Access to Research at National University of Ireland, Galway
Research at National University of Ireland, GalwayNot a member yet
15712 research outputs found
Sort by
A stochastic prototypical network for few-shot intrusion detection in CAN-based IoV network
No full textThe Controller Area Network (CAN) acts as the backbone of intra-vehicle communication in modern Internet of Vehicles (IoV) systems, enabling real-time coordination among critical automotive subsystems. Despite its widespread adoption, CAN lacks essential security mechanisms such as encryption and message authentication, rendering it highly vulnerable to cyberattacks that can jeopardize vehicle safety and operational integrity. Developing an effective Few-Shot Learning (FSL)-based Intrusion Detection System (IDS) for CAN networks presents challenges due to data scarcity, noisy traffic, dynamic attack patterns, and the need for real-time efficiency. Existing FSL approaches often rely on deterministic models that struggle to capture the uncertainty and variability inherent in CAN network traffic. To address these challenges, we propose a Stochastic Prototypical Network based on a Random Neural Network (RaNN) for few-shot intrusion detection in CAN-based networks. RaNNs are inherently stochastic, enabling them to model uncertainty and variability in network traffic. By integrating RaNN with the prototypical network, the proposed framework computes stochastic prototypes that represent the distribution of normal and attack behaviors, improving robustness in noisy and dynamic environments. Additionally, the framework quantifies uncertainty in its predictions, enabling the system to flag ambiguous cases for further analysis, thereby reducing the risk of both false positives and negatives. The proposed approach demonstrates high classification performance across all FSL scenarios, achieving a maximum accuracy of 99.17% in a 15-shot configuration. The framework shows impressive computational efficiency with millisecond inference times and minimal training overhead, making it suitable for real-time deployment.peer-reviewe
DiaSafety-CC: Annotating dialogues with safety labels and reasons for cross-cultural analysis
No full textA dialogue dataset developed in a language can have diverse safety annotations when presented to raters from different cultures. What is considered acceptable in one culture can be perceived as offensive in another culture. Cultural differences in dialogue safety annotation is yet to be fully explored. In this work, we use the geopolitical entity, Country, as our base for cultural study. We extend DiaSafety, an existing English dialogue safety dataset that was originally annotated by raters from Western culture, to create a new dataset, DiaSafety-CC. In our work, three raters each from Nigeria and India reannotate the DiaSafety dataset and provide reasons for their choice of labels. We perform pairwise comparisons of the annotations across the cultures studied. Furthermore, we compare the representative labels of each rater group to that of an existing large language model (LLM). Due to the subjectivity of the dialogue annotation task, 32.6% of the considered dialogues achieve unanimous annotation consensus across the labels of DiaSafety and the six raters. In our analyses, we observe that the Unauthorized Expertise and Biased Opinion categories have dialogues with the highest label disagreement ratio across the cultures studied. On manual inspection of the reasons provided for the choice of labels, we observe that raters across the cultures in DiaSafety-CC are sensitive to dialogues directed at target groups compared to dialogues directed at individuals. We also observe that GPT-4o annotation shows a more positive agreement with DiaSafety labels in terms of F1 score and phi coefficient.This publication has emanated from research conducted with the financial support of Research Ireland under Grant Number 12/RC/2289_P2 - In sight Research Ireland Centre for Data Analytics
On the impact of rain on computer vision in automotive applications
No full textThe core image processing and machine learning technologies used to enable Connected and Autonomous Vehicles (CAV) are typically developed in ideal conditions, particularly good weather conditions. Given the functional safety requirements expected of such systems, they must also operate reliably under adverse environmental conditions; however, there is a comparative lack of research into the performance of such technology in adverse weather conditions, compared to benign weather conditions. This thesis addresses this gap by investigating the impact of rain on technology
for CAV, with a particular focus on visible-spectrum cameras. The primary contributions of the thesis include:
- Firstly, a comprehensive review of the literature, including the effects of rain on the imaging pipeline and its downstream impact on deep learning-based perception algorithms. Inspired by communications system design, a novel “Image Formation Framework” (IFF) that treats image formation and perception as a communications problem, is introduced to segment the problem space, offering insights into the changes caused by rain as perceived by automated vehicles. The literature review also evaluates how well publicly available datasets capture these effects, revealing a significant gap in academic research concerning the impact of rain on autonomous vehicles.
- Secondly, a detailed study investigates the performance of state-of-the-art object detection algorithms when applied to rainy versus clear images, using the Berkeley Deep Drive (BDD) dataset of real-world images. In addition to high level object detection results, the study includes an in-depth error analysis, identifying where and how the models fail under adverse conditions. Additionally, an object-level analysis is presented based on pixel-level metrics. The limited meta-data included in publicly available datasets restricted the analysis of factors such as rain intensity, droplet distribution, and scene-dependent visibility degradation. This study underscores the need for custom datasets, or enhanced labelling in existing datasets, to better capture the variability introduced by adverse weather.
