263134 research outputs found
Sort by
Mechanobiologie van het ontlasten van de tussenwervelschijf: een orgaanmodelstudie met een nieuw bioreactorsysteem
No full textLow back pain (LBP) is a major cause of long-term disability in adults worldwide and it is frequently attributed to intervertebral disc (IVD) degeneration. So far, no consensus has been reached regarding appropriate treatment and LBP management outcomes remain disappointing. Spine unloading or traction protocols are one of the non-surgical approaches to treat LBP. These treatments are widely used and result in pain relief, decreased disability or reduced need for surgery. However, the underlying mechanisms -namely, the IVD unloading mechanobiology (i.e. the biological response of the IVD to those protocols)- have not been studied yet. Hence, we believe that studying the biological response of the degenerative IVD to unloading is an important step towards a better understanding of this organ and ultimately towards improved LBP management. In the lab, we will first adapt the actual organ model mimicking disc degeneration, what means, we will create, in disc samples, different degeneration stages. As second step, we will characterize these samples via magnetic resonance imaging (MRI) and standard biological analyses in order to identify and correlate imaging and biological markers that represent the IVD biological state. Afterwards, we will assess the influence of unloading on the previously defined markers, in our tailored organ model. Finally, we aim to translate the results from the lab to the clinics. Therefore, we will look, in patients with IVD degeneration and suffering from LBP, to the modification of the MRI markers following spine unloading protocols. The comparison of MRI and biological markers will allow establishing clinical markers that reflect the biological state of the disc, which would be a breakthrough in the field. Furthermore, the study of disc unloading will help improving the clinical spine traction protocols and LBP management overall.status: Publishe
Understanding migrant views on deservingness: The role of generation, region of origin and perceived discrimination
No full textstatus: Published onlin
Bringing the farm to the living room. Farmers, television and perceptions on the relationship between agriculture, landscape and the environment in Flanders, 1980s-1990s
No full textstatus: Accepte
Correcting for Coupling Delays in Real-Time Co-Simulation using Iterative Learning Control
No full textstatus: Accepte
Dust populations from 30 to 1000 au in the debris disk of HD 120326: Panchromatic view with VLT/SPHERE, ALMA, and HST/STIS
No full textsponsorship: C.D. thanks for fruitful discussions Elodie Choquet regarding handling HST/STIS data, Gaspard Duchene concerning scattering phase function and disk modeling, Veronica Roccatagliata concerning any potential flybys, Natalia Engler about albedo derivation for dust grains, and Anne-Marie Lagrange, regarding the presence of putative planets in HD 120326. C.D. also thanks Karl Stapelfeldt, Elisabeth Matthews, Sophia Stasevic, Xie Chen, Bin Ren, and Oliver Absil for useful discussions. C.D. is also grateful to Anthony Boccaletti, Deborah Padgett, and Sasha Hinkley, who obtained as PI some of the datasets used in this papers. Last but not least, we thank the referee for their comments which improve the quality of this article. This work is based on observations collected at the European Southern Observatory at Paranal with SPHERE under ESO programmes 095.C-0607(A), 095.C-0487(A), 097.C-0060(A), 097.C-0949(A), 097.C-0865(F), 0101.C-0128(D), 0101.C-0016(A), with HST/STIS (program: 12998). In addition, this paper makes use of the following and also observations collected with ALMA (project ID: 2022.1.00968.S) andALMA data: ADS/JAO.ALMA#2022.1.00968.1. ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada), MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ. The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. This work has made use of the High Contrast Data Centre, jointly operated by OSUG/IPAG (Grenoble), PYTHEAS/LAM/CeSAM (Marseille), OCA/Lagrange (Nice), Observatoire de Paris/LESIA (Paris), and Observatoire de Lyon/CRAL, and supported by a grant from Labex OSUG@2020 (Investissements d'avenir - ANR10 LABX56). C.D. is part of Labex OSUG (ANR10 LABX56). C.D. acknowledges support from the European Research Council under the European Union's Horizon 2020 research and innovation program under grant agreement No. 832428-Origins. G.M.K. is supported by the Royal Society as a Royal Society University Research Fellow. A.A.S. is supported by the Heising-Simons Foundation through a 51 Pegasi b Fellowship. T.D.P. is supported by a UKRI/EPSRC Stephen Hawking Fellowship. F.M. has received funding from the European Research Council (ERC) under the European Union's Horizon Europe research and innovation