Showing papers by "Dublin City University published in 2016"

Findings of the 2016 Conference on Machine Translation

Abstract: This paper presents the results of the WMT16 shared tasks, which included five machine translation (MT) tasks (standard news, IT-domain, biomedical, multimodal, pronoun), three evaluation tasks (metrics, tuning, run-time estimation of MT quality), and an automatic post-editing task and bilingual document alignment task. This year, 102 MT systems from 24 institutions (plus 36 anonymized online systems) were submitted to the 12 translation directions in the news translation task. The IT-domain task received 31 submissions from 12 institutions in 7 directions and the Biomedical task received 15 submissions systems from 5 institutions. Evaluation was both automatic and manual (relative ranking and 100-point scale assessments). The quality estimation task had three subtasks, with a total of 14 teams, submitting 39 entries. The automatic post-editing task had a total of 6 teams, submitting 11 entries.

Shallow and Deep Convolutional Networks for Saliency Prediction

Abstract: The prediction of salient areas in images has been traditionally addressed with hand-crafted features based on neuroscience principles. This paper, however, addresses the problem with a completely data-driven approach by training a convolutional neural network (convnet). The learning process is formulated as a minimization of a loss function that measures the Euclidean distance of the predicted saliency map with the provided ground truth. The recent publication of large datasets of saliency prediction has provided enough data to train end-to-end architectures that are both fast and accurate. Two designs are proposed: a shallow convnet trained from scratch, and a another deeper solution whose first three layers are adapted from another network trained for classification. To the authors' knowledge, these are the first end-to-end CNNs trained and tested for the purpose of saliency prediction.

A systematic review of correlates of sedentary behaviour in adults aged 18–65 years: a socio-ecological approach

Abstract: Background: Recent research shows that sedentary behaviour is associated with adverse cardio-metabolic consequences even among those considered sufficiently physically active. In order to successfully develop interventions to address this unhealthy behaviour, factors that influence sedentariness need to be identified and fully understood. The aim of this review is to identify individual, social, environmental, and policy-related determinants or correlates of sedentary behaviours among adults aged 18-65 years. Methods: PubMed, Embase, CINAHL, PsycINFO and Web of Science were searched for articles published between January 2000 and September 2015. The search strategy was based on four key elements and their synonyms: (a) sedentary behaviour (b) correlates (c) types of sedentary behaviours (d) types of correlates. Articles were included if information relating to sedentary behaviour in adults (18-65 years) was reported. Studies on samples selected by disease were excluded. The full protocol is available from PROSPERO (PROSPERO 2014:CRD42014009823). Results: 74 original studies were identified out of 4041: 71 observational, two qualitative and one experimental study. Sedentary behaviour was primarily measured as self-reported screen leisure time and total sitting time. In 15 studies, objectively measured total sedentary time was reported: accelerometry (n = 14) and heart rate (n = 1). Individual level factors such as age, physical activity levels, body mass index, socio-economic status and mood were all significantly correlated with sedentariness. A trend towards increased amounts of leisure screen time was identified in those married or cohabiting while having children resulted in less total sitting time. Several environmental correlates were identified including proximity of green space, neighbourhood walkability and safety and weather. Conclusions: Results provide further evidence relating to several already recognised individual level factors and preliminary evidence relating to social and environmental factors that should be further investigated. Most studies relied upon cross-sectional design limiting causal inference and the heterogeneity of the sedentary measures prevented direct comparison of findings. Future research necessitates longitudinal study designs, exploration of policy-related factors, further exploration of environmental factors, analysis of inter-relationships between identified factors and better classification of sedentary behaviour domains.

FTIR analysis and quantification of phenols and flavonoids of five commercially available plants extracts used in wound healing

Abstract: Natural products are used in wound healing in order to prevent infection. Propolis is a well known antimicrobial with phenolic compounds and flavonoid content which vary according to the propolis origin. Besides propolis (from both Brazilian and UK sources), pomegranate, dragon's blood and sage are possible antimicrobials to be used in biomaterials. The goal of this work was to analyze the amount of phenols and flavonoid compounds in these natural products, their antioxidant activities and the bonds present by FTIR. The FTIR analysis revealed the presence of active compounds in all drug samples. The phenols quantification showed that Brazilian propolis was rich in phenols compared to the other drugs, followed by pomegranate and UK propolis. UK propolis was the most rich in flavonoids, which is expected on account of its origin. Pomegranate, UK propolis and Dragon's blood presented the highest antioxidant activity. All samples presented antioxidant activity > 82%.

Antibodies and antibody-derived analytical biosensors.

