26199 research outputs found
Sort by
A Deep Learning Framework Based on Novel Hierarchical-LSTM Model for Enhanced Machinery Prognostics
Get PDFMachinery prognostics has garnered increasing research attention due to its critical role in industries such as manufacturing and renewable energy. Data-driven techniques, particularly recurrent neural networks (RNNs) and convolutional neural networks (CNNs), have shown promise in accurately extracting features for estimating the remaining useful life (RUL) of machinery. However, the non-stationary and non-linear nature of machinery signals poses significant challenges to achieving accurate prognostics. This study introduces a novel hierarchical recurrent neural network method called hierarchical long short-term memory (H-LSTM) that is based on the long short-term memory (LSTM) model. H-LSTM is meant to address the problems with traditional RNNs that only use the previous time step for sequential data learning. It incorporates a hierarchical structure, enabling influence from multiple preceding time steps at each current step. Experimental evaluation on the FEMTO benchmark bearing dataset under varying operational conditions demonstrates that the proposed H-LSTM approach achieves up to fourfold improvements in performance compared to state-of-the-art methods, particularly for low signal-to-noise ratio (SNR) signals
The Talent Project and the validation of standards for the identification of student-athlete talent
No full textThe aim of the TALENT project is to promote equality in education, prevent exclusion, support dual careers (sport and school), create new role models for the benefit of young talents and prepare them for lifelong learning and professional sport from an early age. It is promoted by a European consortium of 7 partner institutions and runs from December 2022 to May 2025. It consists of five work packages.In the first work phase, developing the WP2 (from December 2022 to October 2023), under the coordination of UNIPA, NIS University, KMOP and EAS standards for talent recognition were identified and validated. Initially, 12 focus groups were conducted with teachers (77 teachers) and coaches (73 coaches) on creating talent identification standards; subsequently, workshops were held with dual career experts to validate these standards. This was a key piece of work that enabled the establishment of clear guidelines and protocols to identify and support talented young people in their dual careers.A final list of 41 shared statements was identified: 20 related to teachers and 21 related to coaches. For example, teachers emphasized the need for multidisciplinary approaches and early identification of talent, while coaches underlined the importance of psychological readiness and collaboration with schools and families. These statements not only provide structured reference points for talent identification but also highlight actionable needs across educational and sport systems. As such, they represent a solid foundation for developing standard operating procedures in talent recognition and dual career support
The use of horizontal force-velocity profile in soccer: a rapid systematic review
Get PDFBackground: The ability to accelerate and reach high sprinting velocities is crucial to soccer performance. In this context, the horizontal force-velocity profile (H-FVP) has emerged as a tool to evaluate neuromuscular capabilities relevant to sprinting. This rapid review aims to critically describe the application of H-FVP in soccer and summarize the characteristics of the methodologies employed in its measurement and calculation. Methods: A rapid systematic review was conducted in accordance with the Cochrane Rapid Reviews Guidance and PRISMA guidelines. A search on MEDLINE (via PubMed), SPORTDiscus (via EBSCOhost), and Web of Science databases was conducted in February 2025. Studies were considered eligible if they assessed the H-FVP in soccer players of any competitive level and both sexes. Results: Fourteen studies met the inclusion criteria, analysing a total of 1320 soccer players across different competitive levels. Most studies explored the relationship between H-FVP parameters and sprint or change of direction performance. Additional studies addressed variations according to playing position differences, biological maturation, fatigue responses, or injury profile. The predominant testing protocols involved linear sprints ranging from 30 to 40 m, often with split-distance measurements. The Samozino method was consistently used for H-FVP computation. Commonly reported parameters included theoretical maximal force (F0), velocity (V0), and power (Pmax), with some studies also including the ratio of force (RF) and its decrease with speed (DRF). Radar devices, photocell systems and mobile applications were the primary measurement tools utilized. Conclusion: This systematic review highlights the potential of the H-FVP as an approach to be used to improve sprint performance in soccer players across competitive levels. However, methodological inconsistencies among studies highlight the need for standardized testing protocols to improve their practical application. Identified gaps in the literature point out the necessity for further investigation in future research
ABA‐Feed Infant Feeding Training for Peer Supportersand Coordinators: Development and Mixed‐Methods Evaluation
No full textThe Assets-based feeding help Before and After birth (ABA-feed) intervention aims to improve breastfeeding rates by offering proactive peer support to first-time mothers, regardless of feeding intention. Based on behaviour change theory and an assets-based approach, the intervention involved training existing peer supporters to become Infant Feeding Helpers (IFHs). A train-the-trainer model was used, with coordinators delivering four two-hour training sessions to IFHs. Training covered a study overview, IFH role, role-play scenarios, and signposting to local assets. Due to COVID-19, training was delivered online.