program (grant agreement No. 101053020, project Dust2Planets). J.M. acknowledges support from FONDECYT de Postdoctorado 2024 #3240612 V.F. acknowledges funding from the National Aeronautics and Space Administration through the Exoplanet Research Program under Grants No. 80NSSC21K0394 (PI: S. Ertel) and No 80NSSC23K0288 (PI: V. Faramaz). M.B. is supported by the European Union's Horizon 2020 research and innovation programme under grant agreement no. 951815 (AtLAST). (ALMA|095.C-0607, ALMA|095.C-0487(A), ALMA|097.C-0060(A), ALMA|097.C-0949, ALMA|097.C-0865(F), ALMA|0101.C-0128, ALMA|0101.C-0016, ALMA|12998, ALMA|2022.1.00968, PYTHEAS/LAM/CeSAM (Marseille)|2022.1.00968.1, OCA/Lagrange (Nice), Labex OSUG@2020, European Research Council under the European Union, Royal Society, Heising-Simons Foundation through a 51 Pegasi, UKRI/EPSRC Stephen Hawking Fellowship, European Research Council (ERC) under the European Union's Horizon Europe research and innovation program|101053020, FONDECYT de Postdoctorado 2024, National Aeronautics and Space Administration through the Exoplanet Research Program|3240612, European Union|80NSSC21K0394, European Union|80NSSC23K0288, 951815)status: Publishe
Toward Adaptive Spoken Dialogue Systems for Language Learning: Predicting Task Completion from Learning Process Data
No full textsponsorship: Flanders Innovation & Entrepreneurship|HBC.2020.2302status: Publishe
Real-time hybride-fysiek-virtueel testen in de industriële praktijk brengen
No full textReal-Time Hybrid-Physical-Virtual Testing (HPVT) enables full-system testing by combining physical prototypes with simulation models in real-time, offering significant advantages for industrial product development: reduced testing costs, accelerated development cycles, safe testing of high-risk scenarios, and early-stage system validation. HPVT relies on recent technical advancements in model-based techniques such as hybrid simulation, x-in-the-loop testing, and co-simulation. However, widespread industrial adoption of these advancements is still at an early stage.
This thesis aims at bringing real-time HPVT methods into industrial practice by tailoring and incorporating them into industrial real-time solutions. This should result in an efficient, performant and scalable way of applying real-time HPVT to industrial use-cases in sectors such as the automotive and aerospace ones. Two fundamental challenges are addressed: monitoring and improving the HPVT performance, to achieve robust stability while balancing accuracy and computational performance.
For performance monitoring, energy-based coupling performance indicators (residual power and energy) are investigated for industrial HPVT applications using an originally-designed e-motor test bench. An extensive error analysis reveals how these indicators are influenced by subsystem dynamics, sensor characteristics, coupling parameters, and coupling faults introduced through systematic fault injection. Results demonstrate that residual power and energy serve as robust coupling performance indicators for HPVT systems. The investigated capabilities to detect both gradual parameter changes and sudden system faults opens new possibilities for real-time monitoring and adaptive control strategies in fail-safe HPVT systems.
For performance improvement, Iterative Learning Control (ILC) methodologies are developed to compensate for errors at two levels: coupling delays between subsystems and actuator dynamics within subsystems.
For coupling-level compensation, two similar ILC approaches are developed to compensate for coupling delays. They are validated on both fully simulated and HPVT setups, showing that the coupling error is reduced with several orders of magnitude. However, non-repeating noise present in the HPVT setup hinders ILC convergence, and filtering countermeasures have to be implemented.
For subsystem-level compensation, again two ILC approaches are developed that compensate for HPVT actuator dynamics. They are applied to a representative RTHS benchmark. The first approach involves model-inversion ILC combined with feedback control, to cancel out the dynamics introduced by the actuator and system-under-test. The second approach revolves around a data-driven adaptive inverse control approach that achieves near-perfect tracking, except for an introduced time delay. Here, ILC is applied to compensate for this artificial time delay.
A comparative analysis of all ILC-based methodologies establishes necessary conditions for successful ILC application in industrial HPVT: system stability or stabilisation via feedback control, and reasonable model assumptions, such that robust performance and fast convergence can be achieved in the presence of iteration-varying references. The developed methodologies are real-time compatible and require minimal subsystem information, making them suitable for industrial environments.
