Peruvian Union University

IntroductionSubstance use disorder (SUD) is a complex neurobiological disorder characterized by the consolidation of maladaptive neuroplasticity affecting dopaminergic, glutamatergic, and neurotrophic systems, as well as cortical and subcortical networks critical for executive control, emotional regulation, and associative learning.MethodsThis systematic review was conducted in accordance with PRISMA 2020 guidelines and integrated 57 studies published between 2020 and 2025 to analyze neuroplastic mechanisms involved in vulnerability to substance use disorder and brain recovery following chronic substance exposure.ResultsThe findings revealed consistent alterations in synaptic density, BDNF/TrkB signaling, glutamatergic homeostasis, and epigenetic regulation, along with structural and functional neuroimaging changes in regions such as the prefrontal cortex (PFC), nucleus accumbens (NAc), and amygdala. Four core therapeutic domains for neuroplastic restoration were identified: neuromodulation approaches (including repetitive transcranial magnetic stimulation, transcranial direct current stimulation, and deep brain stimulation), compounds that promote neuroplasticity via neurotrophic signaling, epigenetic and anti-inflammatory interventions, and psychological therapies based on memory reconsolidation processes. These strategies demonstrated the capacity to normalize prefrontal activity, modulate reward networks, strengthen emotional regulation, and reduce craving.ConclusionDespite significant advances, important gaps remain, including methodological heterogeneity, scarcity of longitudinal studies, and limited clinical generalizability. Overall, the evidence suggests that recovery from substance use disorder requires multimodal interventions simultaneously targeting molecular, synaptic, and circuit-level plasticity, with growing emphasis on personalized approaches guided by neurobiological biomarkers.

Background/Objectives: In the modern healthcare landscape, breast cancer remains one of the most prevalent malignancies and a leading cause of mortality among women worldwide. Early and accurate prediction of breast cancer plays a pivotal role in effective diagnosis, treatment planning, and improving survival outcomes. However, due to the complexity and heterogeneity of medical data, achieving high predictive accuracy remains a significant challenge. This study proposes an intelligent hybrid system that integrates traditional machine learning (ML), deep learning (DL), and ensemble learning approaches for enhanced breast cancer prediction using the Wisconsin Breast Cancer Dataset. Methods: The proposed system employs a multistage framework comprising three main phases: (1) data preprocessing and balancing, which involves normalization using the min-max technique and application of the Synthetic Minority Over-sampling Technique (SMOTE) to mitigate class imbalance; (2) model development, where multiple ML algorithms, DL architectures, and a novel ensemble model are applied to the preprocessed data; and (3) model evaluation and validation, performed under three distinct training-testing scenarios to ensure robustness and generalizability. Model performance was assessed using six statistical evaluation metrics-accuracy, precision, recall, F1-score, specificity, and AUC-alongside graphical analyses and rigorous statistical tests to evaluate predictive consistency. Results: The findings demonstrate that the proposed ensemble model significantly outperforms individual machine learning and deep learning models in terms of predictive accuracy, stability, and reliability. A comparative analysis also reveals that the ensemble system surpasses several state-of-the-art methods reported in the literature. Conclusions: The proposed intelligent hybrid system offers a promising, multidisciplinary approach for improving diagnostic decision support in breast cancer prediction. By integrating advanced data preprocessing, machine learning, and deep learning paradigms within a unified ensemble framework, this study contributes to the broader goals of precision oncology and AI-driven healthcare, aligning with global efforts to enhance early cancer detection and personalized medical care.