Post-training questionnaires were completed by 22/30 (73.3%) coordinators and 119/193 (61.7%) IFHs, and qualitative interviews were conducted with 24 coordinators and 72 IFHs. Researchers observed training at five sites, assessing fidelity, engagement, and delivery quality. Questionnaire data were analysed descriptively, and qualitative data using Framework Analysis.
Findings indicated that coordinators valued the train-the-trainer model, particularly information on formula feeding and antenatal discussions. IFHs found training engaging and felt prepared, though some were apprehensive about formula feeding support. While online training was convenient, challenges included monitoring discussions and role-play in breakout rooms. Most participants favoured a hybrid approach, with in-person sessions for interactive activities. Observations showed high training fidelity, participant engagement, and confidence in delivering intervention components.
The ABA-feed training was acceptable to coordinators and IFHs and was delivered with fidelity. Future training should adopt a hybrid approach, incorporating diverse resources and prioritising in-person interactive components such as role-play
MODELLING IMMUNE CELL INTERACTIONS WITH AN IN VITRO BLOOD BRAIN BARRIER-GLIOMA MODEL
No full textAIMS The presence of the blood brain barrier (BBB) may reduce disposition of novel therapies to the brain. Cell-cell interactions in the glioma microenvironment also influence tumour progression. There is a need for an in vitro BBB-glioma model to mimic the complex tumour microenvironment (TME) for prediction of patient outcomes. A dynamic BBB model, that incorporates an immune microenvironment and glioblastoma spheroids, will be used to test delivery and efficacy of novel therapies. METHODS The BBB model consisted of primary derived brain endothelial cells (HBMEC), astrocytes (HA) and pericytes (HBVP) grown on 2D transwell or 3D inserts. Glioblastoma spheroids were grown from patient-derived cells. An inflammatory immune microenvironment was created using conditioned media from microglia (BV2, HMC3 and IM) activated with LPS. RESULTS Confirmation of the presence of a physical cell barrier was measured by TEER, exclusion of varying molecular weights of FitC-dextran and the expression of tight junction proteins ZO-1 and Claudin-1. Initial validation studies illustrated a stronger BBB was created with co-cultures compared to an endothelial monolayer. The presence of an inflammatory immune microenvironment was confirmed by measuring an increased release of IL-6, IL-1β, and TNF-α from microglial cells stimulated with LPS compared to no LPS control. When the condition media from activated microglial cells was added to the BBB, there was a further increased detection of IL-6, IL-1β, and TNF-α. CONCLUSION This study confirmed that all BBB and microglia cells could produce a measurable inflammatory environment upon activation. The next stage is to introduce patient derived glioma spheroids to the BBB-microglia model to measure how the tumour creates an inflammatory microenvironment and how the inflammatory markers IL-6, IL-1β, and TNF-α respond. Ultimately this will provide a more realistic model of the TME so glioma biology and drug response can be determined
From Transaction to Transformation: AI and Machine Learning in FinTech
No full textArtificial Intelligence technology and Machine Learning operate in FinTech industries by automating processes while giving complete risk assessments, plus private customized solutions. This research studies how AI and ML technology support modern FinTech business functions, specifically fraud prevention, business credit rating, automated investment advice, and automatic market trades. The research conducts the empirical evaluation of Large Language Model integration within Zero Trust security structures by using adversarial testing together with case studies and explainability assessment methods. Research indicates that LLMs improve threat detection performance by 21% and traditional systems yet they fail in 38% of cases when prompt injection happens. The study highlights the need for adversarial training combined with fairness auditing and regulatory oversight in order to establish AI deployment safety in cybersecurity. The research pairs contemporary studies with specific examples to uncover AI and ML's main advantages of better results, less manual work, and better services for customers. This study researches both data protection risks and official rules while looking at the weaknesses of programmed systems. Through research the paper determines how applications based on AI machine learning propel financial services toward automated systems that adjust their behavior for optimal results. The analysis generates new ideas to develop FinTech innovation models and provides useful directions for business creators plus oversight agencies plus financial establishments. This study brings a new method to view AI/ML integration while showing the need to align ethical and legal rules in future FinTech systems
