Rega Institute for Medical Research

Lirias
Not a member yet
    263134 research outputs found

    Verder kijken dan de score: Machine Learning-technieken om (vrouwen)voetbal te begrijpen

    No full text
    The rapid growth in the amount of data collected from soccer matches has transformed how the game is analyzed. Soccer clubs increasingly rely on data in their decision-making processes. Player recruitment and match preparation are two critical tasks in professional soccer that could benefit from the use of data. Player recruitment encompasses finding the right player to acquire for your team. However, the global player pool is too large to monitor manually and player performance in one league may not transfer directly to the other due to differences in playing style and level. Additionally, successful transfers depend not only on individual performance but also on tactical and team fit. Match preparation is equally demanding: coaching staff must analyze opponents and prepare set pieces with only a few days between matches. Although video analysis supports this process, time constraints limit how thoroughly staff can assess footage. The increasing availability of complex data from matches provides opportunities to address these challenges using Artificial Intelligence (AI) and Machine Learning (ML) techniques. Quantifying player performance is one of those challenges that has been addressed using AI techniques. Traditionally, player performance was measured using simple counting statistics such as the number of shots on target or the percentage of successful passes. The introduction of more sophisticated data enabled building metrics that better measure player performance. One of the first metrics was Expected Goals (xG), which measures the likelihood of a shot yielding a goal. Since shots only constitute a small proportion of the actions in a match, other metrics have been introduced that value all on-the-ball actions such as passes, dribbles and interceptions. However, these metrics often ignore the broader context in which they were executed. That is, they ignore context such as the player's teammates and the mental aspect. This dissertation addresses this gap by proposing novel AI-based frameworks to 1) quantify player chemistry and 2) player performance under mental pressure. Furthermore, the field of soccer analytics has largely focused on men's soccer, with limited attention to the women's game. Most existing performance models were trained on men's data. Given the machine learning principle that models should be trained on representative data, it raises the question whether such metrics could simply be transferred to the women's game. This dissertation addresses this question. Furthermore, it provides the first large-scale data-driven analysis of the technical and tactical developments in the women's game. It also introduces a novel metric to measure a player's pace of play. Finally, data-driven analysis of penalty kicks is limited and often models penalty kicks as a two-player zero-sum game, assuming that goalkeepers and kickers choose their actions independently. However, this oversimplifies reality as kickers and goalkeepers may adapt their decisions based on the opponent's movements. Moreover, not all players can adapt their actions in this way, and small sample sizes make it hard to assess individual player capacities. This dissertation introduces a simulation framework using AI techniques to optimize goalkeeper policies taking into account these action capacities. In summary, this dissertation has three main contributions to the field of soccer analytics: 1) understanding women's soccer using data, 2) capturing more context in data-driven recruitment, and 3) optimizing goalkeeper policies for penalty kicks.status: Publishe

    Is exposure to chytrid fungus and ranavirus higher in ponds invaded by American bullfrogs?

    No full text
    sponsorship: This research was supported by the Research Foundation Flanders (FWO; No. 1S01822N and No. 1S23822N). (Research Foundation Flanders (FWO)|1S01822N, Research Foundation Flanders (FWO)|1S23822N)status: Publishe

    Een moleculaire blik op de verspreiding en ecologische gevolgen van een uitheemse amfibiesoort: Grondslagen voor doeltreffend beheer

