263134 research outputs found
Sort by
Testing and improving the robustness of amortized bayesian inference for cognitive models
No full textstatus: Accepte
Ontrafelen van endotheelcel mechanica in gezondheid en ziekte aan de hand van in vitro modellen
No full textThe vascular system is a highly dynamic network that carries blood to and from tissues through large arteries and veins, and facilitates nutrient and waste exchange via its extensive microvascular (MV) network. Endothelial cells (ECs) form the inner lining of these vessels, simultaneously acting as a restrictive barrier and the primary location for molecule and cellular exchange. Maintaining appropriate cell morphology, gene expression and activity is critical for endothelial function. However, this optimal phenotype varies across vessel types and vascular beds, reflecting the remarkable heterogeneity and adaptability of ECs. Upon stimulation by proangiogenic factors, ECs undergo phenotypic changes that initiate angiogenesis - the formation of new blood vessels from the pre-existing vasculature. This process in vital in vascular maintenance and repair, with ECs playing a central role. Dysregulated angiogenesis can lead to severe pathologies, including Cerebral Cavernous Malformations (CCMs). EC mechanics are fundamental to both vascular homeostasis and angiogenesis, encompassing intercellular junction stability and interactions with the extracellular matrix (ECM) such as adhesion, matrix degradation and traction force generation during migration and invasion. Quantifying these cellular forces provides crucial insight into endothelial migration and vascular remodeling under physiological and pathological conditions. In this context, robust invitro models compatible with Traction Force Microscopy (TFM) - particularly those that are high-throughput and user friendly - are invaluable tools for elucidating endothelial mechanical behavior.
Two-dimensional (2D) single-cell assays, enable detailed investigation of the fundamental mechanical behavior of individual ECs without interference from neighboring cells. Extension of this model by incorporation of paracrine signaling and ECM remodeling further allows the investigation of indirect cell-cell communication mechanisms. However, the key advantage of studying individual cells also represents the main limitation of this approach, as EC behavior within the endothelium is intrinsically influenced by interactions with adjacent cells. To address this limitation, cell patterning techniques can be applied to organize ECs in defined multicellular configurations while maintaining high throughput and compatibility with TFM. Various methods for selective collagen functionalization are available and can be chosen according to the experimental objectives and technical constraints. While 2D models are valuable tools to study the quiescent endothelium, they inherently fail to reproduce three-dimensional (3D) angiogenic invasion - an essential process in both physiological and pathological vascular remodeling. To capture this EC behavior, a 3D TFM-compatible invasion assay was developed and optimized. The utility of this model is demonstrated in two studies: (1) investigation of mechanosensitive ion channels in CCMs and (2) assessment of organ-specific mechanical heterogeneity among ECs.
CCMs are mulberry-like vascular lesions that predominantly develop in low-flow regions in the central nervous system. Because of this low-flow aspect, mechanosensitive ion channels - typically sensitive to shear stress modulations - have emerged as molecules of interest in CCM research. Within the broader context of this project, upregulation of PIEZO1, TRPV2, and TRPV4 was identified in CCM lesions and CCM2-mutant primary ECs. This upregulation was shown to play a critical role in CCM pathogenesis: PIEZO1 and TRPV4 contribute to increased endothelial permeability, while PIEZO1 and TRPV2 drive excessive ECM degradation; in addition, TRPV2 modulated 2D cellular force generation. The optimized 3D invasion assay was further employed to investigate early angiogenic invasion dynamics of CCM2-mutant cells. This study revealed TRPV2 as a key regulator of angiogenic invasion and wildtype cell recruitment, whereas PIEZO1 was found to be essential in 3D force generation. Collectively, the identification of these three upregulated channels as contributors to CCM pathology highlights their potential as therapeutic targets - an especially significant finding given the current lack of pharmacological treatment for CCM.
Endothelial heterogeneity manifests across organs and vessels, resulting in distinct genetic profiles and functional profiles among EC populations. However, mechanical heterogeneity across these populations remains poorly understood. To address this, ECs derived from the umbilical vein, as well as from the microvasculature of the brain, lungs, and skin, together with MV lymphatic cells from the skin, were compared using the 3D invasion assay. This study elucidated key invasion mechanisms, including the importance of tightly regulated ECM degradation during the initial sprouting phase and filopodia formation throughout the sprouting process. Furthermore, traction force quantification revealed clear organ-specific mechanical heterogeneity, with MV cells from the skin exhibiting higher force generation than those from other vascular origins.
