Irish Universities
Not a member yet
78452 research outputs found
Sort by
Abdominal pain in children and adolescents with Autism Spectrum Disorder: a systematic review
No full textThe aim of this study was to review the existing literature on abdominal pain in children and adolescents with autism spectrum
disorder. Systematic search of four databases (PsycINFO, ERIC, PubMed, MEDLINE) identified 13 studies that met the
inclusion criteria. Articles were analyzed for common themes, including the prevalence of abdominal pain and gastrointestinal
(GI) symptoms, associations between abdominal pain/GI symptoms and behavioral and emotional concerns, associations between
abdominal pain/GI symptoms, and other comorbid disorders and treatment options based on gut bacteria, diet, and
probiotics. Reasons for varying prevalence rates, persistence of symptoms over time, comorbidities, and different treatment
options are discussed. Clinical implications and recommendations for future research are also discussed.Open Access funding provided by the IReL Consortiu
Extracting knowledge from deep neural networks through graph analysis
No full textThe popularity of deep learning has increased tremendously in recent years due to its ability to efficiently solve complex tasks in challenging areas such as computer vision and language processing. Despite this success, low-level neural activity reproduced by Deep Neural Networks (DNNs) generates extremely rich representations of the data. These representations are difficult to characterise and cannot be directly used to understand the decision process. In this paper we build upon our exploratory work where we introduced the concept of a co-activation graph and investigated the potential of graph analysis for explaining deep representations. The co-activation graph encodes statistical correlations between neurons’ activation values and therefore helps to characterise the relationship between pairs of neurons in the hidden layers and output classes. To confirm the validity of our findings, our experimental evaluation is extended to consider datasets and models with different levels of complexity. For each of the considered datasets we explore the co-activation graph and use graph analysis to detect similar classes, find central nodes and use graph visualisation to better interpret the outcomes of the analysis. Our results show that graph analysis can reveal important insights into how DNNs work and enable partial explainability of deep learning models
Ultrathin Silicon Membranes for in Situ Optical Analysis of Nanoparticle Translocation across a Human Blood-Brain Barrier Model
No full textHere we present a blood-brain barrier (BBB) model that enables high-resolution imaging of nanoparticle (NP) interactions with endothelial cells and the capture of rare NP translocation events. The enabling technology is an ultrathin silicon nitride (SiN) membrane (0.5 μm pore size, 20% porosity, 400 nm thickness) integrated into a dual-chamber platform that facilitates imaging at low working distances (∼50 μm). The platform, the μSiM-BBB (microfluidic silicon membrane-BBB), features human brain endothelial cells and primary astrocytes grown on opposite sides of the membrane. The human brain endothelial cells form tight junctions on the ultrathin membranes and exhibit a significantly higher resistance to FITC-dextran diffusion than commercial membranes. The enhanced optical properties of the SiN membrane allow high-resolution live-cell imaging of three types of NPs, namely, 40 nm PS-COOH, 100 nm PS-COOH, and apolipoprotein E-conjugated 100 nm SiO2, interacting with the BBB. Despite the excellent barrier properties of the endothelial layer, we are able to document rare NP translocation events of NPs localized to lysosomal compartments of astrocytes on the "brain side" of the device. Although the translocation is always low, our data suggest that size and targeting ligand are important parameters for NP translocation across the BBB. As a platform that enables the detection of rare transmission across tight BBB layers, the μSiM-BBB is an important tool for the design of nanoparticle-based delivery of drugs to the central nervous system.European Commission - Seventh Framework Programme (FP7)Science Foundation Ireland -- replaceEuroNanoMed IIINational Natural Science Foundation Of ChinaSFI-NSFC Partnership ProgrammeNIHLundbeck Foundation2021-05-11 JG: PDF replaced with correct versio
Characterising Concrete Mixes for 3D Printing