    No full text
    Humanity has reached a point where its influence on the natural environment has become a defining characteristic of a new geological epoch, the Anthropocene. Global trade and transport are eroding biogeographic boundaries that have shaped species distributions for millions of years at an unprecedented pace. Pet trade, aquaculture, hunting, and biological contamination of traded goods are among the many pathways that facilitate the intentional and unintentional introduction of species beyond their native ranges. Some of these species can survive in their introduced range, reproduce, and spread, while often displacing native species, inflicting economic damages, and impacting cultural values along the way. These processes, known as biological invasions, are among the key contributors to the degradation, homogenisation, and devaluation of natural diversity across the globe. To mitigate this multifaceted threat, global, regional, and national policies are put in place to prevent new introductions, remove emerging populations as soon as possible, and contain or reduce the impacts of non-native species that are already widespread. Biological invasions are highly context-dependent processes, with the introduction, establishment, spread, and impact shaped by the abiotic and biotic conditions in which they unfold. Because generalisations across contexts are therefore unreliable, biological invasion management is only as effective as the monitoring data on which it is based. Conventionally, monitoring has relied on capture, visual observation, or auditory detection of target species. However, these approaches face multiple challenges: they require taxonomic expertise that is increasingly scarce, often fail to detect species at low abundances, can be harmful to organisms and ecosystems, are methodologically limited in estimating population sizes, become laborious and costly when applied over large spatial and temporal scales, and face difficulties capturing detailed spatiotemporal invasion dynamics. There is thus a need to revisit the techniques used for monitoring native and non-native species, and to embrace technological innovations. In recent decades, molecular techniques have begun to revolutionize the acquisition of species occurrence data and help improve the understanding of non-native species spread and impact. Environmental DNA (eDNA)-based analyses allow more sensitive and accurate detection of species presence and density, particularly in freshwater ecosystems, while population genetics enable the reconstruction of invasion histories and the assessment of population connectivity. Despite their potential, a persistent reluctance among managers and practitioners to adopt genetic tools for applied purposes continues to hinder their widespread application. This thesis seeks to address this knowing-doing gap by demonstrating the potential of eDNA-based analyses and population genetics to generate actionable knowledge about amphibian invasions in particular, and freshwater invasions more broadly. To do so, it focusses on the American bullfrog (Lithobates catesbeianus) invasion in Flanders (Belgium), a high-profile invader that has become so widespread that managers are left uncertain on how to tackle this ecological challenge most effectively. Introduced in the 1990s, this large pond-breeding frog has rapidly spread in multiple locations in Flanders, but mostly in the Grote Nete river valley. Through competition, predation, and the likely transmission of chytrid fungus and ranavirus, bullfrogs are expected to pose a considerable threat to native amphibians. More specifically, this thesis explores (i) improvements in eDNA-based detection and quantification of bullfrog abundances/densities, (ii) their historic and contemporary distribution and spread, and (iii) their impacts on native amphibians. Experiments showed that bullfrog abundances/densities correlate strongly with eDNA concentrations obtained through quantitative barcoding. Monthly quantification of bullfrog eDNA concentrations in two ponds over two subsequent years revealed clear seasonal patterns in bullfrog activity, providing insights that can be used to define optimal temporal sampling windows for detection. Parallel fyke netting confirmed strong correlations with conventional population-size estimates and showed that breeding sites are consistently characterized by higher eDNA concentrations than ponds used for refuge, foraging, or as stepping stones. Large-scale eDNA surveys demonstrated a widespread occurrence of bullfrogs in the Grote Nete river valley, with occurrences spatially clustered into three main groups. Complementary population genetic analyses supported these groups and revealed that (i) bullfrogs were introduced at multiple locations within the river valley, (ii) culverted river sections function as dispersal barriers, (iii) bullfrogs spread both in an upstream and downstream direction along the floodplain, and (iv) downstream dispersal is likely accelerated by occasional flood events. Importantly, only a small fraction of ponds showed eDNA concentrations indicative of breeding activity. From a metapopulation perspective, these breeding ponds act as source populations driving further spread and were consistently associated with a permanent hydroperiod, extensive emergent vegetation, and limited shading by trees. The remaining ponds were highly variable in habitat characteristics. Further eDNA-based sampling of chytrid fungus and ranavirus showed that the presence and environmental load of chytrid fungus, but not ranavirus, were elevated in ponds where and at times when bullfrogs are present and abundant. While year-round tadpole presence in water bodies are likely to prolong the temporal exposure of native amphibians, seasonal juvenile emigration presumably facilitates the spatial spread of chytrid fungus. By combining quantitative eDNA barcoding with eDNA metabarcoding, the ecological impacts of bullfrogs on native amphibians were found to be density-dependent, species-specific, and site-specific. Native species with a high phenological overlap and a narrow environmental niche were most strongly affected, whereas species with broader niches and lower overlap were less impacted. Sites with high aquatic structural complexity buffered the effects of bullfrogs, but the strength of this buffering varied by species and bullfrog density. These findings reaffirm the context-dependency of invasion impacts and highlight that ecosystem-based management strategies, such as enhancing aquatic structural complexity, are not panaceas but must be combined with conventional removal approaches and aligned with specific conservation objectives. While it is highly unlikely that Flanders will ever get rid of bullfrogs, the molecular tools applied in this thesis identified new avenues for containment and impact mitigation that diverge markedly from current management practices. These insights were translated into a modular, scalable spatial prioritisation scheme to guide managers in allocating limited resources for maximum conservation benefit, alongside a containment strategy that identifies areas to contain and spread routes to prevent. Together, these molecular approaches have provided an evidence-based foundation for bullfrog management in Flanders, serving as a model for managing other pondscape invaders. It further underscores that only when technological innovations are effectively deployed in the field can the ambitious goals of the Convention on Biological Diversity be achieved, and biodiversity safeguarded from the persistent and often irreversible impacts of biological invasions.status: Publishe