Abstract: The rapid diagnosis of many diseases and timely initiation of appropriate treatment are critical determinants that promote optimal clinical outcomes and general public health. Biosensors are now being applied for rapid diagnostics due to their capacity for point-of-care use with minimum need for operator input. Antibody-based biosensors or immunosensors have revolutionized diagnostics for the detection of a plethora of analytes such as disease markers, food and environmental contaminants, biological warfare agents and illicit drugs. Antibodies are ideal biorecognition elements that provide sensors with high specificity and sensitivity. This review describes monoclonal and recombinant antibodies and different immobilization approaches crucial for antibody utilization in biosensors. Examples of applications of a variety of antibody-based sensor formats are also described.

Screen-printed electrodes for environmental monitoring of heavy metal ions: a review

Abstract: Heavy metals such as lead, mercury, cadmium, zinc and copper are among the most important pollutants because of their non-biodegradability and toxicity above certain thresholds. Here, we review methods for sensing heavy metal ions (HMI) in water samples using screen-printed electrodes (SPEs) as transducers. The review (with 107 refs.) starts with an introduction into the topic, and this is followed by sections on (a) mercury-coated SPEs, (b) bismuth-coated SPEs, (c) gold-coated SPEs (d) chemically modified and non-modified carbon SPEs, (e) enzyme inhibition-based SPEs, and (f) an overview of commercially available electrochemical portable heavy metal analyzers. The review reveals the significance of SPEs in terms of decentralized and of in situ analysis of heavy metal ions in environmental monitoring.

Management of relapsed multiple myeloma: recommendations of the International Myeloma Working Group

Abstract: The prognosis for patients multiple myeloma (MM) has improved substantially over the past decade with the development of new, more effective chemotherapeutic agents and regimens that possess a high level of anti-tumor activity. In spite of this important progress, however, nearly all MM patients ultimately relapse, even those who experience a complete response to initial therapy. Management of relapsed MM thus represents a vital aspect of the overall care for patients with MM and a critical area of ongoing scientific and clinical research. This comprehensive manuscript from the International Myeloma Working Group provides detailed recommendations on management of relapsed disease, with sections dedicated to diagnostic evaluation, determinants of therapy, and general approach to patients with specific disease characteristics. In addition, the manuscript provides a summary of evidence from clinical trials that have significantly impacted the field, including those evaluating conventional dose therapies, as well as both autologous and allogeneic stem cell transplantation. Specific recommendations are offered for management of first and second relapse, relapsed and refractory disease, and both autologous and allogeneic transplant. Finally, perspective is provided regarding new agents and promising directions in management of relapsed MM.

XIPE: the x-ray imaging polarimetry explorer

Abstract: XIPE, the X-ray Imaging Polarimetry Explorer, is a mission dedicated to X-ray Astronomy. At the time of writing XIPE is in a competitive phase A as fourth medium size mission of ESA (M4). It promises to reopen the polarimetry window in high energy Astrophysics after more than 4 decades thanks to a detector that efficiently exploits the photoelectric effect and to X-ray optics with large effective area. XIPE uniqueness is time-spectrally-spatially- resolved X-ray polarimetry as a breakthrough in high energy astrophysics and fundamental physics. Indeed the payload consists of three Gas Pixel Detectors at the focus of three X-ray optics with a total effective area larger than one XMM mirror but with a low weight. The payload is compatible with the fairing of the Vega launcher. XIPE is designed as an observatory for X-ray astronomers with 75 % of the time dedicated to a Guest Observer competitive program and it is organized as a consortium across Europe with main contributions from Italy, Germany, Spain, United Kingdom, Poland, Sweden.

Using Filter Banks in Convolutional Neural Networks for Texture Classification

Abstract: Deep learning has established many new state of the art solutions in the last decade in areas such as object, scene and speech recognition. In particular Convolutional Neural Network (CNN) is a category of deep learning which obtains excellent results in object detection and recognition tasks. Its architecture is indeed well suited to object analysis by learning and classifying complex (deep) features that represent parts of an object or the object itself. However, some of its features are very similar to texture analysis methods. CNN layers can be thought of as filter banks of complexity increasing with the depth. Filter banks are powerful tools to extract texture features and have been widely used in texture analysis. In this paper we develop a simple network architecture named Texture CNN (T-CNN) which explores this observation. It is built on the idea that the overall shape information extracted by the fully connected layers of a classic CNN is of minor importance in texture analysis. Therefore, we pool an energy measure from the last convolution layer which we connect to a fully connected layer. We show that our approach can improve the performance of a network while greatly reducing the memory usage and computation.