Multi‐objective optimization of green concrete incorporating recycled plastic and sawdust waste as fine aggregates using response surface methodology
No full textThis study investigates the development of Green Normal Concrete (GNC) by incorporating plastic waste aggregates (PWAs) and sawdust waste (SDW) as partial replacements for natural sand (NS). The research follows three experimental stages using Response Surface Methodology (RSM) and the Absolute Volume (AV) method to optimize compressive strength, workability, and density. Stage I establishes a normal concrete (NC) control mix based on British standards. Stage II examines the effects of replacing NS with gray and black ABS plastic granules (GABSR and BABSI), identifying the optimal mix (OPW) with 0.04% NS-GABSR and 24.11% NS-BABSI, achieving 20.50 MPa compressive strength, 25 mm workability, and 2276 kg/m3 density. Stage III introduces sawdust (SDF) as a fine aggregate replacement in OPW, with silica fume (DSF) and superplasticizer (SP) enhancing performance. The optimal mix (OPWSD), with 60.32% BABSI-SDF and 20.01% NS-SDF, achieves 25.10 MPa compressive strength, 124.21 mm workability, and 2055 kg/m3 density. The results highlight the feasibility of PWAs and SDW in concrete, emphasizing the need for further studies on long-term durability and environmental impact
Investigating climate change through argumentation: Purposeful questioning supports argumentation and knowledge acquisition.
Get PDFOver several weeks, 125 young adolescents engaged deeply with the topic of climate change in a discourse-based program designed to build argumentation skills. We put to a test the hypothesis that information on this complex and critical topic is best acquired and made use of in argument if acquiring it is experienced as having purpose and able to fulfill a role in argument. Activities in an experimental condition followed the program’s practice of making available topic-related information in the form of brief questions and answers on an as-requested basis. Offered to them as a potential resource in peer dialogs on the topic, throughout the activity participants selected questions they wished answers to, and these were provided. Students in a comparison condition followed the traditional classroom practice of being assigned to read an introductory text as background information on the topic. It contained information identical to that in the questions and answer cards experimental group participants chose to access. Under both conditions, all information remained available once accessed. Both groups benefited in knowledge gain, as well as skill development in coordinating evidence with claims in final essays. However, the experimental group showed greater knowledge as well as skill gain, and a difference we suggest is attributable to the knowledge gained having an anticipated purpose making them more likely to make use of it
Student nurse burnout
No full textIn this editorial, the author discusses the increasing levels of burnout in mental health nursing students and the importance of taking urgent measures to prevent further strain on the NHS workforce
Personality disorders: A preface
No full textPersonality disorders (PDs) are complex psychiatric conditions marked by persistent maladaptive patterns of cognition, emotion, and behaviour. This Special Issue of Neuropharmacology compiles cutting-edge research and reviews exploring the neurobiological, psychological, and therapeutic dimensions of PDs. Contributions span diverse topics, including neurobiological mechanisms, hypothesis for novel treatments, gene-environment interactions, and immune-metabolic influences. Key findings include the role of fatty acid amide hydrolase (FAAH) in aggression in borderline personality disorder (BPD), excitation/inhibition imbalances in antisocial personality disorder (ASPD), and the potential of psychedelics and gut-microbiota modulation as therapeutic interventions. Additionally, research highlights the interplay between genetic vulnerabilities (e.g., genes affecting serotonergic/dopaminergic pathways) and environmental factors (e.g., childhood trauma) in shaping PD phenotypes. This collection underscores the need for precision medicine approaches, integrating biomarkers and targeted therapies to improve clinical outcomes. By bridging preclinical and clinical insights, this Special Issue advances our understanding of PDs and paves the way for innovative treatments