To conclude, this research provides practical solutions for industrial HPVT deployment, demonstrated and validated for representative applications covering co-simulation, RTHS structural testing, and automotive electric powertrain testing. The findings pave the way for wider adoption of HPVT in industry, enabling more efficient and reliable product development processes.status: Accepte
Onder druk: hoe lncRNA's de tolerantie voor geneesmiddelen beïnvloeden
No full textLong non-coding RNAs (lncRNAs) are a class of transcripts longer than 200 nucleotides with no coding potential. They are involved in the regulation of epigenetic and post-transcriptional events and growing evidence indicates the link between lncRNAs aberrant expression and cancer development and resistance to therapy. One pending question is how such poorly expressed transcripts impacts the activity of multiple and often abundant proteins. There is recent evidence that osmotic stress, which affects the cellular protein concentration and interaction, induces phenotype switching of cancer cells into an invasive and drug-tolerant state in melanoma. One hypothesis is that osmotic stress induces lncRNAs upregulation in these subpopulations, contributing to molecular crowding and driving liquid-liquid phase separation. The latter is a crucial physical process in signaling cascades and therefore could be responsible for several tumor phenotypes. It is thus interesting to characterize lncRNAs upregulated by osmotic stress, investigate their functions in melanoma and assess whether they can be used as therapeutic targets.status: Accepte
Modellering van Cerebrovasculaire Autoregulatie bij Traumatisch Hersenletsel: Evaluatie van Gevestigde Methoden en Ontwikkeling van Nieuwe Machine Learning-Modellen
No full textSevere traumatic brain injury (TBI) is a life-threatening condition requiring specialized neurocritical care. Following the initial structural damage caused by primary injury, a cascade of secondary pathophysiological events may occur, and because these secondary insults can significantly worsen both short- and long-term outcomes, their prevention or early detection is central to current TBI management. Cerebrovascular autoregulation (CA) plays a pivotal role in preserving brain health by maintaining stable cerebral blood flow (CBF) despite fluctuating cerebral perfusion pressures (CPP). However, CA can be dynamically impaired following the primary injury, which can further compromise cerebral perfusion and oxygenation, exacerbating tissue injury. Maintaining an active CA by continuously monitoring CA using neuromonitoring in TBI patients is therefore essential to prevent the transition from primary to secondary injuries.
Genetic variation may influence baseline CA capacity by altering the myogenic, metabolic, endothelial, and neurogenic pathways regulating CBF, including variants in calcium-signaling (DUSP5, TRPM2, TRPM4, TRPM8) and renin-angiotensin system genes (ACE, AGTR2, AGT, ENPEP, ATP2B1), endothelial mediators such as NOS isoforms (NOS1, NOS2, NOS3), adenosine receptors (ADORA1, ADORA2A), endothelin genes (EDN1-3, EDNRA, EDNRB), prostaglandin/eicosanoid-related genes (PHACTR1, PTGER2), and CYP4A11 involved in 20-HETE production. The exploratory study outlined here found several physiologically plausible associations between these genes - both coding and common non-coding variants - and cerebrovascular reactivity (CVR) after TBI. The main limitations of the study concerned a small sample size and noisy CVR metrics, limiting definitive interpretation. Most variants showed the expected direction of effect with respect to what was previously found in the literature in relation to associated biological mechanisms. Most significant associations involved non-coding variants. Future studies should aim to gather larger cohorts of patients and improved CVR definitions, which will be essential to validate these findings and enable investigation of rare variants. Integrating genetic data into TBI research may help identify key molecular pathways affecting CA, enhance prediction models, and support development of targeted therapies for prevention of failing CA after TBI.
An important concept in TBI research is the dose-response concept, visualizing the relationship of sustained physiological events for a certain duration with outcome using important physiological parameters such as intracranial pressure (ICP) and CPP. Here this concept was explored using a novel prospectively collected pediatric TBI dataset from the multi-center, multi-national KidsBrainIT consortium (n=104). Minute-by-minute ICP and CPP time series were transformed into intensity-duration episodes and linked to 6-month outcome using dose-response visualizations. The analysis validated the previously published pediatric ICP dose-response plot, demonstrating that higher ICP levels were tolerated only for short periods along an exponential transition curve and that any ICP exceeding 20 mmHg was associated with poorer outcome. Moreover, the first pediatric CPP dose-response plots were generated, revealing a similar exponential separation between good and poor outcomes as in adult TBI CPP dose-response plots. Together, these findings validate pediatric ICP dose-response relationships, establish the first pediatric CPP dose-response thresholds, and emphasize the need to reconsider current treatment targets in childhood TBI. However, unlike in adult TBI cohorts, 'normal' ICP and CPP values in children vary across age groups because both baseline arterial blood pressure (ABP) and ICP change with development, with CPP = ABP - ICP. Hence, future studies should increase pediatric datasets to further investigate the effect of age by stratifying along age bands.