While the high-throughput in vitro assays used here have certain limitations - most notably the lack of flow - this work illustrates their strong utility for investigating EC mechanics in both health and disease.status: Publishe
Robuuste akoestische monitoring van roterende machines
No full textThe robust condition monitoring of rotating machinery using non-contact acoustic sensors is a critical objective for modern industry, yet its practical deployment remains challenging. Industrial acoustic environments are characterized by complex soundscapes, strong background interference, and limited availability of fault data for training diagnostic models. In addition, fault-related acoustic signatures are often subtle and highly specific, further complicating reliable detection. This PhD research addresses these challenges by proposing a suite of advanced methodologies that together enhance the robustness, applicability, and practicality of acoustic monitoring systems. The contributions are structured around three central themes: robust source separation, data-efficient learning under fault data scarcity, and the exploitation of physics-informed signal characteristics.
First, to improve robustness against acoustic interference and background noise, this PhD research develops and progressively refines a deep-learning-based framework for unambiguous acoustic source separation. Conventional blind source separation (BSS) techniques suffer from a fundamental ambiguity, resulting in so-called "virtual" rather than physically meaningful sources. To address this limitation, the proposed TUNE-BSS framework learns a transformation from virtual components to physical source maps. This framework is further enhanced by replacing an adversarial loss with a spatial entropy regularization term, which serves as a physically motivated prior and improves both accuracy and zero-shot generalization. Importantly, this approach is not limited to inverse acoustic imaging methods but is also, for the first time, successfully adapted through transfer learning to improve the outputs of conventional beamforming techniques, representing a significant methodological advance for industrial applications.
Second, this PhD research tackles the critical challenge of fault data scarcity, which severely restricts the deployment of data-driven diagnostic models in real-world settings. A novel simulation-driven partial domain adaptation strategy is introduced, in which a physics-based bearing model is used to generate synthetic fault data. These simulated fault representations are aligned with real-world healthy data through contrastive learning, enabling the training of an effective fault classifier in the realistic scenario where only healthy data from the physical asset are available. This approach substantially improves the feasibility of applying deep learning techniques to industrial fault diagnosis.
Finally, this PhD research demonstrates that further robustness can be achieved by explicitly exploiting the physical characteristics of bearing fault signatures. A cost-effective acoustic monitoring method based on a sparse microphone array is presented, which applies beamforming to the low-frequency envelope of acoustic signals to localize amplitude-modulated fault signatures using minimal hardware resources.
Taken together, the contributions of this PhD research provide a comprehensive methodological toolkit for robust acoustic monitoring of rotating machinery. All proposed approaches are rigorously validated using experimental and simulated data, demonstrating significant progress toward making acoustic-based condition monitoring a reliable and practical solution for industrial maintenance.status: Publishe
Grof- en Ultrafijnkorrelige Fe- en Cu-gebaseerde Materialen onder Glijden: Evolutie en Stabilisatie van de Microstructuur onder het Oppervlak
No full textThis PhD research investigates how microstructural design and processing strategies can enhance the strength, stability, and tribological performance of metal-based materials subjected to reciprocating sliding. Modern engineering components—from aerospace actuators to precision mechanical systems—operate under demanding sliding contact conditions where friction, wear, and subsurface damage critically influence reliability and efficiency. Traditional alloy design has struggled to balance high strength with wear resistance, as hard materials often crack under cyclic loading while softer materials deform excessively.
Recent advances in ultrafine-grained (UFG) metals offer a pathway toward improved mechanical performance, yet their behaviour under sliding remains insufficiently understood, particularly regarding subsurface microstructural evolution. This thesis addresses this gap by examining three representative material systems: pure Fe and Cu, Cu-Al₂O₃ metal matrix composites, and Fe-Cu immiscible alloys. Together, they capture the contrasting deformation mechanisms of BCC and FCC metals, the stabilizing role of ceramic dispersoids, and the effects of dual-phase architectures.
