Biodiversity is under pressure worldwide, particularly in the Netherlands under sustained anthropogenic stressors. Governments are responding through policy measures that require impact quantification, creating the need for scalable monitoring methods. Geospatial Foundation Models (GFMs) can learn patterns across space, time, and context, potentially enhancing biodiversity monitoring compared to conventional remote sensing methods. Yet the ecological community has not widely embraced GFMs, likely due to their rigidity to changes in modalities without re-pretraining, misalignment between pretraining modalities and downstream tasks, and the unknown effect of data-gaps during discriminatory phenological periods. This study adapts GFMs for biodiversity monitoring via two research questions, addressing the challenges in reverse order: how data-gaps in GFM embeddings impact vegetation species mapping capacity, and how modular GFMs trained on multimodal data affect this capacity. SpheerFM, a Sentinel-2 pixel-in-time GFM pretrained on Dutch nature reserves, is evaluated on the Veluwe, with height data added as a complementary modality. Embedding sensitivity is assessed through period-based ablation scenarios, and height data is incorporated using a modular combination methodology to test multimodal sensitivity against Sentinel-2 perturbations. Results show spatiotemporal data-gaps are prevalent over the Veluwe, also during discriminatory vegetative periods, with likelihood varying by year and period due to cloud cover and tile availability. Unimodal embeddings show sensitivity to data-gaps; masking discriminatory periods reduces discriminability, while masking assumed redundant periods generally improves it. No existing method satisfied all modular and cross-modal requirements. The resulting multimodal embeddings generally prevented worsened classification following data-gaps when the added modality’s features are discriminatory and complementary. ... Read more

Mutational processes leave characteristic patterns of somatic mutations, traditionally studied in tumour tissue. Far less is known about whether they can be observed in normal tissue, particularly blood. Detecting the mutational imprint of disease-associated processes there could enable earlier detection and intervention. Here, we investigate whether the mutational signal detected in blood can be explained by biological and clinical factors, in particular age, DNA repair deficiencies, and cancer diagnoses. Using whole-genome sequencing of blood-derived DNA from 17,419 UK Biobank participants, we developed a filtering strategy to isolate somatic mutations and analysed four views of the mutational landscape: mutation burden, mutation channel composition, exposures to de novo signatures and exposures to COSMIC signatures. We modelled their relation to these factors using regression analysis. Across all four views, sequencing provider was the dominant predictor, far outweighing other predictors. Among non-technical predictors, BRCA (p = 0.024) and POLE (p = 0.030) variants were significantly associated with a higher mutation burden. A leukaemia diagnosis was the strongest signal across the remaining views, appearing in both the mutation channel composition and the exposure to the clock-like signature SBS1 (p = 2.10e-06). De novo signature exposures clustered by sequencing provider, and no association survived when extraction was performed separately for each provider. Our results show that some biological and clinical factors do explain part of the mutational signal in blood, but that technical variation between sequencing providers dominates the mutational landscape and must be addressed before blood can serve as a reliable substrate for mutational analysis. ... Read more

Many real-world time-series, including behavioral, financial, and medical data are highly personal. Event detection models trained on such data typically learn global patterns, capturing differences between time-series but not within them. This treats time-series as interchangeable, which is not always a valid assumption. This is particularly problematic in medical settings because a heart rate that is healthy for one patient may signal a clinical deterioration in another. This thesis investigates how historical personal time-series data can be used to personalize and improve machine learning models for time-series prediction tasks. This is explored in the Intensive Care Unit (ICU), where substantial personal health data is available but rarely fully exploited.

Three personalization methods were implemented: z-score normalization, Empirical Cumulative Distribution Function (ECDF) percentile scoring, and Isolation Forest anomaly detection. Each personalization method constructed patient-specific features by comparing current windows against the patient’s
own historical distribution, also called their baseline. These features were evaluated using static models (XGBoost and Logistic Regression) and a dynamic model (Temporal Convolutional Network) using ECG and PPG waveform data from the MIMIC-III database to predict sepsis and mortality hours in advance.