Optimal CPP (CPPopt) represents the CPP at which CVR - as surrogate for CA - is most effective and CBF remains stable. Although numerous algorithms have been proposed to compute CPPopt and linked to functional outcomes, studies vary widely in methodological choices, including inclusion criteria, CPPopt variable definitions, preprocessing, autoregulation indices, calculation methods, and statistical modeling. Here, a comprehensive literature review was performed followed by a large-scale multiverse analysis on a novel high-resolution adult (n=57) and low-resolution pediatric (n=202) severe TBI datasets. We systematically applied 7,497,520,044 plausible analytical combinations to the adult cohort and a subset of these combinations on the pediatric and/or adult cohort. More specifically, for the low-resolution dataset, only LAx, L-PRx, and UL-PRx could be computed to derive CPPopt. Of 34 studies meeting inclusion criteria, 41.45% of reported CPPopt-derived summary measures showed significant association with the Glasgow outcome score (GOS). The multiverse analysis revealed a near-uniform p-value distribution for CPPopt-derived predictors in relation to GOS at 6 months, with only 7.73% with a p-value below 0.05. Significant results arose mainly from 3 out of 10 statistical tests, with varying influence of other methodological choices such as exclusion criteria, autoregulation metric, or summarization strategy of CPPopt. These findings suggest that the strength of associations of CPPopt-derived variables with outcome appear to be highly sensitive to the choice of parameters and tests used, and highlight the need for methodological standardization.
Reliable real-time CA monitoring in intensive care units (ICU) remains challenging. Using a porcine cranial window dataset (n=20) with concurrent ABP and ICP recordings, a deep representation learning model was developed to learn an active CA representation from 300-second signal segments. The model achieved high reconstruction accuracy for active CA segments, while reconstruction errors increased steadily as CBF deviated from baseline, reflecting failing CA, with distinct frequency components driving errors during inactive states. Moreover, the observed interaction between increases in error for ABP and ICP and CA state suggests that the model can capture differential CA behavior with increasing versus decreasing CPP. Classifiers built on reconstruction error or latent features substantially outperformed the conventional pressure-reactivity index (PRx) in classifying CA state, improving precision from 0.14 to 0.77 and recall from 0.62 to 0.87. This approach further suggested that relevant CA state information resides in both lower and higher frequency components of ABP and ICP, and that leveraging the full complexity of these signals can provide more accurate dynamic monitoring of CA than traditional correlation-based methods.
Lassen's classic triphasic CA curve defined lower and upper limits of autoregulation (LLA and ULA), however, recent evidence suggests a more complex pressure-flow relationship. Building on the concept of gradual CA failure derived from the quadriphasic model, a novel nonlinear continuous static autoregulation metric (cSARm) was introduced. This method generates a pentaphasic CA curve, capturing gradual transitions between active and impaired CA at both the hypotensive and hypertensive side of the CA curve and provides refined estimates of gradually failing CA between the respective autoregulatory limits. Using a porcine cranial window dataset (n=20) with pial arteriole diameter measurements, cSARm determined multiple LLA and ULA breakpoints (LLA2, LLA1, ULA1, ULA2) based on mathematical features of the first and second derivatives of a nonlinear fit, while mapping the progression from fully active to fully inactive CA onto a continuous 0 to 1 score. Both quadriphasic and cSARm breakpoints were successfully computed, with cSARm generally estimating higher lower limits (LLA1) and slightly lower upper limits (ULA1) in absolute CPP, and provided a smoother transition from fully active to fully inactive CA. Namely, LLA1 reflected the onset of gradual CBF decline, while ULA1 captured early hypertensive responses in smaller arterioles, offering greater granularity than the quadriphasic model. Overall, cSARm extends the quadriphasic framework into a pentaphasic model that more continuously characterizes partially impaired CA zones under both hypotensive and hypertensive conditions.
Adequate CBF is critical for brain health, yet continuous bedside monitoring in the ICU is limited by surrogate measurement techniques, since no reliable and robust CBF probe exists to date. Therefore the feasibility of using Temporal Fusion Transformer (TFT) models - trained on a porcine cranial window dataset (n = 60) - to predict continuous CBF from routinely collected ICU time series was investigated, offering a data-driven approach to real-time monitoring. Raw signals were preprocessed and downsampled to 0.1 Hz, and four modeling conditions were tested to forecast 60-second laser Doppler flow (LDF) time series, with two conditions incorporating historical LDF data. Hyperparameter optimization using a Bayesian Tree-structured Parzen Estimator was performed, followed by full model training. TFT models without historical LDF data failed to capture meaningful temporal dependencies, whereas autoregressive models including past LDF substantially improved predictive performance, enabling accurate short-term forecasts. Attention analyses suggested that poor performance in non-autoregressive setups was driven by misallocated attention and standardization mismatches of the LDF signal. These findings indicate that autoregressive TFT models can support short-term CBF prediction in neurocritical care, particularly when brief calibration periods are available.status: Accepte