Introduction: Corporate social responsibility (CSR) has always played an essential role in the market, contributing to a more competitive and efficient business environment that must be incorporated across sectors, including financial institutions. This research examined whether CSR, customer satisfaction, and corporate image are related to customer loyalty. Methods: An explanatory study was conducted with 424 Peruvians aged 18-68 (M = 32.70, SD = 10.66). Data were collected using a self-report scale of CSR, corporate image, satisfaction, and customer loyalty. The theoretical model was evaluated using the Partial Least Squares Structural Equation Modeling (PLS-SEM). Results: A measurement model with adequate fit was obtained (alpha = 0.901-0.950; CR = 0.902-0.950; AVE = 0.746-0.910). Based on the results, a positive related was demonstrated between CSR with the customer loyalty (beta = 0.264; p < 0.000; t = 4.593) and corporate image (beta = 0.235; p < 0.000; t = 4.615); between satisfaction with the customer loyalty (beta = 0.317; p < 0.000; t = 4.061) and corporate image (beta = 0.645; p < 0.000; t = 12.766), and between corporate image with the customer loyalty (beta = 0.235; p = 0.005; t = 2.810). Discussion: The study offers a valuable theoretical contribution by situating the proposed model within the integrative framework of sustainability psychology, stakeholder theory, and the Triple Bottom Line (TBL). Furthermore, it provides a significant empirical contribution by arguing that strategic investments in CSR, such as environmental finance, digitalization, and inclusive initiatives, yield tangible environmental and social benefits. These practices align with the Sustainable Development Goals (SDGs) and strengthen customer loyalty. The findings encourage decision-makers to integrate measurable sustainability actions informed by sustainability psychology to understand the individual and collective psychological processes that promote sustainable practices across organizational, social, and environmental contexts.

This study aims to compare the feedback provided by human professors and ChatGPT on university students’ work and to report on students’ perceptions of both types of feedback. A systematic review was conducted following PRISMA 2020 guidelines. Databases research included Web of Science, Scopus, EBSCO, ACM Digital Library, and IEEE Xplore, with additional gray literature sources, until October 2024. Inclusion criteria were cross-sectional studies evaluating university students’ work, comparing feedback from ChatGPT with human professors. Data extraction was performed using a standardized form, and risk of bias was assessed with the Joanna Briggs Institute critical appraisal tool. A narrative synthesis of the results was made. PROSPERO registration number: CRD42024566691. This review included 8 studies with 461 students. ChatGPT feedback was detailed and rapid, while human feedback was valued for its personalization and emotional support. Students appreciated the detailed and immediate nature of ChatGPT feedback but noted its lack of emotional nuance and context-specific guidance. Human feedback was preferred for addressing individual learning needs and providing affective support. A combination of both types of feedback to maximize benefits. ChatGPT can assist human teachers by providing detailed and timely feedback to university students. However, human supervision is essential to ensure feedback is nuanced and contextually appropriate. A hybrid approach can optimize the learning experience in higher education. Further research is necessary to explore AI applications in educational settings and understand their impact on learning outcomes.

To culturally adapt the Patient Health Questionnaire (PHQ-9) for Aymara-speaking populations in Peru and Bolivia and to evaluate its psychometric properties. A study was conducted with a non-probabilistic sample of 1,607 Aymara speakers from Peru (n = 969) and Bolivia (n = 638), aged 18 to 74 years, with both sexes represented. In the first phase, forward and backward translations were carried out, followed by expert review to ensure cultural contextualization, and focus group with Aymara speakers. In the second phase, internal structure validity, external criterion validity, group-based invariance, and reliability were assessed. The Aymara PHQ-9 demonstrated a unidimensional factor structure with satisfactory fit indices (CFI TLI > 0.95 and RMSEA SRMR < 0.08) across Aymara variants in Peru and Bolivia. Measurement invariance was established across age, sex, residential area, education level, and marital status. External validity revealed significant relationships with the GAD-7 and WHO-5. Both country-specific versions showed high reliability coefficients (Peru α = 0.847 to 0.849; Bolivia ω = 0.943 to 0.942). The Aymara PHQ-9 for Peru and Bolivia demonstrated robust internal structure validity, external validity, measurement invariance, and reliability. This version is therefore recommended for depression screening among Aymara populations.