scientometric analysis examined peer-reviewed journal articles on mental health in the Middle East from January 2000 to June 2025. Data were retrieved from Web of Science, Scopus, and PubMed. Eligible publications were screened for regional affiliation and topical relevance. Bibliometric indicators were analyzed using Microsoft Excel and VOSviewer to evaluate authorship, institutional collaboration, country distribution, keyword co-occurrence, and thematic clusters. A total of 15,214 articles were identified across the three databases. Iran, Turkey, and Saudi Arabia accounted for the largest share of publications. Co-authorship and institutional networks revealed concentrated collaboration hubs. Keyword co-occurrence analysis identified five major thematic clusters focusing on trauma, anxiety, depression, stigma, and youth mental health, along with several emerging topics. Mental health research in the Middle East has expanded substantially over the past 25 years, though notable geographic disparities and thematic gaps persist. Strengthening cross-national collaboration and developing context-specific interventions remain essential priorities.

This study examined how Iranian experts in mental health, information technology, and public health perceive the potential role of artificial intelligence (AI) in counseling for anxiety and depression. Particular attention was given to perceived effectiveness, requirements for cultural adaptation, and ethical considerations surrounding the integration of AI into clinical practice. A qualitative design was employed using semi-structured interviews with 25 experts from seven regions of Iran (Ardabil, Tehran, Mashhad, Sanandaj, Zahedan, Tabriz, and Ahvaz). Interviews were conducted between May and August 2024 and analyzed through reflexive thematic analysis following Braun and Clarke’s framework. Data collection and analysis proceeded iteratively, and thematic saturation was reached after the twenty-third interview. Five overarching themes captured expert perspectives: (1) perceived usefulness of AI for individuals with mild to moderate symptoms (reported by 21 of 25 experts); (2) opportunities for improving accessibility in underserved regions (noted by 23 experts); (3) the central importance of cultural and religious alignment in system design (highlighted by 18 experts); (4) ethical and relational concerns, including privacy, data governance, and the absence of genuine empathy (raised by 20–22 experts across subthemes); and (5) recommendations for responsible implementation, such as regulatory oversight, clinician training, and the use of culturally representative datasets (noted by 17–19 experts). Experts viewed AI-assisted counseling as a potentially helpful complementary resource rather than a substitute for human-delivered care. Successful integration requires culturally sensitive design, transparent data governance, and national guidelines for safe implementation. The findings provide actionable insights for policymakers, developers, and mental-health authorities planning to incorporate AI-driven tools within Iran and other culturally comparable settings.

This systematic review and meta-analysis evaluated speech-based computational approaches for diagnosing schizophrenia and assessing symptom severity. This study is a systematic review and meta-analysis conducted in accordance with PRISMA guidelines. A comprehensive search of PubMed, Scopus, and Web of Science databases was performed from inception to August 2025. Eligible studies included individuals diagnosed with schizophrenia and employed computational analyses of speech or language, such as natural language processing, acoustic feature extraction, or large language models, and reported diagnostic accuracy metrics or associations with validated clinical scales. Two independent reviewers screened titles, abstracts, and full texts, extracted data, and assessed methodological quality. When sufficient quantitative data were available, random-effects meta-analysis models were applied to pool diagnostic performance and correlation estimates. Diagnostic performance refers to supervised classification accuracy relative to clinician-assigned reference diagnoses rather than autonomous clinical diagnosis. Thirty-nine studies met inclusion criteria, with 16 contributing data to meta-analysis. Pooled diagnostic performance was high (AUC≈0.84), while speech-derived features showed moderate association with symptom severity (r ≈ 0.34). Large language models and multimodal pipelines outperformed traditional machine-learning and unimodal systems. Speech measures were most sensitive to negative symptoms, including alogia and reduced affect, whereas coherence-based metrics captured thought disorder. Sensitivity analyses confirmed robustness, and no meaningful publication bias was observed. Computational speech analysis demonstrates strong potential as an objective, scalable adjunct for schizophrenia assessment. Future research should prioritize standardized protocols, larger multicenter cohorts, and real-world clinical integration to support translation into practice. Not applicable. Not applicable.

