8383 research outputs found
Sort by
Prevalence and Reporting Rates of Extraspinal Findings for Lumbar Spine Magnetic Resonance Imaging in a Ghanaian Tertiary Hospital
Get PDFBackground
Extraspinal findings are commonly detected on magnetic resonance imaging of the lumbar spine, but these findings are sometimes omitted from radiological reports. Failing to report these findings could have a clinical impact on the patients. The purpose of this study was to determine the prevalence and reporting rates of extraspinal findings on lumbar spine magnetic resonance imaging.
Methods
Retrospective analysis was done on lumbar spine magnetic resonance images done at the Korle-Bu Teaching Hospital between January 2020 and December 2021. A total of 1267 patients underwent lumbar spine magnetic resonance imaging within the period. The degree of clinical significance of the extraspinal findings was ascertained using the computed tomography colonography reporting and data system classification scheme. The reporting rate was determined by referring to the archived radiological reports. Statistical analysis was done using IBM SPSS Statistics for Windows, Version 25 (Released 2017; IBM Corp., Armonk, New York, United States).
Results
A total of 737 extraspinal findings were detected from 530 patients. The overall reporting rate of extraspinal findings was 62.6% (461/737). The most common extraspinal finding was a simple renal epithelial cyst (n = 333). Clinically significant findings were detected in 107 out of the 530 patients; 36.4% of the clinically significant findings were not reported when compared with the archived reports.
Conclusion
Extraspinal findings on lumbar spine imaging were common in our study population. When radiologists are reporting lumbar spine magnetic resonance imaging, it is crucial to be aware of the risk of missing clinically significant findings
Efficacy of the Best Possible Self intervention for generalised anxiety: exploration of mediators and moderators
Get PDFGeneralised anxiety is increasingly prevalent, yet access to therapeutic interventions remains limited. We present two randomised control trials aimed to investigate the efficacy of the Best Possible Self (BPS) technique as an intervention for reducing anxiety in a non-clinical sample. The BPS was delivered online using survey software, and changes in anxiety were assessed over two weeks. Across both studies, the BPS significantly reduced anxiety, as measured with the Generalised Anxiety Disorder-7 Questionnaire (GAD-7). Evidence was found for the potential mediating role of self-esteem, and analysis of intervention frequency demonstrated that completing two or more sessions of the BPS intervention led to significant reductions in anxiety. Participants who completed only one session reported no significant change in symptoms. Evidence was not found for a moderating role of imagery capacity. These findings suggest that the BPS technique could be an accessible, cost-effective intervention for reducing generalised anxiety
Impact of Tropical Cyclone on Coastal Phytoplankton Blooms and Underlying Mechanisms
Get PDFThis study examines the impact of tropical cyclone (TC) "Wipha" (2019) on phytoplankton chlorophyll-a (Chl-a) dynamics, using observations from two buoy stations (S1 and S2). Results indicate that persistently high turbidity at the inner bay (station S1) restricted underwater light availability, resulting in an insignificant change in mean daily Chl-a concentrations, despite sufficient nutrients. Conversely, at the outer bay (station S2), Chl-a significantly increased after the storm, exhibiting notable delayed correlations with elevated turbidity (r = 0.87, p < 0.01) and aerosol deposition (r = 0.90, p < 0.01). The differential phenomenon at two locations highlights that distinct environmental control the responses of phytoplankton dynamics to the tropical cyclone, primarily related to light availability and nutrient sources.
New Hydrological Insights for the Region:
In contrast to prior studies, the nutrient source leading to increased Chl-a at the outer bay may result from wet deposition of aerosols and re-suspension of suspended matter, rather than direct terrestrial nutrient inputs. Additionally, the prolonged turbidity recovery period (up to 5 days) at the inner bay substantially limited phytoplankton growth, highlighting TC-induced turbidity as a critical factor constraining phytoplankton blooms in eutrophic coastal environments.