- To gain a more comprehensive understanding of the effects of rain on automotive cameras under controlled conditions (in contrast to real-world conditions), experimental data were collected at the AVL Mobility and Sensor Test Center in Roding, Germany. This controlled environment enables the collection of data while isolating the effects of rain from other environmental factors, enabling direct comparisons between rain and no-rain conditions and minimising the impact of confounding variables.
- Using the data collected in controlled conditions, two key studies were completed, focusing, firstly, on the degradation of image quality due to rain, and secondly, the subsequent effect of rain on perception performance using deep learning models. These studies allowed for deeper insights into the impact of rain on camera images, and object detection systems.
- Finally, a computationally-efficient framework for simulation of rain was developed (working with an industry collaborator). Overall, the thesis contributes important insights into the effect of rain on the computer vision pipeline in CAV, that goes beyond high-level object detection metrics
to encompass detailed object and image-level analysis of the effects of rain. The contributions should provide useful information to designers of computer vision systems for CAV applications operating in inclement weather conditions.This work was jointly funded by Taighde Éireann - Research Ireland (previously Science Foundation Ireland) grants 13/RC/2094 P2 and 18/SP/5942, and co-funded under the European Regional Development Fund through the Southern and Eastern Regional Operational Programme to Lero - the Research Ireland Centre for Software (www.lero.ie), and by Valeo Vision Systems
Using distributed acoustic sensing to monitor Galway bay’s ocean noise
No full textThe global ocean environment consists of various complex mechanisms whose better understanding is crucial to study topics such as climate change and the interactions between the ocean bottom and solid Earth. The emergent technology of Distributed Acoustic Sensing (DAS) is presenting itself as a significant tool to help overcome some of these barriers.
DAS utilises the optical phase difference of Rayleigh backscattered light from within an optical fibre to measure the dynamic strain of the fibre in response to mechanical signals interacting with it, producing a spatially dense dataset of strain-meters, known as channels.
Within this thesis, an optical fibre connected to the Galway SmartBay, Galway, Ireland, an offshore observatory hosting multiple ocean environment sensors, was interrogated for two acquisitions.
The spatial density of DAS provides more information which can be utilised to better understand the oceanographic wavefield behaviour in the Galway Bay area. This included characterising the impacts of the tides, swells generated from distant storms and ocean currents on the wavefield, providing another source of information for climate and coastal hazard monitoring.
Following this, the presence of a local secondary microseisms and ambient high frequency Scholte waves was identified and studied. These wavefields were observed to correlate temporally as well as with local weather conditions, indicating the presence of a potential coupling mechanism between the atmosphere, sea surface, and solid Earth. This highlights the feasibility of near-shore case studies on this coupling mechanism, as well as the opportunity to use DAS to further study ocean processes and their implications on the seismic wavefield in Galway Bay.
Lastly, some applications of DAS were explored, including the earthquake detection and acoustic signal tracking capabilities of the fibre. These show great potential for incorporation of DAS measurements within offshore facilities, such as the Galway SmartBay.Geological Survey of Irelan
Association between Body Roundness Index and cancer risk, with further stratification by cardiometabolic disease status: Findings from three national longitudinal cohorts
No full textObjective
This study investigates the association between body roundness index (BRI) and overall cancer incidence in a large population-based cohort and explores the consistency of this association in individuals with and without cardiometabolic disease (CMD).
Design
A prospective cohort study.
Setting
Data for this study was extracted from three cohorts: the English Longitudinal Study of Aging, the Health and Retirement Study, and the China Health and Retirement Longitudinal Study.
Participants
Among 33,624 individuals aged 50 and older, 2999 reported a cancer diagnosis.
Measurements
BRI was calculated using measured height and waist circumference, while cancer and CMD were based on self-reported physician diagnoses. Restricted cubic spline curves assessed nonlinearity and identified BRI cutoff points. Cox proportional hazards models were used to evaluate the association between BRI and cancer incidence, with subgroup and interaction analyses conducted by CMD status.
Results
A non-linear relationship was identified between BRI and cancer incidence, with an inflection point at BRI = 5.06. Participants with BRI > 5.06 had a 10% higher risk of cancer (HR: 1.10; 95% CI: 1.02–1.19) compared to those with BRI ≤ 5.06, while increases in BRI below this threshold were not significantly associated with risk. This remained consistent across CMD-stratified analyses. No significant interaction was detected between BRI and CMD status.