The objective of this study was to investigate the psychometric properties of the Persian version of the social motivation questionnaire. The current research is a descriptive-correlational investigation. The research community comprised all students of Farhangian, Amol, and Shamal universities who were studying in the academic year 2020-2021 at Farhangian, Amol, and Shomal universities. Using the electronic sampling procedure, 502 students (425 girls and 77 boys) from this community completed the social motivation questionnaire (SMQ). The data were used with statistical software (SPSS version 20) and (LISREL) version 8.80 to check the reliability and validity of the questionnaire, internal consistency coefficient, and face validity and confirmatory factor analysis. The sample size was arbitrarily divided into two groups to conduct statistical analysis in accordance with the study’s objectives. The first group underwent exploratory factor analysis, while the second group underwent confirmatory factor analysis. The results of exploratory factor analysis indicated that 8 items were assigned onto 2 factors, which collectively accounted for 58% of the variance in the social motivation questionnaire. The model’s fit indices were at the optimal level, with CFI=0.98, NFI=0.96, NNFI=0.97, GFI=0.97, AGFI=0.94, and RMSEA=0.05. Additionally, the Cronbach’s alpha coefficient was calculated to be 0.71 for the entire 8-item questionnaire, indicating that the items possessed a high level of internal consistency. The confirmatory factor analysis results demonstrated that the two-factor model of the questionnaire is consistent with the data, as demonstrated by the structural equation modeling method and LISREL software. In general, the results suggested that the Persian version of the social motivation questionnaire’s psychometric properties were suitable for the Iranian population, and that this instrument could be employed in research.

Background: Suicidal ideation, a serious mental health concern, threatens individuals’ well-being and has significant social and psychological consequences. This study aimed to compare the effectiveness of the Modified Unified Protocol (UP) and Transcranial Direct Current Stimulation (tDCS) on cognitive flexibility, emotional regulation strategies, and repetitive negative thinking in women with suicidal ideation.Methods: This study was quasi-experimental with pre-test and post-test design and a six-month follow-up. The participants included women with suicidal ideation, recruited from counseling centers affiliated with the State Welfare Organization in Ardabil, Iran between 2022 and 2024. Using criterion‑based sampling and random assignment, 75 participants were divided into three groups (n=25 each): Modified Unified Protocol (UP), transcranial Direct Current Stimulation (tDCS), and control. The intervention lasted 12 weeks (UP: individual sessions on emotion regulation & cognitive flexibility; tDCS: 2 mA stimulation at F3/F4 areas). Data were collected at pre-test, post-test, and six-month follow-up using the Cognitive Flexibility Inventory (CFI), the Difficulties in Emotion Regulation Scale–Short Form (DERS-SF), and the Repetitive Thoughts Questionnaire (RTQ). For data analysis, ANCOVA was used to examine between-group differences, and repeated measures ANOVA was used to assess within-group changes over time. All tests were conducted using SPSS version 27, with statistical significance set at P<0.05.Results: For cognitive flexibility, mean scores increased in the UP group (76.80±10.06 to 100.45±14.12) and tDCS group (79.40±11.07 to 91.35±10.49), with minimal change in the control group. Significant differences were found between UP and control groups (P=0.001) and between tDCS and control groups (P=0.008). In cognitive emotion regulation difficulties, scores decreased in the UP group (72.55±7.72 to 52.25±7.90) and tDCS group (72.45±8.48 to 61.70±9.29), while the control group remained unchanged. Post-test comparisons showed significant differences; UP vs. control (P=0.001), tDCS vs. control (P=0.001), and UP vs. tDCS (P=0.001). Repetitive negative thinking decreased in the UP group (39.20±5.85 to 20.85±5.14), moderately in the tDCS group (29.05±6.72), and showed no change in the control group. All between-group comparisons in post-test were significant (P=0.001).Conclusions: Both the Modified Unified Protocol (UP) and tDCS were effective in enhancing cognitive flexibility, improving emotion regulation strategies, and reducing repetitive negative thinking in women with suicidal ideation. However, UP showed a more substantial and sustained impact over time, particularly in emotion regulation and repetitive negative thought reduction.