Keywords: tropical cyclone; buoy observation; turbidity; lagging correlatio
Emerging technologies and innovative approaches to combat antimicrobial resistance: A narrative review of next-generation therapeutic strategies
Get PDFAntimicrobial resistance (AMR) is one of the most pressing global health challenges, with approximately 700,000 deaths annually directly attributable to resistant bacterial infections. This alarming trend threatens to undermine decades of medical progress. The widespread misuse and overuse of antibiotics have accelerated the emergence of multidrug-resistant (MDR) pathogens, leading to increased morbidity, mortality, and healthcare costs. This review examines the intricate mechanisms underlying the development of AMR and discusses innovative next-generation therapeutic strategies and emerging approaches for combating resistant pathogens. CRISPR-based antimicrobials demonstrated over 90 % in vitro efficacy in selectively eliminating MDR pathogens. Nanotechnology-based solutions, such as those utilizing silver and gold nanoparticles, have demonstrated potent bactericidal activity in preclinical settings; however, toxicity and regulatory concerns persist. Bacteriophage therapy and antimicrobial peptides (AMPs) are advancing through early clinical trials, offering targeted activity and immune-modulating effects. Artificial intelligence (AI)-driven drug discovery has already been clinically integrated, accelerating the design of antibiotics and predicting resistance with high efficiency. Comparative analysis reveals that AI tools possess the highest readiness level, while CRISPR and AMPs are promising but remain in early development stages. These emerging strategies collectively present significant potential to complement or replace conventional antibiotics in addressing AMR. Despite their potential, these technologies face significant implementation challenges, including technical limitations, economic barriers, ethical considerations, and regulatory complexities. This review emphasizes the critical need for multidisciplinary collaboration, sustainable funding models, and global policy frameworks to effectively translate these innovations into clinical practice. The AMR crisis can only be addressed through international collaboration, combining scientific innovation and supportive policy environments
Explainable machine learning models for early Alzheimer's disease detection using multimodal clinical data.
Get PDFAlzheimer's disease (AD) represents a significant global health challenge requiring early and accurate prediction for effective intervention. While machine learning models demonstrate promising capabilities in AD prediction, their black-box nature limits clinical adoption due to a lack of interpretability and transparency. This study aims to develop and evaluate explainable artificial intelligence (XAI) frameworks for AD prediction using comprehensive multimodal patient data, with a focus on enhancing model interpretability through SHAP and LIME techniques. A comprehensive dataset of 2,149 patients aged 60-90 years was obtained from Kaggle, encompassing demographic, medical history, lifestyle, clinical measurements, cognitive assessments, and symptom data. Rigorous preprocessing included MinMax normalisation, Synthetic Minority Over-sampling Technique (SMOTE) for class imbalance, and Backward Elimination Feature Selection reduced 32 features to 26 optimal predictors. Six machine learning models were evaluated: K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Logistic Regression (LR), XGBoost, Stacked Ensemble, and Random Forest (RF). RF's optimal hyperparameters were obtained using Ant colony Optimization Model interpretability was enhanced using SHAP and LIME frameworks for both global and local explanations. The optimised Random Forest with backward elimination feature selection and ant colony optimisation achieved superior performance with 95 % accuracy, 95 % precision, 94 % recall, 94 % F1-score, and 98 % AUC. SHAP analysis identified functional assessment, activities of daily living (ADL), memory complaints, and Mini-Mental State Examination (MMSE) as the most influential predictors. LIME provided complementary local explanations, validating the clinical relevance of identified features. The integration of explainable AI techniques with machine learning models provides clinically meaningful insights for AD prediction, enhancing transparency and fostering trust in AI-driven diagnostic tools whilst maintaining high predictive accuracy. Future work should focus on external validation, clinical workflow integration, and addressing computational requirements for real-world deployment. [Abstract copyright: Copyright © 2025 The Author(s). Published by Elsevier B.V. All rights reserved.
Examining the effect of AI advertising involvement disclosure on advertising value and purchase intentions
No full textPurpose
The integration of artificial intelligence (AI) into advertising has revolutionized how brands create and deliver marketing messages. The involvement of AI in ad creation introduces a critical question: if the AI origin of an advertisement is disclosed, how this transparency affects advertising value and purchase intentions. Grounded on the Persuasion Knowledge Model, this study investigates the effect of AI disclosure on these two key outcomes.