The results show that adding personalized features consistently improves predictive performance when using a sliding baseline over using global features alone. The AUROC when predicting sepsis improved from 0.72 to 0.82 when adding z-score normalized features, and mortality prediction AUROC improved from 0.77 to 0.82 when adding the Isolation Forest anomaly score. The optimal baseline length is dependent on the task. Using all available prior data is best for predicting sepsis, while a baseline constructed from the previous 4-8 hours is optimal for mortality prediction. A sliding baseline, which uses the most recent history for each prediction window, consistently outperforms a fixed baseline placed at the start of the stay.

The Isolation Forest anomaly score, which was computed individually per patient, was the most important feature for both sepsis and mortality prediction when combining all personalized feature sets with global features. Unlike other personalized features, the anomaly score also improved Logistic Regression performance, suggesting a more linear relationship with mortality risk.

Models trained on PPG data show similar improvement to ECG-based models, and approximately the same optimal baseline lengths are found across both signal types, for both sepsis and mortality prediction. This suggests that the optimal baseline is event-specific rather than just dataset-specific. The strong performance of PPG data is particularly promising because wearable devices could provide
a healthy pre-admission personal baseline, potentially mitigating the cold start problem and improving
performance further.

These results demonstrate that incorporating personal historical baselines into clinical prediction models is a practical and effective approach to improving early warning systems in the ICU. These methods are likely broadly applicable to any time-series domain where individual patterns carry predictive value. ... Read more

As the energy transition accelerates, improving wind energy efficiency and forecasting becomes increasingly critical. One key challenge lies in reconstructing high-fidelity atmospheric boundary layer (ABL) flow fields from sparse measurements, especially in regions influenced by wind turbines. This thesis explores the use of Latent Diffusion Models (LDMs) to reconstruct physically plausible ABL flow fields from limited spatial data. Where previous work focused on homogeneous, neutral ABL states, this research extends the methodology to more diverse and realistic conditions by including both stable and neutral boundary layers with embedded wind farms. A conditional diffusion model is trained in the latent space of an autoencoder (AE), using both local measurements and global atmospheric labels. The model is evaluated using statistical, physical, and spectral metrics, and its ability to generalize across multiple ABL regimes is analyzed. Results show that the model can generate realistic reconstructions from extremely sparse inputs, including turbine wake structures, and can potentially serve as a tool for initializing Large Eddy Simulations (LES). These findings mark an important step toward integrating generative models into operational wind forecasting and control systems. ... Read more

Estimating bike trip times is becoming more and more important in many different areas such as urban mobility and route planning. However, especially in real-world, the GPS data used to generate these estimations is frequently noisy, irregularly sampled, or incomplete. With an emphasis on how these strategies interact with trip length and speed variance, this study intends to examine the effects of various data resampling techniques on the precision of bicycle travel time estimations. To analyze the impact of different preprocessing methods, we apply and assess a graph neural network model using various resampling techniques. Instinctively, the assumption that we expect to be concluded from this research is that there is no single resampling technique works well for every kind of trip. Rather, trip parameters like duration and speed fluctuation have a significant impact on accuracy. ... Read more

This research investigates the effectiveness of graph-based data augmentation techniques in improving the performance of DG4b, a deep learning model designed to estimate bicycle travel times in urban environments. Given the limitations of real-world cycling datasets, particularly data scarcity and trip-length imbalance, we propose two augmentation methods: Graph Stitching (GS), which combines segments of existing trips to form new trajectories, and Graphon-Inspired Trip Generation (GITG), which uses an empirically estimated transition kernel to simulate realistic trip patterns through probabilistic sampling.
Despite limited improvements, this study establishes a foundation for future research in graph-based trajectory augmentation. Integrating richer trip-level features, such as dynamic environmental conditions or behavioral data, with structural augmentation could lead to more effective training data and improved model generalization. ... Read more

Accurate estimation of bicycle trip travel times remains a challenge due to the limited availability of structured cycling data. This paper investigates how graph-based data augmentation can be used to address this limitation, specifically within the context of the DG4B model, a Graph Convolutional Neural Network for travel time estimation. We explore and evaluate three augmentation techniques: Random Walk (with and without node revisiting), Dijkstra Walk and Subgraph Stitching. These methods generate new trips by traversing or recombining paths within an existing road network graph, aiming to expand the training dataset while preserving realistic routing behavior. The augmented data is evaluated both statistically, using metrics like mean, variance and Frobenius norm, and in terms of model performance using RMSE, MAE and MAPE. Experimental results show that Subgraph Stitching and Dijkstra Walk yield the most effective improvements in model accuracy, with each method exhibiting strengths across different trip duration ranges. This work demonstrates that carefully designed graph-based data augmentation can improve GCNN-based travel time predictions in settings with limited cycling trip data. ... Read more