This randomized trial compared the efficacy of Internal Family Systems (IFS) and Emotion-Focused Training for Self-Compassion and Self-Protection (EFT-SCP) on psychophysiological stress reactivity, anger rumination, and fear of compassion among adolescents with family-related traumatic experiences. From 180 screened adolescents in Out-Of-Home Care (OOHC), 75 were randomized, with 67 participants (IFS = 23, EFT-SCP = 22, Control = 22) completing the two-month follow-up. Data were analyzed using Multivariate Repeated-Measures Analysis of Variance (MANOVA) and Galvanic Skin Response (GSR). Findings revealed significant group-by-time interactions (p < .001). Psychologically, EFT-SCP demonstrated superior immediate post-test efficacy in reducing anger rumination and fear of compassion. However, the IFS group exhibited an incremental improvement trajectory, significantly outperforming EFT-SCP at follow-up in reducing perceived stress reactivity and fear of compassion. Physiologically, while both interventions decreased autonomic arousal, no significant difference in GSR was observed at post-test. Notably, only at follow-up did the IFS group achieve a more stable and superior reduction in physiological stress reactivity (GSR) compared to EFT-SCP. EFT-SCP facilitates rapid psychological stabilization, whereas IFS promotes progressive, long-term transformations across both subjective and physiological markers. This underscores the need for multi-level, process-oriented interventions to address the complex psychosocial consequences of childhood trauma.

The present study aimed to examine the causal model of negative emotionality based on cognitive fusion, with the mediating roles of acceptance and cognitive flexibility among adolescents. This applied study employed a descriptive correlational design using structural equation modeling (SEM). The statistical population consisted of all junior and senior high school students in Urmia during the 2024–2025 academic year. A total of 394 participants were selected through multistage cluster sampling. Data were collected using the Positive and Negative Affect Schedule (PANAS; Watson & Tellegen, 1988), the Cognitive Fusion Questionnaire (CFQ; Gillanders et al., 2014), the Cognitive Flexibility Inventory (CFI; Dennis & Vander Wal, 2010), and the Acceptance and Action Questionnaire–II (AAQ II; Bond et al., 2011). Data analysis was conducted using Pearson correlation coefficients and structural equation modeling through SPSS and AMOS software. The findings indicated that cognitive fusion had a positive and significant direct effect on negative emotionality, whereas acceptance and cognitive flexibility had negative and significant direct effects on negative emotionality. Furthermore, bootstrap analysis revealed that cognitive fusion exerted a significant indirect effect on negative emotionality through acceptance and cognitive flexibility, with the 95% confidence interval for the indirect effect excluding zero. Overall, the results suggest that cognitive fusion plays a significant role in increasing adolescents' negative emotionality, and this effect is strengthened through reduced levels of acceptance and cognitive flexibility. These findings may contribute to the development of interventions based on third wave cognitive behavioral therapies aimed at reducing negative emotionality among adolescents.

Cognitive dysfunction is a major nonmotor feature of Parkinson's disease (PD) that impairs quality of life and increases caregiver burden. Its intellectual structure and thematic development remain underexplored. This bibliometric and scientometric study analyzed English-language publications on PD-related cognitive dysfunction indexed in Web of Science, Scopus, and PubMed (2000-May 2025). After screening and duplicate removal, 8721 articles were included. VOSviewer 1.6.20, Python, and Microsoft Excel were used for co-authorship, bibliographic coupling, co-word, and citation analyses. Results showed 8721 publications authored by 34,311 researchers, with the United States, China, and the United Kingdom leading in output and international collaboration (collaboration index 8.02). Four thematic clusters were identified: diagnostic criteria and staging, neurobiological mechanisms, nonmotor symptoms and quality of life, and therapeutic interventions. Findings highlight a mature but evolving research field, emphasizing multidisciplinary integration, stronger collaboration, and the need to explore biomarkers, neuroinflammation, and novel therapeutic approaches.

BACKGROUND:Adolescents in Out-of-Home Care (OOHC) often suffer multiple Adverse Childhood Experiences, leading to high rates of distorted Post-Traumatic Cognitions (PTC) and Fear of Compassion (FC). Tailored, evidence-based psychological interventions are essential for this group. OBJECTIVE:This mixed-methods study aimed to design, validate, and empirically evaluate a Internal Family Systems (IFS)-based group intervention for trauma-exposed adolescents in Out-of-Home Care (OOHC). PARTICIPANTS AND SETTING:Fifty trauma-exposed adolescents (27 female, 23 male) were purposively selected from 180 OOHC residents and randomly assigned to the IFS (n = 25) or control (n = 25) group. METHODS:The IFS protocol was first developed and validated qualitatively via content analysis and expert review. The quantitative quasi-experimental phase utilized a pre-post-follow-up design to assess PTC and FC using standardized measures across three time points. RESULTS:IFS led to significant reductions across all PTC subscales. The strongest and most stable effect was on Negative Self-Cognitions (η2 = 0.40), while the most immediate change, aligning with IFS theory regarding Protector Parts, was seen in Self-Blame (η2 = 0.22). FC also decreased significantly, with the largest improvements in FC-for oneself (η2 = 0.31) and FC-for others (η2 = 0.27). These effects were maintained at the two-month follow-up. CONCLUSIONS:Findings support the effectiveness of the IFS in improving trauma-related cognitive and emotional outcomes. Crucially, the change pattern suggests that IFS promotes recovery by first establishing internal self‑leadership and self-compassion, which then gradually generalizes to interpersonal relationships. This provides a promising model for structural trauma recovery in vulnerable youth.