    Interplay between microstructure deformation, oil absorption, and physicochemical changes to starch during deep-frying

    No full text
    Starch constitutes the major macronutrient constituent in numerous deep-fried foods. In this study, the impact of starch physicochemical properties during deep-frying on the corresponding microstructural deformations were investigated for wheat starch (WS) and potato starch (PS) hydrated to 40 and 50 % moisture content (MC) using complementary techniques assessing (micro)structure in-situ (time-resolved X-ray microcomputed tomography and variable-temperature time domain proton nuclear magnetic resonance) and ex-situ (differential scanning calorimetry, swelling properties and scanning electron microscopy). Within 5 s of deep-frying, WS and PS swelling increased and most starch gelatinized, resulting in fully amorphous starch chains exhibiting high molecular mobility. Structure expansion during deep-frying (5–30 s), triggered by rapid steam generation in pre-existing air pores, was supported by starch chains in a rubbery state. Notably, PS exhibited greater swelling, more extensive disruption of granule morphology, and formed larger pores in the microstructure compared to WS. With longer deep-frying times, WS and PS granular integrity was progressively lost. Expansion of steam-filled pores, and thus global structure, ceased when starch in the outer regions of the structure lost sufficient water and transitioned to a glassy state. Oil absorption during deep-frying strongly depended on the starting porosity of formulations. PS and WS at 40 % MC were highly porous before deep-frying and displayed considerable oil absorption immediately upon their submersion in oil when vapor pressures were still low. However, PS and WS at 50 % MC did not exhibit porous microstructures before deep-frying and showed no oil absorption during the process (60 s).sponsorship: FWO|G090319N, FWO|I013518Nstatus: Published onlin

    Long legs and small joints: The locomotor capabilities of <i>Homo naledi</i>

    No full text
    The lower limb of Homo naledi presents a suite of primitive, derived and unique morphological features that pose interesting questions about the nature of bipedal movement in this species. The exceptional representation of all skeletal elements in H. naledi makes it an excellent candidate for biomechanical analysis of gait dynamics using modern kinematic software. However, virtual gait analysis software requires 3D models of the entire lower limb kinematic chain. No single H. naledi individual preserves all lower limb elements, and what material is preserved is fragmentary. As an antecedent to future kinematic analysis, a 3D lower limb skeleton was reconstructed from the most complete fossil bones of different H. naledi individuals. As both juvenile and adult H. naledi were used, we tested if the knee joint remained congruent throughout ontogeny in a sample of great apes (N = 143) and modern humans (N = 70). The reconstruction and subsequent comparative analysis reveal that H. naledi had remarkably small joint sizes for their body size, a hyper-elongated tibia, and a high crural index (90.2). We consider that the lower limb morphology of H. naledi could have improved locomotor economy, but the exceptionally small joints cast doubt on its capabilities for long distance travel, including endurance running. The unusual mixture of primitive and derived traits in H. naledi remains intriguing and might indicate that this hominin engaged both in bipedal walking and climbing, demonstrating that kinematic diversity in hominins persisted well into the Middle Pleistocene.sponsorship: This research has been partly funded by BELSPO through the "Neandertal_3D" project for the adaptation of the lhpFusionBox for paleoanthropological aims. Special thanks to Dr. F. Moiseev and D. T. Jerbi for their collaboration during these adaptations. LhpFusionBox is available for noncommercial purposes as part of a research agreement with ULB (contact [email protected]). The lhpFusionBox software was originally developed as part of the LHDL project using the MAF application framework developed by the BioComputing Competence Center, Bologna, Italy (). For the recovery and continuing analysis of the fossil material, the authors thank the National Geographic Society, the South African National Research Foundation, and the Gauteng Provincial Government. We acknowledge access to the site provided by the Jacobs Family and the Lee R. Berger Foundation for Exploration and the permit authority of the South African Heritage Resource Agency and Cradle of Humankind UNESCO World Heritage Site Management Authority. We would also like to thank the University of the Witwatersrand and the Evolutionary Studies Institute, as well as the South African National Centre of Excellence in PalaeoSciences, for the curation of the fossil material and for providing open access to the data used in this study. The authors would like to thank Patrick Semal and Hans Kainz for discussion on this paper and Caroline Polet for the independent observation of the bicondylar angle on the 3D printed reconstructed femur. We further thank Kimberly A. Congdon and William Harcourt-Smith for assistance with an early version of this paper. (BELSPO through the "Neandertal_3D" project, National Geographic Society, South African National Research Foundation, Gauteng Provincial Government, Lee R. Berger Foundation, University of the Witwatersrand, South African National Centre of Excellence in PalaeoSciences)status: Publishe