No full textThe 2nd RILEM International Conference on Concrete and Digital Fabrication, Eindhoven, The Netherlands (held online due to Coronavirus outbreak), 6-8 July 2020The construction industry is currently experiencing significant change as building information modeling (BIM), digital design and construction automation are exerting intense pressure on traditional technologies. As an advanced manufacturing technology, three-dimensional (3D) printing has significant potential applications in the construction sector, by utilizing a programmable robotic arm with a nozzle jet, 3D printing can enable us to construct complex concrete structures layer by layer. This new construction technique offers an advanced approach that can potentially accelerate the construction time and improve efficiency. They can work 24/7, even in a hazardous environment while minimising human errors. However, as an advanced cutting-edge technology, there are several remaining challenges to be overcome in comparison to traditional concrete casting, and these include appropriate pumpability, extrudability, buildability, compressive strength and open time for printing concrete. To overcome these challenges this project will investigate the effect of nanoclay to improve the fresh properties of printing concrete. It is considered that by utilizing different amounts of nanoclay and superplasticiser in the mix it will be possible to significantly affect the fresh properties of 3D printed concrete. Printable materials, like any other cementitious materials, flow only when submitted to stresses higher than a critical yield stress. In this project, a shear vane test was used to measure the yield stress of cement-based materials in the lab.University College Dubli
Nickel catalysts for C1 reactions: recollections from a career in heterogeneous catalysis
No full textThis paper presents a brief review of some research projects carried out in the author’s laboratories over a number of years. The work reported concerns the use of nickel containing catalysts for a range of C1 reactions: the steam reforming of methane,
the methanation of CO, the oxidative coupling of methane and the dry reforming of methane. A number of novel catalysts have been developed in the course of this work, mostly in collaborative projects with industrial organisations, and some of the background to this work is discussed. The paper emphasises the importance in work of the type described of the establishment of contacts between the academic laboratory and industrial researchers such as Mike Spence
Classification and biological identity of complex nano shapes
No full textEverywhere in our surroundings we increasingly come in contact with nanostructures that have distinctive complex shape features on a scale comparable to the particle itself. Such shape ensembles can be made by modern nano-synthetic methods and many industrial processes. With the ever growing universe of nanoscale shapes, names such as “nanoflowers” and “nanostars” no longer precisely describe or characterise the distinct nature of the particles. Here we capture and digitise particle shape information on the relevant size scale and create a condensed representation in which the essential shape features can be captured, recognized and correlated. We find the natural emergence of intrinsic shape groups as well-defined ensemble distributions and show how these may be analyzed and interpreted to reveal novel aspects of our nanoscale shape environment. We show how these ideas may be applied to the interaction between the nanoscale-shape and the living universe and provide a conceptual framework for the study of nanoscale shape biological recognition and identity.Irish Research CouncilScience Foundation Ireland -- replaceGuangdong Provincial Education Department Key Laboratory of Nano-Immunoregulation Tumor
MicroenvironmentChinese Scholarship CouncilCNP
Traffic Aware Resource Allocation Schemes for Multi-Cell MIMO-OFDM Systems
No full textWe consider a downlink multi-cell multiple-input multiple-output (MIMO) interference broadcast channel (IBC) using orthogonal frequency division multiplexing (OFDM) with multiple users contending for space-frequency resources in a given scheduling instant. The problem is to design precoders efficiently to minimize the number of backlogged packets queuing in the coordinating base stations (BSs). Conventionally, the queue weighted sum rate maximization (Q-WSRM) formulation with the number of backlogged packets as the corresponding weights is used to design the precoders. In contrast, we propose joint space-frequency resource allocation (JSFRA) formulation, in which the precoders are designed jointly across the space-frequency resources for all users by minimizing the total number of backlogged packets in each transmission instant, thereby performing user scheduling implicitly. Since the problem is nonconvex, we use the combination of successive convex approximation (SCA) and alternating optimization (AO) to handle nonconvex constraints in the JSFRA formulation. In the first method, we approximate the signal-to-interference-plus-noise ratio (SINR) by convex relaxations, while in the second approach, the equivalence between the SINR and the mean squared error (MSE) is exploited. We then discuss the distributed approaches for the centralized algorithms using primal decomposition and alternating directions method of multipliers. Finally, we propose a more practical iterative precoder design by solving the Karush-Kuhn-Tucker expressions for the MSE reformulation that requires minimal information exchange for each update. Numerical results are used to compare the proposed algorithms to the existing solutions.European Commission - European Regional Development FundScience Foundation IrelandFinnish Funding Agency for Innovation (Tekes)Nokia NetworksXilinxElektrobitRiitta ja Jorma Takanen FoundationAcademy of Finlan