Abstract Objective This study aimed to compare levels of moral, emotional, and spiritual intelligence in parents of aggressive versus non‐aggressive children. Methods A causal‐comparative design was employed. The study population included parents of elementary school students in Ardabil, Iran. Using a multistage sampling method, 60 participants were selected. Research instruments included the Shahim Aggression Questionnaire (2006), Lennick and Kiel's Moral Intelligence Questionnaire (2011), Bar‐On Emotional Intelligence Questionnaire (1997), and King's Spiritual Intelligence Questionnaire (2008). Data were analyzed using multivariate analysis of variance (MANOVA). Basic demographic information was collected, and sampling procedures aimed to reduce socioeconomic variability between groups. Results MANOVA revealed a significant overall group difference across parental moral, emotional, and spiritual intelligence; F(3, 58) = 27.10, p < 0.001, η2 = .421. Parents of aggressive children showed lower mean scores across all domains, with the largest difference in emotional intelligence (M = 2.21 vs. 3.11). Follow‐up univariate analyses confirmed significant group differences for moral intelligence, F(1, 58) = 61.43, p < 0.001; emotional intelligence, F(1, 58) = 78.64, p < 0.001; and spiritual intelligence, F(1, 58) = 56.34, p < 0.001. These results indicate associations between parental intelligence dimensions and child aggression, without implying causal direction. Conclusions The findings highlight that parental moral, emotional, and spiritual intelligence are linked to differences in children's aggressive behavior. These results suggest that parent‐focused interventions, such as programs to enhance emotional regulation, ethical decision‐making, and meaning‐oriented coping skills, may help parents strengthen their interactions with children, supporting adaptive behaviors and potentially reducing maladaptive outcomes.

The global demand for mental health services is increasing, yet access to professional therapists remains limited. In this context, artificial intelligence (AI)-based solutions, especially therapeutic chatbots, are gaining attention as complementary tools. However, user adoption is hindered by trust issues. This study aimed to develop a grounded theory model to explore how trust in AI-driven therapeutic systems is built, maintained, and potentially lost. Trust is conceptualized as a multi-dimensional construct, incorporating system reliability (from human-computer interaction), healthcare credibility (including safety and ethics), and relational trust reflecting therapeutic alliance elements like empathy. A qualitative grounded theory approach was used, analyzing 557 public text units from social media, mental health forums, and related online platforms. Through open, axial, and selective coding, a mid-level conceptual framework was developed, resulting in the "Credential Reappraisal" model. This model outlines three stages: initial evidence appraisal, interactional testing based on empathy and accuracy, and trust stabilization or collapse. Social feedback and the system’s error management were found to be central to strengthening or undermining trust. The study reveals that trust in AI-based therapists is not static but dynamic, continuously renegotiated through interactions with the system and feedback from others. Transparency, accuracy, and error management are key factors in influencing trust. The study emphasizes the need for trust-by-design in developing AI-driven therapeutic systems, ensuring transparency, explainable responses, and active user feedback mechanisms for continuous improvement.