Quantifying radiographic knee osteoarthritis severity using deep convolutional neural networks

Abstract: This paper proposes a new approach to automatically quantify the severity of knee osteoarthritis (OA) from radiographs using deep convolutional neural networks (CNN). Clinically, knee OA severity is assessed using Kellgren & Lawrence (KL) grades, a five point scale. Previous work on automatically predicting KL grades from radiograph images were based on training shallow classifiers using a variety of hand engineered features. We demonstrate that classification accuracy can be significantly improved using deep convolutional neural network models pre-trained on ImageNet and fine-tuned on knee OA images. Furthermore, we argue that it is more appropriate to assess the accuracy of automatic knee OA severity predictions using a continuous distance-based evaluation metric like mean squared error than it is to use classification accuracy. This leads to the formulation of the prediction of KL grades as a regression problem and further improves accuracy. Results on a dataset of X-ray images and KL grades from the Osteoarthritis Initiative (OAI) show a sizable improvement over the current state-of-the-art.

Demands or resources? The relationship between HR practices, employee engagement and emotional exhaustion within a hybrid model of employment relations

Abstract: This article explores the ways in which employees may experience and respond to tensions inherent in the mix of potentially confl icting human resource (HR) practices that compose hybrid models of employment relations. By drawing on the job demands–resources (JD-R) literature and viewing HR practices as “demands” and “resources,” we explore the impact of performance manage-ment and employee voice practices on employee well-being, as exemplifi ed by engagement and emotional exhaustion, in a large public-sector organization in Ireland. Our fi ndings suggest that employee voice mechanisms may act as a resource in both enhancing engagement and in counterbalancing the demands presented by a performance management system, thus reducing the deleteri-ous effects of emotional exhaustion. Our study extends understanding of hybrid models of human resource management (HRM) and of the ways in which employ-ees manage the contradictory signals that such models may send in terms of performance expectations.

Bags of Local Convolutional Features for Scalable Instance Search

Abstract: This work proposes a simple instance retrieval pipeline based on encoding the convolutional features of CNN using the bag of words aggregation scheme (BoW). Assigning each local array of activations in a convolutional layer to a visual word produces an assignment map, a compact representation that relates regions of an image with a visual word. We use the assignment map for fast spatial reranking, obtaining object localizations that are used for query expansion. We demonstrate the suitability of the BoW representation based on local CNN features for instance retrieval, achieving competitive performance on the Oxford and Paris buildings benchmarks. We show that our proposed system for CNN feature aggregation with BoW outperforms state-of-the-art techniques using sum pooling at a subset of the challenging TRECVid INS benchmark.

Governance in Blockchain Technologies & Social Contract Theories

Abstract: This paper is placed in the context of a growing number of social and political crit iq ues of blockchain technologies. We focus on the supposed potential of blockchain technologies to transform political institutions that are central to contemporary human societies, such as money, property right s regimes , and systems of democratic governance. Our aim is to examine the way blockchain technologies can bring about - and justify - new models of governance . To do so, w e draw on the philosophical works of Hobbes , Rousseau , and Rawls , analyzing blockchain governance in terms of contrasting social contract theories . We begin by comparing the justifications of blockchain governance offered by members of the blockchain developers ’ community with the justifications of governance presented with in social contract theories . We then examine the extent to which the model of governance offered by blockchain technologies reflect s key governance themes and assumptions located within social contract theories , focusing on the notions of sovereignty, the initial situation, decentralization and distributive justice .

Highly Efficient F, Cu doped TiO2 anti-bacterial visible light active photocatalytic coatings to combat hospital-acquired infections.

Abstract: Bacterial infections are a major threat to the health of patients in healthcare facilities including hospitals. One of the major causes of patient morbidity is infection with Staphylococcus aureus. One of the the most dominant nosocomial bacteria, Methicillin Resistant Staphylococcus aureus (MRSA) have been reported to survive on hospital surfaces (e.g. privacy window glasses) for up to 5 months. None of the current anti-bacterial technology is efficient in eliminating Staphylococcus aureus. A novel transparent, immobilised and superhydrophilic coating of titanium dioxide, co-doped with fluorine and copper has been prepared on float glass substrates. Antibacterial activity has demonstrated (by using Staphylococcus aureus), resulting from a combination of visible light activated (VLA) photocatalysis and copper ion toxicity. Co-doping with copper and fluorine has been shown to improve the performance of the coating, relative to a purely fluorine-doped VLA photocatalyst. Reductions in bacterial population of log10 = 4.2 under visible light irradiation and log10 = 1.8 in darkness have been achieved, compared with log10 = 1.8 under visible light irradiation and no activity, for a purely fluorine-doped titania. Generation of reactive oxygen species from the photocatalytic coatings is the major factor that significantly reduces the bacterial growth on the glass surfaces.