Conclusion
BRI is associated with cancer risk in a non-linear manner, with increased risk above a defined threshold, irrespective of CMD status. These findings highlight the potential of BRI as a simple tool for cancer risk screening in older adult
Public order and the internal security apparatus: Affective tension monitoring as police epistemology
No full textFarmer develops the concept of the civil order to help understand the function of criminal law, but civil order is a particularly capacious concept. In this article I use it to frame the field of public order and the internal security apparatus. The internal security apparatus names those elements of the state which ensure a docile populace. This article suggests that internal security should be considered as an affective apparatus, arguing that at its core is the need to understand and manage the affective life of the populace. To show how this operates, the article develops a genealogy of the technique of ‘tension monitoring’ in Britain. The article argues that tension monitoring is a particular form of ‘police epistemology’ within the internal security apparatus, one which is oriented around resonance rather than veracity; and which produces the collective subjects of ‘community’ and ‘the populace’, as the objects of its interventions
High-performance THz nanometamaterial absorber with negative permittivity (0.1–10 THz) for early cancer detection via circulating exosomes
No full textCancer remains one of the leading causes of mortality worldwide, with nearly 10 million cancer-related deaths and approximately 30 million new cases projected by 2030. While significant advances have been made in therapeutic strategies, such as targeted treatments, immunotherapy, and gene therapy, early diagnosis remains crucial for improving patient outcomes. However, conventional diagnostic tools like magnetic resonance imaging (MRI), computed tomography (CT), and histopathology often suffer from limited sensitivity and specificity. Emerging technologies, including liquid biopsy, nanophotonics, and terahertz (THz) sensing, offer promising alternatives by enabling faster, more accurate detection. In particular, extracellular vesicles (EVs)—specifically exosomes, which are nanometer-sized vesicles involved in cancer progression—have gained attention as biomarkers for early stage cancer detection. Traditional methods for detecting exosomes typically rely on labeling with antibodies or dyes, which can be costly, time-consuming, and prone to false positives. This study presents a novel THz nanometamaterial absorber designed for early cancer detection through circulating exosomes. The absorber features ultrawideband operation from 0.1 to 10 THz and achieves absorption rates exceeding 92%, demonstrating exceptional sensitivity and precision. With a compact footprint of 100×100 nm2 and a thickness of just 30 nm, the design incorporates silver (Ag) resonators, dual dielectric substrates (silicon dioxide [SiO2] and titanium dioxide [TiO2]), and a nickel (Ni) backplane. This configuration enables optimized plasmonic and dielectric interactions for effective electromagnetic wave absorption. Key performance features include polarization insensitivity, reduced noise, and the ability to achieve negative permittivity at 4.85 THz—enhancing the sensor’s responsiveness to subtle refractive index changes in nanoscale biological samples such as cancerous exosomes. Comprehensive numerical analyses, including field distributions, surface currents, and scattering parameters, validate the sensor’s performance. Benchmark comparisons highlight the proposed sensor’s superior absorption efficiency, sensitivity, and nanoscale precision, setting a new standard for noninvasive, early stage cancer diagnostics.Co-funded by the European Union.
Besides that, this publication has emanated from research jointly funded by Taighde Éireann – Research Ireland under Grant number 13/RC/2094_2, the European Union’s Marie Sklodowska-Curie Actions under grant number 101126578 and was supported in part by University of Galway. In addition, the authors would like to thanks to the Universiti Kebangsaan Malaysia research grant Dana Impak Perdana 2.0 (DIP 2.0) for funding through the grant number: DIP-2023-028. Furthermore, part of this research has been supported by the Icelandic Research Fund (Grant No. 2410297), by the National Science Centre of Poland (Grant No. 2020/37/B/ST7/01448)
The unrealized potential of phenomenology in understanding addiction: A critical exploration
No full textThis paper seeks to make a contribution to addiction research by introducing some key phenomenological concepts and discussing how they can illuminate the lived experience of people with addiction. We begin by briefly sketching the historical and philosophical background of phenomenology, highlighting its focus on subjective experience and its clinical relevance. In the second part of the paper we introduce some fundamental aspects of lived experience, e.g., temporality, affectivity, embodiment and intersubjectivity, and explore alterations in these aspects that occur during addiction. We also discuss how an awareness of the socio-cultural dimension of such alterations has the potential of improving clinical encounters. In the third part of the paper we focus on three articles on addiction published in Philosophy, Psychiatry and Psychology in 2024. We explore how phenomenology can add nuance to some of the ideas discussed in these pieces. Our key message is that phenomenological concepts and a critical phenomenological attitude can make a crucial contribution to current research initiatives
Supramolecular approaches for protein framework modulation