BACKGROUND:This study presents a bibliometric and scientometric analysis of research trends in schizophrenia genetics over nearly 7 decades (1957-2025). The field has evolved from early heritability studies to genome-wide association studies and multi-omics approaches. The objective was to map the historical trajectory, thematic evolution, key contributors, and global collaboration patterns in relation to clinically relevant psychological constructs. METHODS:An integrated bibliometric and scientometric approach was applied. A total of 5679 publications were included after systematic retrieval and PRISMA-guided screening from Scopus, PubMed, and Web of Science (initial n = 6193; final n = 5679). Data preprocessing included deduplication, keyword normalization, and metadata verification. VOSviewer (version 1.6.20) was used to construct co-authorship, keyword co-occurrence, and source co-citation networks. Thresholds were defined empirically to balance network interpretability and coverage. RESULTS:The findings suggest a gradual increase in research activity beginning in the 1990s, followed by a marked acceleration after 2010. Four thematic clusters were identified: neurobiological mechanisms and endophenotypes; basic genetic foundations; pharmacological treatments; and clinical comorbidities. Temporal patterns indicate a shift between 2012 and 2016 from genetic association studies toward functional genomics and cognitive neuroscience. Prominent contributors include J. van Os, R.E. Gur, and M.T. Tsuang. The United States may play a prominent role within the global collaboration network. However, clinically relevant psychological constructs may be less centrally integrated within the network structure. CONCLUSION:This study provides a structured bibliometric mapping of schizophrenia genetics research from 1957 to 2025. The findings may suggest increasing movement toward large-scale, consortium-based research models and a possible partial separation between clinical and basic science domains. Importantly, the results highlight a relative underrepresentation of clinically interpretable psychological dimensions. These findings may indicate the need for stronger integration between genetic discoveries and clinical psychological frameworks to enhance translational relevance.

BACKGROUND AND OBJECTIVE:This study provides a comprehensive bibliometric and scientometric analysis of research trends in schizophrenia genetics from 2020 to 2025. Driven by rapid technological advances such as genome-wide association studies (GWAS) and next-generation sequencing (NGS), the field has experienced significant growth. The objective was to map the thematic evolution, key contributors, and collaboration networks shaping psychiatric genomics research. METHODS:An integrated bibliometric and scientometric approach was employed. A dataset of 5001 English-language publications authored by 27,692 researchers was retrieved from PubMed, Scopus, and web of science (Wos) databases. Following deduplication and eligibility screening, VOSviewer (version 1.6.20) was used to generate network visualizations analyzing co-authorship, keyword co-occurrence, citation, and co-citation patterns. RESULTS:The analysis revealed a notable surge in publications after 2022, increasing from 431 articles in 2020 to 1,645 articles in 2022, before stabilizing above 900 publications annually in 2023 and 2024. Six major thematic clusters were identified: (1) genetic and cellular mechanisms; (2) neurochemical and behavioral studies; (3) neuroimaging; (4) clinical and epidemiological features including polygenic risk scoring; (5) methodological approaches; and (6) comorbidities with pharmacogenetic implications. Prominent contributors included Ole A. Andreassen and Vince D. Calhoun, while the United States, China, and the United Kingdom emerged as central hubs of international collaboration. CONCLUSION:This study elucidates the leading researchers, prevailing themes, and collaborative networks driving schizophrenia genetics research from 2020 to 2025. The findings demonstrate the efficacy of bibliometric and scientometric methods, particularly VOSviewer, in revealing the dynamic landscape of psychiatric genomics and guiding future multidisciplinary research efforts.This graphical abstract illustrates the integrated workflow of the study "mapping the scientific landscape of schizophrenia genetics (2020-2025)". The process combines bibliometric and scientometric analyses of 5,001 publications using VOSviewer to identify thematic clusters, key contributors, and international collaboration networks that define global research trends in schizophrenia genetics.

Introduction: Children with learning disorders show memory and language deficits that may be related to oxidative stress. Objective: The aim of the present study was to investigate the possible relationships between cognitive deficits and oxidative stress markers in children with specific learning disorders through a systematic review. Method: The present study was a systematic review. The search period for articles in Persian and English was from the beginning of 2000 to the end of May 2025, which was conducted in databases (Science Direct, SID, WOS, PubMed, Scopus) and the Google Scholar search engine. During the review process, the following keywords were used: ("oxidative stress"), ("cognition" or "cognitive disorder"), ("learning disorder" or "dyslexia" or "mathematical learning disorder"), and "antioxidant". Finally, 7 relevant articles were selected based on the inclusion and exclusion criteria and were reviewed and analyzed. Results: The findings showed that in total, 564 individuals with SLD were compared with 596 normal individuals. A total of 9 oxidant biomarkers and 9 oxidant-related biomarkers were examined. Among oxidants, MDA was studied 3 times, and in all 3 cases it was higher in the SLD group. The antioxidant biomarkers GPx, SOD, and CoQ10 were each examined twice, and in all cases, the findings showed that the SLD group had lower values ​​than normal individuals. Results regarding the association between oxidative stress-related indices and intelligence and test scores related to SLD were inconsistent. Conclusion: The results indicate that some indicators of oxidative stress may be significantly associated with cognitive and learning performance in children with learning disorders. Therefore, it is better to further investigate these indicators so that they can be used as valid criteria for evaluating treatment, diagnosis, or even the severity of SLD.