The research aims to clarify how processing routes affect microstructure and strength, how friction and wear develop in each system, and how subsurface layers evolve during sliding. Ultimately, the work seeks to establish microstructural design principles for creating wear-resilient, high-strength metallic materials for advanced engineering applications.status: Accepte
On the connected (sub)partition polytope
No full textLet k be a positive integer and let G be a graph with n vertices. A connected ksubpartition
of G is a collection of k pairwise disjoint sets (a.k.a. classes) of vertices
in G such that each set induces a connected subgraph. The connected k-subpartition
polytope of G, denoted by P(G, k), is defined as the convex hull of the incidence vectors
of all connected k-subpartitions of G. Many applications arising in off-shore oil-drilling,
forest planning, image processing, cluster analysis, political districting, police patrolling,
and biology are modeled in terms of finding connected (sub)partitions of a graph. This
study focuses on the facial structure of P(G, k) and the computational complexity of the
corresponding separation problems. We first propose a set of valid inequalities having
non-zero coefficients associated with a single class that extends and generalizes the ones
in the literature of related problems, show sufficient conditions for these inequalities to
be facet-defining, and design a polynomial-time separation algorithm for them. We also
devise two sets of inequalities that consider multiple classes, prove when they define
facets, and study the computational complexity of associated separation problems. Finally,
we report on computational experiments showing the usefulness of the proposed
inequalities.sponsorship: KU Leuven Internal Funds|C14/22/026status: Published onlin
Gezicht-lichaam interacties in de temporale visuele cortex
No full textWe perceive a human body as a whole and not as an arbitrary concatenation of a head with a face and a headless body. However, the large majority of cognitive neuroscience studies treat faces and (headless) bodies as separate units. This is also evident in the existence of a large research tradition on face perception and a separate (but less developed) research line on (faceless) body perception. Behavioral and neuroimaging studies in humans suggested, however, interactions between the processing of faces and bodies. In this project, we aim to study systematically visual interactions between faces and bodies in single neurons of areas in the temporal visual cortex of nonhuman primates. First, we will employ fMRI to map patches in the temporal cortex that are activated by visual images of whole bodies. Then, with single unit recordings, we will examine interactions of body- and face stimulus dimensions in the patches revealed by fMRI. We will determine neural interactions amongst multiple patches of the network with simultaneous recordings during performance of a whole body discrimination task, followed by causal tests in which we will reversibly inactivate nodes of the network to reveal the information flow within the network. These studies will relate the different research traditions of body and face processing, and advance our understanding of the integration of face and body information in temporal visual cortical areas.status: Accepte
In-class crossover of enthusiasm between teachers and students: The moderating role of teacher-student relationship quality and math ability
No full textstatus: Publishe
Het belang van personhood: een exploratie van persoonsgerichte zorg en personhood voor mensen met hoge ondersteuningsnoden
No full textPerson-centred care and support places the whole person in the centre of care and support. The focus is hereby not only on the disabilities or illnesses of the person, but on getting to know the person in all their aspects. To deliver person-centred care and support, it is therefore important to acknowledge the personhood of the person receiving the care and support. However, understanding and acknowledging personhood can be challenging when an individual has high-support needs and is non- or minimally-speaking, such as people with profound intellectual and multiple disabilities and people with late-stage dementia. These individuals have limited symbolic awareness and function on a sensorimotor level of cognitive functioning. Both groups typically require care and support in almost every aspect of their life.
The aim of this doctoral dissertation is to explore person-centred care and personhood for people with high-support needs by defining person-centred care and personhood for these groups (Chapter 2 and 3) and accessing and mapping personhood (Chapter 4, 5 and 6). The first part of this doctoral dissertation is about defining. Chapter 2 presents a rapid literature review in which the preconditions, methods, outcomes and challenges of person-centred care for people with profound intellectual and multiple disabilities are described. This review shows that person-centred care can be seen as an umbrella term with different conceptualisations, methods and focuses. This diversity in the application of person-centred care leads to sometimes conflicting results. In Chapter 3, a concept mapping study is described in which elements of personhood are identified by family members of people with profound intellectual and multiple disabilities and people with late-stage dementia, staff members of residential care facilities and scientific experts. Seven elements were found for people with dementia: identity, vulnerabilities, preferences, personal approach and interaction, experiences of the social context, social background, and vision of life. Also, seven different overarching elements for people with profound intellectual and multiple disabilities were found: capabilities, preferences and experiences, social relations, life story, profile of personal characteristics, communication, and basic safety. Those elements formed the basis for the second part of this dissertation focusing on accessing and mapping personhood.