Concepts and characteristics of the 'COST Reference Microplasma Jet'

Abstract: Biomedical applications of non-equilibrium atmospheric pressure plasmas have attracted intense interest in the past few years. Many plasma sources of diverse design have been proposed for these applications, but the relationship between source characteristics and application performance is not well-understood, and indeed many sources are poorly characterized. This circumstance is an impediment to progress in application development. A reference source with well-understood and highly reproducible characteristics may be an important tool in this context. Researchers around the world should be able to compare the characteristics of their own sources and also their results with this device. In this paper, we describe such a reference source, developed from the simple and robust micro-scaled atmospheric pressure plasma jet (μ-APPJ) concept. This development occurred under the auspices of COST Action MP1101 'Biomedical Applications of Atmospheric Pressure Plasmas'. Gas contamination and power measurement are shown to be major causes of irreproducible results in earlier source designs. These problems are resolved in the reference source by refinement of the mechanical and electrical design and by specifying an operating protocol. These measures are shown to be absolutely necessary for reproducible operation. They include the integration of current and voltage probes into the jet. The usual combination of matching unit and power supply is replaced by an integrated LC power coupling circuit and a 5 W single frequency generator. The design specification and operating protocol for the reference source are being made freely available.

Cross‐level effects of high‐performance work systems (HPWS) and employee well‐being: the mediating effect of organisational justice

Abstract: In this cross-level study, we examine the mediating influence of employee perceptions of the fairness of human resource practices associated with the high-performance work systems model. Data were collected from 187 employees in three companies in Ireland. Using cross-level analyses, employee perceptions of distributive, procedural and interactional justice were found to mediate the relationship between high-performance work systems and job satisfaction, affective commitment and work pressure. The findings also point to a ‘management by stress’ HPWS relationship, suggesting diminished employee well-being, less satisfaction and lower commitment. The research adds to our understanding of the mechanisms through which human resource practices influence employee outcomes and contributes to debates that move beyond the polemic high versus low employee well-being debates of HRM. The discussion reviews the theoretical and practical implications of these results.

A prospective view of animal and human Fasciolosis

Abstract: Fasciolosis, a food-borne trematodiasis, results following infection with the parasites, Fasciola hepatica and Fasciola gigantica. These trematodes greatly affect the global agricultural community, infecting millions of ruminants worldwide and causing annual economic losses in excess of US $3 billion. Fasciolosis, an important zoonosis, is classified by WHO as a neglected tropical disease with an estimated 17 million people infected and a further 180 million people at risk of infection. The significant impact on agriculture and human health together with the increasing demand for animal-derived food products to support global population growth demonstrate that fasciolosis is a major One Health problem. This review details the problematic issues surrounding fasciolosis control, including drug resistance, lack of diagnosis and the threat that hybridization of the Fasciola species poses to future animal and human health. We discuss how these parasites may mediate their long-term survival through regulation and modulation of the host immune system, by altering the host immune homeostasis and/or by influencing the intestinal microbiome particularly in respect to concurrent infections with other pathogens. Large genome, transcriptome and proteomic data sets are now available to support an integrated One Health approach to develop novel diagnostic and control strategies for both animal and human disease.

Passive Sampling in Regulatory Chemical Monitoring of Nonpolar Organic Compounds in the Aquatic Environment

Abstract: We reviewed compliance monitoring requirements in the European Union, the United States, and the Oslo-Paris Convention for the protection of the marine environment of the North-East Atlantic, and evaluated if these are met by passive sampling methods for nonpolar compounds. The strengths and shortcomings of passive sampling are assessed for water, sediments, and biota. Passive water sampling is a suitable technique for measuring concentrations of freely dissolved compounds. This method yields results that are incompatible with the EU's quality standard definition in terms of total concentrations in water, but this definition has little scientific basis. Insufficient quality control is a present weakness of passive sampling in water. Laboratory performance studies and the development of standardized methods are needed to improve data quality and to encourage the use of passive sampling by commercial laboratories and monitoring agencies. Successful prediction of bioaccumulation based on passive sampling is well documented for organisms at the lower trophic levels, but requires more research for higher levels. Despite the existence of several knowledge gaps, passive sampling presently is the best available technology for chemical monitoring of nonpolar organic compounds. Key issues to be addressed by scientists and environmental managers are outlined.

