19200 research outputs found
Sort by
Diet and mental health in school-aged children: a mini review of school-based dietary intervention studies
No full textSchool-based dietary interventions are implemented to improve health outcomes in children and adolescents, yet their impact on mental health and wellbeing remains underexplored. This mini-review synthesized findings from seventeen interventions assessing behavioral functioning and mental health symptoms in children (6–12 years) or adolescents (13–18 years). Most studies were conducted across multiple sites, enabling recruitment of large, diverse populations. More studies were conducted in children compared to adolescents. Behavioral outcomes such as hyperactivity, inattention, and oppositional behavior were commonly assessed in younger children via parent or teacher reports, while adolescent studies more frequently measured mental health symptoms, including depression and anxiety, through self-report. Supplementation, particularly in the context of nutrient deficiencies, was associated with modest improvements in behavioral functioning in children and mental health symptoms in adolescents. However, outcomes varied by the assessor (parent or teacher), and some studies showed placebo effects. In contrast, food reformulation interventions showed no significant impact on mental health outcomes. Despite the use of validated tools, methodological limitations, and variation in participants’ nutritional status limit interpretation. Overall, school-based dietary interventions show potential to improve mental health by reaching large, diverse populations. Further research is needed using standardized, age-appropriate measures and incorporating assessment of nutritional status to understand how diet can support and improve mental health in children and adolescents
ESA-YOLOv5m: a lightweight spatial and improved attention-driven detection for brain tumor MRI analysis
Get PDFIntroduction: The early and accurate detection of brain tumors is vital for improving patient outcomes, enabling timely clinical interventions, and reducing diagnostic uncertainty. Despite advances in deep learning, conventional Convolutional Neural Network (CNN)-based models often struggle with small or low-contrast tumors. They also remain computationally demanding for real-time clinical deployment. Methods: This study presents an Enhanced Spatial Attention (ESA)-integrated You Only Look Once v5 medium (YOLOv5m) architecture, a lightweight and efficient framework for brain tumor detection in MRI scans. The ESA module, positioned after the Spatial Pyramid Pooling-Fast (SPPF) layer, enhances feature discrimination by emphasizing diagnostically relevant regions while suppressing background noise, thereby improving localization accuracy without increasing computational complexity. Experiments were conducted on the Figshare brain tumor MRI dataset containing three tumor classes: glioma, meningioma, and pituitary. Results: ESA-YOLOv5m achieved a Precision of 90%, Recall of 90%, and mean Average Precision (mAP)@0.5 of 91%, surpassing the baseline YOLOv5m by approximately 11%–12%. An ablation study further confirmed that placing the ESA module after the SPPF layer yields the highest performance ([email protected] = 0.91), while earlier integration produced marginally lower results. Classwise analyses demonstrated consistent gains (mAP range 0.87–0.98), and fivefold cross-validation showed stable performance ([email protected] = 0.910 ± 0.006). Efficiency tests revealed negligible overhead, with less than a 4.3% increase in parameters and an average latency below 10 ms per image. Discussion: Overall, the results validate that integrating a lightweight spatial attention mechanism significantly enhances tumor localization and model generalization while preserving real-time inference. The proposed ESA-YOLOv5m framework provides a reliable and scalable solution for automated brain tumor detection, suitable for clinical decision-support systems and edge healthcare applications
Prediction Optimization for Type 2 Diabetes Mellitus Using Artificial Neural Networks
No full textHigh level of blood glucose is a key indicator of diabetes mellitus, with 90% of the worldwide cases attributed to Type 2 diabetes. This illness requires constant attention, putting a huge monetary toll on patients and their families. Research has shown a worldwide presence of diabetes, which has affected 10.5% of the adult population in the age group of 20–79 years. Projections suggest a stunning raise to 783 million cases by 2045. Diabetes increases the risk and complications of other diseases and may even cause untimely death. Existing literature has shown the implemented machine learning and deep learning approach on traditional datasets. Local and current datasets based on the Indian population are unavailable. The aim of this research is to create an optimized model to predicts the occurrence or non-occurrence of diabetes mellitus, on the Pima Indian diabetes (PID) dataset. The proposed Artificial Neural Network (ANN) model employs several hidden layers in an ANN with RMSprop and Adam optimizers. This study focuses on training the model using epochs, and studying its behavior on the performance of the model. Three hidden layers in an ANN showed a maximum accuracy of 84.42%, using the Adam optimizer. This study demonstrated that, in some scenarios, larger epochs did not reduce the validation loss, and hence did not improve the model’s performance. The model was validated through various performance metrics. Further studies would include evaluating the accuracy of predictions with a higher number of hidden layers. Our research would augment the continual worldwide effort to fight diabetes mellitus, by advancing the computational models for optimized prediction of the disease
Comparative Effectiveness of Human- and Robot-Based Interventions in Increasing Empathy Among Autistic Children
Get PDFQuality, determinants and financial consequences of climate change disclosures: a structured review
No full textRecent years have seen rapid growth in research on climate-related disclosures, yet existing reviews mainly focus on carbon and GHG reporting through the Carbon Disclosure Project (CDP), overlooking the broader framework introduced by the Task Force on Climate-related Financial Disclosures (TCFD). This study conducts a structured review of 134 peer-reviewed articles (2010–2025), all published in journals ranked 2* or higher according to the ABS 2024 Journal Guide, to integrate insights from both CDP and TCFD perspectives. The review makes three key contributions over past studies. First, it provides a structured and integrative synthesis to date by bridging CDP and TCFD based disclosure research, while prior reviews largely examined carbon disclosure in isolation. Second, it develops a thematic framework encompassing disclosure characteristics, quality and compliance, determinants, financial consequences and greenwashing, dimensions that earlier reviews treated separately. Finally, it advances the methodological rigor of prior reviews on climate related disclosures by applying a structured review and by identifying underexplored empirical gaps to guide future research. The findings offer actionable insights for regulators, investors and firms aiming to strengthen the quality, comparability and credibility of climate-related disclosures
Inner Speech Decoding: A Comprehensive Review
No full textInner speech decoding is the process of identifying silently generated speech from neural signals. In recent years, this candidate technology has gained momentum as a possible way to support communication in severely impaired populations. Specifically, this approach promises hope for people with a variety of physical or neurological disabilities who need alternative means of verbal expression. This review covers recording modalities that range from the noninvasive EEG to the high‐density electrocorticography and discusses how linear discriminant analysis, deep convolutional networks, and hybrid fusion of EEG with fMRI are integrated into machine learning strategies to infer covert speech. This review synthesizes evidence to suggest that small vocabularies, under controlled conditions, can yield relatively reasonable accuracy while further refining the decoding outcome via context‐based approaches. The impact of sensor quality, training data size, and domain adaptation is illustrated by focusing on public datasets of imagined or articulated speech. Throughout the article, the methodological standards emerging across laboratories will be discussed, emphasizing that effective inner speech recognition involves high‐quality preprocessing, subject calibration, and informed modeling choices balanced against computational power for interpretability. In addition to technical advancements, this review also examines the ethical, societal, and regulatory challenges surrounding inner speech decoding, including brain data privacy, neural rights, informed consent, and user trust. Addressing these interdisciplinary issues is critical for the responsible development and real‐world adoption of such technologies. This article is categorized under: Neuroscience > Computation Computer Science and Robotics > Machine Learnin