Fertiliser characteristics of stored spent mushroom substrate as a sustainable source of nutrients and organic matter for tillage, grassland and agricultural soils
No full textSpent mushroom substrate (SMS) is an organic manure that can be used with advantage in agriculture. Under European Union (EU) (Good Agricultural Practice for Protection of Waters) Regulations, SMS cannot be applied to land over the winter months and must be stored on concrete surfaces, either covered or uncovered, to prevent nutrient-rich runoff seeping into groundwater. Spent mushroom substrate at four storage facilities, two covered and two uncovered, was analysed for physical and chemical characteristics after storage for up to 12 mo. Significant differences (P<0.05) were identified for all parameters across the four sites, except for pH, but there were no consistent differences that correlated with uncovered or covered storage conditions. The content of nitrogen (N) and manganese (Mn) was significantly lower in uncovered SMS, while the content of iron (Fe) and copper (Cu) was significantly higher. The chemical nitrogen-phospous-potassium (NPK) fertiliser equivalent value of SMS, when applied at a rate of 10 t/ha, was between €105 and €191 per hectare. Nitrogen-phospous-potassium concentrations per kg wet weight were all higher in SMS that was stored under cover, meaning higher chemical fertiliser savings are possible. The high pH of stored SMS (7.8–8.1) means it could be used with good effect on acid soils instead of ground limestone. The low bulk density of SMS (0.545–0.593 g/cm 3) makes it an ideal amendment to soils to improve soil structure and quality. There is some variability in the nutrient content of SMS from different sources, so it is advisable to get the material analysed when including in nutrient management plans.Teagas
Effect of moisture condensation on vapour transmission through porous membranes
No full textPorous membranes find natural application in various fields and industries. Water condensation on membranes can block pores, reduce vapour transmissibility, and diminish the porous membranes\u27 performance. This research investigates the rate of water vapour transmission through microporous nylon and nanofibrous Gore-Tex membranes. Testing consisted of placing the membrane at the intersection of two chambers with varied initial humidity conditions. One compartment is initially set to a high (Rh=95%) water vapour concentration and the other low (Rh=<10%), with changes in humidity recorded as a function of time. The impact of pore blockage was explored by pre-wetting the membranes with water or interposing glycerine onto the membrane pores before testing. Pore blockage was measured using image analysis for the nylon membrane. The mass flow rate of water vapour (ṁv) diffusing through a porous membrane is proportional to both its area (A) and the difference in vapour concentration across its two faces (ΔC), such that m˙v=KAΔC where K is defined as the moisture diffusion coefficient. Correlations are presented for the variation of K as a function of ΔC. Liquid contamination on the porous membrane has been shown to reduce the moisture diffusion rate through the membrane due to pore blockage and the subsequent reduced open area available for vapour diffusion. Water evaporation from the membrane\u27s surface was observed to add to the mass of vapour diffusing through the membrane. A model was developed to predict the effect of membrane wetting on vapour diffusion and showed good agreement with experimental data
Context-Aware Co-attention Neural Network for Service Recommendations
No full textThe 2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW), Macau, China, 8-11 April 2019Context-aware recommender systems are able to produce more accurate recommendations by harnessing contextual information, such as consuming time and location. Further, user reviews as an important information resource, providing valuable information about users\u27 preferences, items\u27 aspects, and implicit contextual features, could be used to enhance the embeddings of users, items, and contexts. However, few works attempt to incorporate these two types of information, i.e., contexts and reviews, into their models. Recent state-of-the-art context-aware methods only characterize relations between two types of entities among users, items and contexts, which may be insufficient, as the final prediction is closely related to all the three types of entities. In this paper, we propose a novel model, named Context-aware Co-Attention Neural Network (CCANN), to dynamically infer relations between contexts and users/items, and subsequently to model the degree of matching between users\u27 contextual preferences and items\u27 context-aware aspects via co-attention mechanism. To better leverage the information from reviews, we propose an embedding method, named Entity2Vec, to jointly learn embeddings of different entities (users, items and contexts) with words in a textual review. Experimental results, on three datasets composed of millions of review records crawled from TripAdvisor, demonstrate that our CCANN significantly outperforms state-of-the-art recommendation methods, and Entity2Vec can further boost the model\u27s performance.Science Foundation IrelandInsight Research Centr