Chapter 4, 5 and 6 are consecutive studies, in which accessing and mapping personhood of the same six participants with high-support needs were the central focus. In Chapter 4, their family members described how they give meaning to the personhood of their relative with high-support needs in an interview study. They described this by focusing on the past, the present and the future. The stories of the family members highlighted the importance of those family relationships and the co-creation of care. In Chapter 5, a shadowing study is described, in which the six participants were shadowed for two days to see how personhood was reflected in their lives. Their personhood was described by themes such as a basic attitude called a so-state, by behaviour, communication and preferences on a person-driven level, and by the influence of the context. In Chapter 6, a multimethod case study is presented, in which methods from educational sciences and methods from design research were combined. The study shows that providing an overview of who a person is, is possible using those research methods, which were complementary and provided different insights.
This doctoral dissertation shows that getting to know a person, within the (care) relationships, is important in person-centred care. Personhood can be accessed and mapped, but the multifaceted nature of the concept, with a central role for relations, embodiment and temporality, should be considered. One of the implications of this doctoral project, alongside some (ethical) reflections, limitations and other implications, is that small actions in which the personhood of an individual is acknowledged, can have a big impact. The findings of this doctoral project, which were part of a bigger interdisciplinary project, were turned into a practical toolkit which can be used in practice.status: Publishe
Progress and perspectives of crop type mapping with remote sensing: A review
No full textCrop type mapping is a core topic in agricultural
remote sensing, playing a strategic role in global food security and resource management. Advances in remote sensing and artificial intelligence have shifted crop type mapping from traditional expert-driven approaches toward data- and knowledge-driven paradigms. However, a systematic synthesis that links key components of crop identification across data sources, methods, and application contexts remains limited. In this review, we analyze the evolution of crop type mapping over the past five decades through a large-scale meta-analysis of more than 19,000 publications retrieved from the Web of Science Core Collection. The literature was examined using a large language model-assisted workflow applied to titles, abstracts, keywords, and publication metadata, enabling scalable identification of thematic patterns and methodological trends. Building on this analysis, we organize existing studies within an analytical framework that connects crop sampling strategies, feature engineering, algorithm architectures, and validation practices. The review critically assesses empirical approaches, machine learning and deep learning algorithm, transfer learning strategies, and hybrid modeling frameworks, highlighting recent progress in deep feature extraction, learning under limited data conditions, and modeling in complex agricultural environments. Rather than proposing a universal solution, this review provides structured methodological guidance by clarifying the applicability, strengths, and limitations of different approaches under varying environmental conditions, data availability, and mapping objectives. Key challenges are also identified, including early-season and large-scale crop mapping, cross-regional generalization, data gaps and feature drift caused by cloud cover and precipitation in mountainous regions, and the interpretability of deep learning models. Finally, we outline promising directions for all-weather, multimodal, and scalable crop type mapping, emphasizing the emerging role of large remote sensing models enabled by pre-training, adaptive fine-tuning, and data fusion. By synthesizing methodological trends and empirical evidence, this review offers a coherent reference framework and practical insights to support future research in remote sensing-based crop type mapping.status: Accepte
Board Dynamics and Shareholder Engagement: Exploring Culture, Responsiveness and Value Creation
No full textThis dissertation examines how internal board dynamics and external shareholder pressures jointly shape corporate governance and firm outcomes. Drawing on textual and archival data from S&P 1500 firms, the three chapters show that board culture and managerial responsiveness play an important role in explaining variation in firm value and in how firms engage with shareholder demands. Together, the findings highlight the relevance of board-level cultural attributes for understanding corporate behavior and demonstrate how firms both influence, and are influenced by, shareholder activism.status: Publishe