Displacement is a focal point of humanitarian aid efforts, since it affects millions of people globally. Mitigating the consequences of forced migration is important for reducing suffering and one way of doing so is through predicting displacement to prioritise resources in advance. To achieve this, machine learning can be used for its ability to analyse larger amounts of data and identify latent structures more efficiently than human experts. Through a systematized literature review, this research evaluates thoroughly six machine learning tools: UNHCR's Jetson, DRC's Foresight and AHEAD, the EU's EUMigraTool and EPS-Forecasting, and the agent-based simulation Flee, in order to assess their suitability to that end. The analysis compares these tools across several criteria, including the way they use data, algorithmic characteristics, and operational use cases. Finally, it makes recommendations about what should be considered and how to choose amongst the tools for displacement prediction. ... Read more

Natural disasters frequently cause casualties and property losses. Predicting and mitigating the impact of such threats is crucial to the work of humanitarian organizations. The interactions between hazards are best represented through a multi-hazard approach, and machine learning models are well suited for natural hazard prediction. This study presents a systematized literature survey of machine learning in multi-hazard disaster forecasting in the years 2019-2025, focusing on the used models and performance metrics, their applications and feasibility of use, as well as potential cross-applications. There is a wide variety of models and metrics used. The most commonly used models are random forest and support vector machine and the most prevalent performance metric is the ROC-AUC score. The machine learning models generally perform well, with AUC scores above 0.8, though patterns in performance are difficult to examine. Feasibility is defined here as readiness to be used in practice, and the models are rated in the factors that define it. Most of the articles are feasible. Consideration of cross-application is rare and should be extended. This research summarizes the main trends in the field of disaster forecasting, providing a clear reference point for other academics. ... Read more

As humanitarian needs increase while donor budgets decrease, anticipatory strategies are essential for effective crisis response. In this context, machine learning (ML) has emerged as a promising tool for crisis forecasting, offering the potential to support timely interventions and humanitarian decision-making. However, despite rapid developments in ML-based prediction models, questions remain about their practical utility and trustworthiness in real-world humanitarian settings. This study presents a systematic scoping review of 32 academic and gray literature sources to assess the reliability and feasibility of ML systems for conflict forecasting. By analyzing these systems across dimensions such as forecasting scope, data sources, modeling approaches, validation practices, and ethical considerations, the study finds that while some models demonstrate strong predictive performance and methodological rigor, many lack transparent validation, robust error analysis, and operational applicability. The review concludes that while ML systems hold substantial potential for enhancing conflict anticipation, their current real-world readiness is uneven and context-dependent. ... Read more

With the worsening of climate change, the complications brought on by floods every year create an increasing need for forecasting systems that humanitarian organizations can use to help populations in danger. This research presents a literature review of machine-learning models for impact-based flood forecasting, and compares them with existing humanitarian projects. The results examine the characteristics of the models surveyed, while the discussion focuses on understanding how these characteristics can define whether the machine learning models proposed can actually be translated to humanitarian settings. The main takeaways include the prevalent choice of deep learning and ensemble models, used to improve the adaptability of the models, the problems with data availability and data quality in different areas considered, and the difference between lead times, usability, and scalability of the models proposed in contrast with already used humanitarian projects. This study then highlights the importance of transparency and reproducibility of the survey by detailing the queries and databases used, ensuring accessibility of selected articles, and explaining the selection criteria and methodology. Ultimately, the review concludes with the key insights on the connection between academic prototypes and real-life humanitarian projects, as well as key areas for future research. ... Read more

Broccoli for the fresh market in the Netherlands is still harvested manually, which is labor-intensive and increasingly difficult to sustain as seasonal labor declines. Existing mechanized harvesters cut entire fields at once and cannot account for plant-to-plant variation, leading to substantial losses when heads differ in maturity. These constraints motivate plant-level, data-driven growth modeling to support selective and more efficient harvesting.