No full textProteins are the building block of choice in nature to make complex, functional frameworks. The choice of 20 amino acids provides a rich diversity of sequence possibilities, allowing for the evolution of a wide range of structures that carry out unique functions. There is great interest in manufacturing highly ordered (supramolecular) protein architectures with a view to equal or exceed the structural properties demonstrated by nature’s examples. However, the structural diversity and complexity of proteins pose a challenge to controlled protein assembly. This thesis describes the use of macrocycles, macrocycle-metal complexes, mutagenesis, seeding, and crystallization to overcome this challenge, specifically exploiting supramolecular interactions between synthetic macrocyclic hosts and macromolecular guests. NMR spectroscopy and X-ray crystallography were used to observe host-guest interactions in solution and in the solid state. The model protein cytochrome c was cocrystallized with a complex of sulfonatothiacalix[4]arene (tsclx4) and zinc to yield two distinct crystal forms depending on the concentration of zinc. Further investigation was conducted with a histidine mutant of the trimeric protein RSL, yielding a dinuclear tsclx4/Zn complex, suggesting that the macrocycle-metal complex is an adaptable protein binder.1 Modulation of a previously described cubic assembly2 of sulfonato-calix[8]arene (sclx8) and RSL was carried out by surface charge modification and cross-seeding. Three mutants of RSL were generated to investigate the effect of replacing aspartic acid with asparagine. The goal was to alleviate charge repulsion with the acidic macrocycle and therefore enable cocrystallization at higher pH. Two cocrystal forms were obtained with one of the mutants, both requiring the same pH as the original structure. Control over which crystal form grew was enabled via seeding. The remaining mutants and wild type RSL were unaffected by cross-seeding in these conditions, suggesting replacement of the acidic side chain with an amide at this site is necessary for framework modulation. When this mutant was crystallized in the same conditions as native RSL,2 , an essentially identical assembly was observed. However, cocrystallization of a histidine mutant under these conditions resulted in an altered framework. The results reveal an attractive method for engineering protein frameworks.3 Complexation of sclx8 and Pent, a designed pentameric β-propeller lectin (developed by the Tawfik laboratory) was evaluated by using NMR spectroscopy and X-ray crystallography. Two crystal forms of the Pent – sclx8 complex were obtained, depending apparently on the ionic strength of the crystallization conditions. In conditions containing PEG as the sole precipitant, a 1:1 complex was obtained, where the calixarene forms multivalent interactions with the cationic protein surface. When the ionic strength is increased, a dimer of proteins sandwiches the calixarene. In this case the protein may be considered the host and the calixarene the guest. Strong binding to the concave, cationic pocket on Pent was observed in NMR, confirmed by a slow exchange process on the NMR time scale. This data,indicative of ~µM affinity, was consistent with the large protein – calixarene interface observed in the crystal structures. The calixarene scaffold, with its large surface area (~1500 Å2 for the pleated loop), is a versatile multivalent protein binder.4 The Pent – sclx8 system was further investigated with different crystallization conditions, yielding an icosahedral assembly of pentamers. This icosahedral assembly was confirmed using SAXS under similar conditions. The high porosity (∼70% solvent content) may have resulted in the lowresolution X-ray data of the crystals. However, Pent – sclx6 cocrystals provided evidence that sclx6 bound to the lysine-containing vertex of the pentamer. These data suggested that sclx8 may bind in a similar fashion, resulting in the formation of an icosahedron.5 Though this subject is still relatively new, much work has been done to understand protein – calixarene interactions. The results described in this thesis highlight the potential for macrocyclemediated protein assembly towards the fabrication of functional materials. Further work expanding on the concept of supramolecular synthons in protein – macrocycle frameworks may lead to precise framework fabrication across a range of different systems
Beyond MALDI: Molecular recognition of gentisic acid by a model protein
No full textThe benzoic acid derivative gentisic acid (2,5-dihydroxybenzoic acid), is a common sample matrix molecule used in matrix-assisted laser desorption/ionization (MALDI) mass spectrometry of proteins. Beyond its use in MALDI, gentisic acid has been shown to exhibit notable polymorphic properties i.e., capable of adopting more than one crystal form under varying crystallization conditions.
In this study, we explored the co-crystallization of gentisic acid with lysozyme, a well characterized model protein known for its ease of crystallization. Motivation for this work stems from the hypothesis that the polymorphic nature of gentisic acid, may play a role in driving and/or directing protein assembly, potentially resulting in novel crystalline architectures of lysozyme.
X-ray crystallographic analysis revealed two distinct binding modes of gentisic acid on the protein surface. These binding modes show gentisic acid engaging in hydrogen bonding, salt bridge and cation-π interactions with residues on the protein surface. Additionally, a ligand-free structure of lysozyme, devoid of gentisic acid was obtained in the presence of Mg²⁺. This result provides some insight into metal coordination and its impact on the protein binding potential of simple ligands such as gentisic acid.
We also investigated co-crystallization of the cationic macrocycle, Bluebox with the neutral model protein Ralstonia solanacearum lectin (RSL), in an attempt to investigate molecular recognition using cationic receptors. However, our attempts yielded two small molecule crystals of Bluebox–Acetate and Bluebox–SCN salts. The host-guest properties of Bluebox may warrant further exploration in protein co-crystallization