Design/methodology/approach
ChatGPT and Stable Diffusion were employed to generate stimuli. 358 consumers recruited via Prolific were exposed to the stimuli. The data were analyzed using a combination of Hayes’ Process models, MANCOVA and ANCOVA.
Findings
The results reveal that AI disclosure diminishes advertising value and purchase intentions. The relationships are negatively mediated by advertising credibility and positively moderated by consumer attitudes towards AI.
Originality/value
Theoretically, the research contributes to a more comprehensive picture of AI-generated advertising evaluation. Practically, the research offers actionable insights for businesses seeking to balance the advantages of AI with human psychology, ultimately optimizing advertising effectiveness in an increasingly AI-driven marketplace
Community Food Insecurity Interventions for Adults Living in the United Kingdom: A Scoping Review
Get PDFFood insecurity is a growing concern worldwide, particularly in the United Kingdom. Despite this, community‐based interventions to address food insecurity remain an under‐researched area. Existing food insecurity reviews have focused on international evidence, limiting investigations to foodbank use and/or interventions targeted towards children. This scoping review aimed to understand the evidence on available community‐based interventions for adults experiencing food insecurity in the United Kingdom and the suggested elements for a feasible, acceptable intervention. A comprehensive electronic search was completed up to January 2024. All study designs were considered. A descriptive analytical approach was used to summarise intervention data. Narrative synthesis explored the data further, using the Food Ladders model as a framework. This review identified a very limited scope and quantity of evidence on community food insecurity interventions for UK adults, with 21 included studies. Over half of interventions (52.4%, n = 11) relied on volunteers, and a high proportion used donated or surplus food. The nutritional quality of emergency food provision was poor, and it was unclear whether providers could adequately cater for special dietary requirements, cultural and/or religious needs. There were very few studies (19.0%, n = 4) assessing the feasibility or acceptability of interventions or their impact on food insecurity. Further research is required into the feasibility, acceptability and effectiveness of community food insecurity interventions for adults in the United Kingdom
Unveiling the ChatGPT Educational Revolution: Assessing the Dynamic Impact on Students and Educators
Get PDFDue to shifting social demands and technology breakthroughs, the higher education environment is changing quickly. Despite initiatives to make education accessible to everyone, accessibility is still a major problem, especially in light of the digital divide. This study investigates how ChatGPT, an AI-powered chatbot, can revolutionize higher education by tackling important problems including resource efficiency, personalized learning, and accessibility. This study intends to improve learning outcomes for both students and educators by comprehending how instructors and students incorporate ChatGPT into instructional methods. Students and instructors were given both quantitative and qualitative questionnaires as part of a mixed-methods approach, in order to gather data on the usage of ChatGPT for different academic tasks, such as lesson preparation, grading, and student help. Results showed that most people believe ChatGPT to be a useful tool that improves productivity, saves time, and helps with grasping difficult subjects. Questions were raised concerning the veracity of the data that ChatGPT offered and the necessity of organized training. ChatGPT and other AI technologies have the potential to enhance educational results by enabling personalized instruction and offering on-demand learning help. In addition to the continuing discussion on the use of cutting-edge technology in higher education, the findings provide insightful information for educational institutions seeking to use AI to improve teaching and learning. Received: 16 September 2024 | Revised: 9 April 2025 | Accepted: 15 July 2025 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement Data are available from the corresponding author upon reasonable request. Author Contribution Statement Swathi Ganesan: Conceptualization, Methodology, Data curation, Writing – Original draft, Writing – review and editing, Visualization, Project administration. Lakmali Karunarathne: Conceptualization, Methodology, Formal analysis, Data curation, Writing – original draft, Writing – review and editing, Visualization, Project administration. Ghanshyam Mahota: Methodology, Formal analysis, Writing – original draft. Sangita Pokhrel: Conceptualization, Writing – review and editing, Visualization