    Superior Biomechanics of Patient-Specific Mandibular Reconstruction Plate: A Validated Computational-Experimental Framework

    No full text
    Patient-specific mandibular reconstruction plates (PSMRPs) have gained prominence for their precise adaptation to mandibular contours and reported enhanced mechanical performance compared to the manual-bent mandibular reconstruction plates (MBMRPs). However, clinical adoption remains cautious due to insufficient biomechanical evidence directly and carefully comparing their performance. Hence, this study investigated the biomechanical behavior between two mandibular reconstruction assemblies with MBMRP and PSMRP, respectively. Mechanical properties of these two reconstruction systems, including yield and ultimate strength, fatigue strength and life, were evaluated through finite element analyses (FEA) and biomechanical tests, during which digital image correlation (DIC) was used to measure full-field displacements and strains. Results revealed that the PSMRP provides higher stiffness and longer fatigue life to the reconstruction system than the MBMRP, signified by above 33% higher stiffness in the quasi-static compression and exceeding 90% more life cycles in the cyclic test, respectively. These findings not only highlight the biomechanical advantages of PSMRP over MBMRP but also underline the strong correlation between FEA predictions and experimental outcomes in the mandibular reconstruction system, validating the FEA's utility for preoperative biomechanical evaluation. Collectively, this work provides critical evidence supporting the biomechanical superiority of PSMRPs in mandibular reconstruction-potentially reducing risks of plate failure and reoperation-and establishes a translational framework that combines computational and experimental biomechanics to advance patient-specific implant design in oral, dental, and craniofacial surgery.status: Publishe

    Uiterst betrouwbare UAV-ondersteunde voertuigcommunicatienetwerken

    No full text
    Urban wireless networks face growing challenges in providing ultra-reliable, low-latency connectivity in dense, dynamic environments, particularly for future Intelligent Transportation Systems (ITS). In this context, 6G networks are expected to integrate Non-Terrestrial Networks (NTNs), where Uncrewed Aerial Vehicles (UAVs) can enhance coverage and reliability through flexible deployment and improved Line-of-Sight (LoS) conditions. This PhD thesis investigates reliable Air-to-Ground (A2G) communication for UAVs operating as both Aerial Base Stations (ABSs) and Aerial User Equipment (AUEs) in urban environments. For UAVs as ABSs, a 3D geometry-based LoS probability (PLoS) model is developed using ITU built-up parameters. Unlike existing models that rely solely on elevation angle, the proposed approach incorporates elevation, azimuth, and user location, yielding improved accuracy and strong agreement with simulation and ray-tracing results across both structured and irregular urban layouts. The thesis further demonstrates that user heterogeneity, such as pedestrians versus vehicles, significantly affects path loss and should be explicitly modeled. For UAVs as AUEs, the thesis presents a measurement-based, trajectory-aware characterization of the A2G channel using real-world UAV data. Furthermore, a Context-Aware Smart Handover (CASH) protocol is proposed, showing a substantial reduction in handovers and outages compared to conventional 5G A3-based schemes. Overall, this thesis combines analytical modeling, simulation, and measurements to improve the reliability and predictability of UAV-assisted vehicular communication, contributing to the design of future 6G-enabled ITS.status: Accepte

    Smiling for Effect? Smiling as a Strategic Tool in Recruiters' Impression Management in Belgian Job Interviews

    No full text
    sponsorship: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: funding by KU Leuven (project C14/21/038#56286704) (KU Leuven|C14/21/038#56286704)status: Published onlin

    Heritagisation by imagery: The significance of ancient Chinese architecture in 'Peking the Beautiful', 1927

    No full text
    This article examines the photographic volume "Peking the Beautiful" (1927) by Herbert Clarence White as a formative intervention in the global heritagisation of Chinese architecture. Produced amid rising cross-cultural interest and Republican China’s emerging preservationist ethos, the work combines photography and text to construct a proto-heritage discourse across six dimensions: historical, artistic, scientific, social, cultural, and environmental. By employing qualitative analysis of images and texts, alongside a typological approach for quantitative analysis, this article situates Peking the Beautiful as an image-text assemblage that rearticulated architectural significance through transnational visual regimes. It argues that White, though not an architectural expert but a missionary and good photographer, functioned as a cultural mediator, advancing early twentieth-century frameworks of architectural significance and authenticity through visual documentation, textual annotation, and wide networks. In doing so, Peking the Beautiful contributed to the reframing of Chinese architecture within a global heritage consciousness.sponsorship: China Scholarship Council|CSC 202106260024status: Published onlin

    15,455

    full texts

    263,134

    metadata records
    Updated in last 30 days.
    Lirias
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