This thesis investigates how field-based measurements can be used to model broccoli growth at the level of individual plants. The work addresses four problems: converting raw field video into plant-level growth curves, investigating which environmental parameters best describe the growth, determining whether cumulative temperature or thermal time better represents broccoli development, and evaluating how different growth models capture head diameter growth. A preliminary study using an external dataset creates the methodological foundation by analysing broccoli growth and benchmarking classical and neural models. A field study in a Dutch production environment expands this work through the development of a data-processing pipeline that includes head detection, plant identification, tracking, diameter estimation, and integration with local weather data. Across both studies, thermal time provides a more biologically meaningful predictor of development than cumulative temperature. Model comparison shows that classical parametric models capture general developmental trends, while a multi-layer perceptron achieves the highest predictive accuracy, with a mean absolute error of 0.583~cm, when multiple environmental variables are included. The results demonstrate that field-based predictive modeling can support precision-agriculture applications such as harvest planning and selective mechanized harvesting. ... Read more

Automatic Speech Recognition (ASR) systems perform well for typical adult speech but remain challenged by children’s speech, especially that of children with Developmental Language Disorder (DLD). This study investigates how ASR performance can be enhanced for DLD speech while maintaining accuracy on typical child speech. Two state-of-the-art ASR models, a conformer-based model and Whisper-large-v3, were evaluated using Dutch typical (Jasmin) and atypical (Auris) child speech. The experiments examine data augmentation methods, including speed perturbation and vocal tract length perturbation, and transfer learning through fine-tuning. Results show that both techniques improve DLD speech recognition without degrading typical speech accuracy. The best performance was achieved by combining augmentation and fine-tuning with domain-matched DLD data, reaching 53.2% WER on the Auris test set, while mismatched fine-tuning reduced gains, particularly for Whisper. Overall, the findings demonstrate that integrating data augmentation and fine-tuning offers an effective, balanced approach toward inclusive and robust ASR for children with DLD. ... Read more

The combination of increasing global food demand with increased food security risks associated with climate change amid a decreasing number of skilled growers necessitates innovative solutions in green- house horticulture. Autonomous growing offers a solution based on greenhouse climate forecasting and (optimal) control. However, current theoretical models developed for greenhouse climate forecast- ing face limitations due to the in-depth physics knowledge required for their use and their dependence on intricate system parameters that are difficult to estimate. Conversely, machine learning models struggle with generalisation to unseen conditions and adherence to physical laws, leading to unrealis- tic predictions in greenhouse environments. This study addresses these challenges by exploring the use of Physics-Informed Machine Learning (PIML) techniques to enhance greenhouse climate fore- casting. A simple differentiable theoretical model for simulating the greenhouse climate is proposed to serve as prior physical knowledge of the system. Subsequently, a novel PIML model is introduced in the form of Controlled Aphynity (CA), which integrates insights from neural controlled differential equations and Aphynity, and is the first such model that allows for the augmentation of incomplete prior knowledge of a dynamical system with data-driven machine learning models while being adaptable to changing dynamics due to forces acting on the system. Experimental results show that CA is capable of successfully augmenting incorrect physics under changing dynamics on the task of humidity deficit prediction in the greenhouse. Furthermore, three ensemble methods combining CA with traditional ma- chine learning techniques are explored and demonstrate promising synergies. A detailed case study evaluates CA and the best-performing ensemble approach on the task of humidity deficit prediction under realistic greenhouse scenarios over a complete crop cycle. Both CA and the ensemble methods exhibit superior adherence to physical laws, lower data requirements, and improved performance on outliers compared to conventional machine learning methods. This research contributes to advancing greenhouse climate modelling, underscoring PIML’s potential in optimising the greenhouse climate.F ... Read more

The bilateral filter is an edge-aware image filter. While it has a variety of applications, its naive implementation is quadratic in nature, hindering the ability to efficiently process multi-megapixel images. If performance is needed, like in a real-time setting, an approximation is necessary. Current literature on Fourier series-based approximations does not explore the capabilities of graphics processing units (GPUs) as viable platforms for this computational problem. This paper proposes an approach for implementing such filtering on a GPU by conducting a series of separable convolutions, and also investigates the use of different range kernels. Our adaption of the bilateral filter is found to be more two times faster than readily available solutions, with frame times showing that real-time performance is possible for large spatial kernel sizes and image resolutions. ... Read more