Air sensitivity of MoS2, MoSe2, MoTe2, HfS2, and HfSe2

Abstract: A surface sensitivity study was performed on different transition-metal dichalcogenides (TMDs) under ambient conditions in order to understand which material is the most suitable for future device applications. Initially, Atomic Force Microscopy and Scanning Electron Microscopy studies were carried out over a period of 27 days on mechanically exfoliated flakes of 5 different TMDs, namely, MoS2, MoSe2, MoTe2, HfS2, and HfSe2. The most reactive were MoTe2 and HfSe2. HfSe2, in particular, showed surface protrusions after ambient exposure, reaching a height and width of approximately 60 nm after a single day. This study was later supplemented by Transmission Electron Microscopy (TEM) cross-sectional analysis, which showed hemispherical-shaped surface blisters that are amorphous in nature, approximately 180–240 nm tall and 420–540 nm wide, after 5 months of air exposure, as well as surface deformation in regions between these structures, related to surface oxidation. An X-ray photoelectron spectroscopy study o...

A Container-Based Edge Cloud PaaS Architecture Based on Raspberry Pi Clusters

Abstract: Cloud technology is moving towards multi-cloud environments with the inclusion of various devices. Cloud and IoT integration resulting in so-called edge cloud and fog computing has started. This requires the combination of data centre technologies with much more constrained devices, but still using virtualised solutions to deal with scalability, flexibility and multi-tenancy concerns. Lightweight virtualisation solutions do exist for this architectural setting with smaller, but still virtualised devices to provide application and platform technology as services. Containerisation is a solution component for lightweight virtualisation solution. Containers are furthermore relevant for cloud platform concerns dealt with by Platform-as-a-Service (PaaS) clouds like application packaging and orchestration. We demonstrate an architecture for edge cloud PaaS. For edge clouds, application and service orchestration can help to manage and orchestrate applications through containers. In this way, computation can be brought to the edge of the cloud, rather than data from the Internet-of-Things (IoT) to the cloud. We show that edge cloud requirements such as cost-efficiency, low power consumption, and robustness can be met by implementing container and cluster technology on small single-board devices like Raspberry Pis. This architecture can facilitate applications through distributed multi-cloud platforms built from a range of nodes from data centres to small devices, which we refer to as edge cloud. We illustrate key concepts of an edge cloud PaaS and refer to experimental and conceptual work to make that case.

Osteogenic cell response to 3-D hydroxyapatite scaffolds developed via replication of natural marine sponges

Abstract: Bone tissue engineering may provide an alternative to autograft, however scaffold optimisation is required to maximize bone ingrowth. In designing scaffolds, pore architecture is important and there is evidence that cells prefer a degree of non-uniformity. The aim of this study was to compare scaffolds derived from a natural porous marine sponge (Spongia agaricina) with unique architecture to those derived from a synthetic polyurethane foam. Hydroxyapatite scaffolds of 1 cm3 were prepared via ceramic infiltration of a marine sponge and a polyurethane (PU) foam. Human foetal osteoblasts (hFOB) were seeded at 1 × 105 cells/scaffold for up to 14 days. Cytotoxicity, cell number, morphology and differentiation were investigated. PU-derived scaffolds had 84–91 % porosity and 99.99 % pore interconnectivity. In comparison marine sponge-derived scaffolds had 56–61 % porosity and 99.9 % pore interconnectivity. hFOB studies showed that a greater number of cells were found on marine sponge-derived scaffolds at than on the PU scaffold but there was no significant difference in cell differentiation. X-ray diffraction and inductively coupled plasma mass spectrometry showed that Si ions were released from the marine-derived scaffold. In summary, three dimensional porous constructs have been manufactured that support cell attachment, proliferation and differentiation but significantly more cells were seen on marine-derived scaffolds. This could be due both to the chemistry and pore architecture of the scaffolds with an additional biological stimulus from presence of Si ions. Further in vivo tests in orthotopic models are required but this marine-derived scaffold shows promise for applications in bone tissue engineering.

Fundamental movement skill proficiency amongst adolescent youth

Abstract: Background: Literature suggests that physical education programmes ought to provide intense instruction towards basic movement skills needed to enjoy a variety of physical activities. Fundamental movement skills (FMS) are basic observable patterns of behaviour present from childhood to adulthood (e.g. run, skip and kick). Recent evidence indicates that children have the developmental potential to master most FMS by 6 years of age during physical education, physical activity (PA) and sport.Purpose: With a noticeable absence in the literature relating to adolescent movement patterns, the present study assessed the performance of 9 FMS during physical education class amongst 12- to 13-year olds. The study further assessed the range of FMS at the behavioural component level with a view to identifying weaknesses within performance across individual skills.Participants and setting: Baseline data were collected in 2010 as part of a larger longitudinal study evaluating the effectiveness of a prescribed adolescent...