Background: The Learning to BREATHE (L2B) curriculum is a school-based mindfulness intervention that has been used over the past two decades to promote adolescent mental health in educational settings. This program draws on mindfulness principles to enhance psychological well-being. In Iran, given the intense cultural and social pressures—particularly those related to highly competitive examinations—there is a growing need for novel, culturally appropriate mental health interventions. Aims: This pilot study aimed to investigate the feasibility, acceptability, safety, and preliminary effectiveness of the Learning to BREATHE curriculum on self-compassion, perceived stress, and mindfulness among Iranian adolescent girls. Methods: This quasi-experimental study used a pre-test and post-test design without a control group. The population included all female high school students aged 15–19 in Ardabil in 2024. Eight students were selected through non-random voluntary sampling. Measurement tools included the Self-Compassion Scale (Neff, 2003), Perceived Stress Scale (Cohen et al., 1983), and Mindfulness Scale (Brown & Ryan, 2003). Data were analyzed using paired t-tests in SPSS-26, with effect sizes calculated via Cohen’s d. Results: The attendance rate of 97.6%, completion rate of 87.5%, and high satisfaction levels confirmed the program’s feasibility and acceptability. No chronic adverse events were reported. The results demonstrated significant improvements in self-compassion (Cohen’s d = 1.12, p < 0.05), significant reductions in perceived stress (Cohen’s d = 1.3, p < 0.05), and significant increases in mindfulness (Cohen’s d = 1.2, p< 0.05) in the post-test compared to the pre-test. Conclusion: The Learning to BREATHE program shows high potential as an effective and culturally compatible intervention for promoting the mental health of Iranian adolescents. Despite limitations such as a small sample size and the absence of a control group, future research with larger samples and long-term follow-up is recommended.

Background:The rising prevalence of mental disorders, coupled with limited access to mental health services, underscores the urgent need for innovative solutions. Artificial Intelligence (AI) offers transformative potential in managing mental health conditions through multimodal data analysis. Objective:This study explores emerging applications of AI in early detection, personalized treatment, and the prevention of symptom escalation in mental disorders. Methods:A narrative review was conducted using comprehensive searches of PubMed, Scopus, and IEEE Xplore databases (2015-2025). Selected sources included studies on natural language processing (NLP), deep learning, and the analysis of multimodal data (eg, voice, text, and biosensor inputs). A qualitative synthesis was employed to identify key patterns, challenges, and innovations. Findings:AI enhances early detection through concepts such as a "psychological digital signature" and reports high performance in some studies (reported accuracies vary widely, eg, up to ~91% in selected cohorts). However, many high-accuracy reports derive from single-site or limited datasets with variable external validation; therefore, these figures should be interpreted cautiously. We discuss study-specific limitations (sample size, validation methods, and population diversity) in the Methods and Critical Appraisal sections. Conclusion:AI provides a patient-centered, preventive framework for reimagining mental health care. However, its effective integration requires robust ethical standards and digital infrastructure. Ethical considerations are critically linked to clinical implementation, particularly regarding privacy, fairness, and transparency in AI-assisted decision-making.

Background:Artificial intelligence (AI), through multimodal deep learning and predictive analytics, holds transformative potential in the prevention and treatment of mental disorders. This study explores the opportunities and challenges of these technologies. Objective:To present a conceptual framework for the responsible application of AI in mental health care. Methods:This integrative review analyzed selected sources from Google Scholar up to June 2025. Both qualitative and quantitative analyses were conducted to identify opportunities and challenges. Results:Key opportunities include early detection, personalized treatment, and enhanced access to mental health services. Major challenges involve ethical concerns, algorithmic bias, and data quality issues. Conclusion:AI can revolutionize mental health care, but it requires standardization and regulatory oversight. Future research should focus on addressing ethical dilemmas and improving data quality.