The bilateral filter is a popular filter in image processing and computer vision. This comes from the fact that it is able to blur images while keeping the structure intact. However, the bilateral filter allows for blurring to happen across edges. This can result in halo-like effects around the edges of structures if both sides are made up of different intensities. In this paper, we propose an extension to the bilateral filter that reduces this phenomenon of blurring across edges. By giving the filter knowledge of the edges beforehand, it is possible to prevent the filter from blurring past them. When we filter a pixel, its surrounding area within the kernel is checked for edges. If a pixel within this area lies on or beyond an edge, its weight for blurring is reduced. As a consequence, pixels that lie past an edge have less influence on blurring. We show that this new edge-aware bilateral filter reduces across-edge blurring compared to the standard bilateral filter. Furthermore, when we allow a bigger range of intensities to mix, the new filter is also able to prevent the filtered image from appearing washed out, unlike the bilateral filter. ... Read more

Traditional bilateral filters, effective in 2D image processing, often fail to account for the 3D structure of meshes, leading to artifacts in texture filtering. This thesis introduces On-Mesh Bilateral Filtering, a novel method that adapts the bilateral filter to work with non-contiguous texture mappings by incorporating 3D spatial distances and face adjacency information into the filtering process. The On-Mesh Bilateral Filter combines mesh surface sampling techniques with heat geodesic distance calculations to create a geometry-aware kernel that achieves more accurate and context-sensitive smoothing operations, respecting both the mesh topology and texture space properties. This paper hopes to encourage further research in the area of geometry-aware texture filters. ... Read more

Unlike traditional blur filters, the bilateral filter exhibits non-linear blur behaviour as its kernel size increases. This atypical blur behaviour makes it challenging to find a good σr . This paper investigates the underlying reasons for this behaviour and proposes methods to align the bilateral filter’s blur scaling linearly with its spatial filter size. Using local frequency analyses to quantify blur levels, we introduce an approach that finds the best σr through iterative search. Results demonstrate that the pro- posed method effectively counters the atypical blur behaviour. However, the proposed method does not perform sufficiently when handling very large kernel sizes. The proposed method can be used to abstract away the σr parameter when seeking linear blur behaviour in the bilateral filter. Further re- search is needed to make it functional for very large kernel sizes. ... Read more

This paper introduces the Quadrilateral filter, an advanced extension of the Bilateral and Trilateral filters aimed at addressing limitations in high-gradient regions of images. While the Bilateral filter effectively preserves edges during smoothing, it struggles with intensity variations, leading to blunted image details. The trilateral filter improves upon this by incorporating local plane geometry approximations but assumes linear pixel intensity distributions, limiting its effectiveness. The proposed Quadrilateral filter utilizes curvature-based geometry approximations to enhance noise reduction, contrast preservation, artifact reduction, and image reconstruction by accounting for nonlinear pixel value distributions. The development of this filter represents the main contribution of the paper while exploring whether the established Bilateral and Trilateral filters’ performance can be further improved through curvature-based local geometry approximations. The findings demonstrate improvements in image quality and detail preservation, with broad implications for applications in image de-noising, tone-mapping, multimedia processing, and beyond. ... Read more

Shunting yards are locations next to train stations that serve as parking places for trains when they are not in operation and often contain facilities for maintenance and cleaning for passenger trains. Planning of the tasks regarding shunting trains involves routing, assignment of tracks, and scheduling tasks. This is done manually and requires a lot of effort, making it inefficient. Identifying patterns specific in the arrival times of trains at shunting yards can help to predict future train arrivals and potential delays throughout the year more accurately. This enables the alignment of staff and equipment with train arrivals, minimizing idle time and optimizing cost efficiency.
This research focuses on extracting and analyzing the arrival times of trains at shunting yards using a dataset consisting of GPS data. It conducts two algorithms to cluster the given data for each train unit within and across days to identify the same train across different days. Distributions and heatmaps of the arrival times and delays are created based on the identified train series. They are analyzed to identify patterns in train arrival times and delays across different months. ... Read more