TU/e MSc academic award nominees

The transition towards a Large-Scale Vortex in fluid turbulence


Turbulent flows driven by convection and influenced by rotation are everywhere in nature, but they are far from understood. We discovered that a remarkably abrupt transition exists in such flows, where a large strong vortex can spontaneously emerge. This sheds light on the origin of the Earth’s magnetic field and other planetary flows.

Understanding the flow of turbulence is essential to all that happens around us: on earth and far beyond.

Towards 3D freehand ultrafast multi-perspective imaging of the abdominal aorta


When an abdominal aortic aneurysm ruptures, only 1 out of 5 patients survives. So, assessment of the aneurysm is of great importance. This can be done by Ultrasound imaging, which is non-invasive and low-cost. This research proposed a 3D freehand ultrafast multi-perspective approach to obtain high-quality images to improve rupture risk assessment.

I believe that we can improve healthcare through technical innovations in ultrasound imaging.

A Multi-Domain Approach to Thermal Comfort in Office Buildings


Thermal discomfort hinders us all but we struggle to design comfortable buildings. This research examines the impact of variables beyond temperature. It shows that thermal conditions in offices also depend on acoustics, occupancy in the building and occupant personality traits. The findings offer suggestions for future research and design practices.

Warm, warmer, hot! We are getting closer to designing thermally comfortable spaces by considering the indoor environment and its occupants.

Low CO2 partial pressure hydrogenation: Can reducible oxides bring us there?


The direct conversion of CO2 from dilute mixtures (atmosphere, biogas, etc.) to valuable small (C1) molecules is a straightforward solution to mitigating the CO2-induced climate change. We identified efficient materials to convert CO2at atmospheric concentration selectively to methane and furthered our understanding of how these materials function.

We should aim for a closed loop system in which we recycle CO2 to make fuels and chemicals instead of relying on fossil fuels.

Correlation Detective: Efficient Multivariate Correlation Discovery


In the era of Big Data, efficient data analysis methods are becoming ever more important. In this project, we developed a fast algorithm that discovers strongly correlated sets of variables. The algorithm reduces computation time by several orders of magnitude, enabling experts to find interesting patterns in much larger datasets.

Efficient algorithms will enable data-driven discovery of complex phenomena.

Improved Aleatoric Uncertainty Quantification in Medical Image Segmentation: A Framework for Modeling Complex Distributions using Normalizing Flows


A significant challenge for medical specialists entails reaching consensus on the delineation of cancerous lesions. Nevertheless, it is essential to consider their disagreement for reliable surgical planning. Our work uses state-of-the-art AI to learn and express the expert disagreement, enabling safer use of Machine Learning in the Medical domain.

I wish I knew earlier what I don’t know.

Pulse based optimization of Variational Quantum Algorithms in Rydberg systems


Future quantum computers have the potential to do a scala of things, ranging from developing new medicine to weather forecasting. My work focusses on developing quantum computer algorithms that are tailored towards current day quantum computers. These systems are the testbeds necessary to propel ourselves into this future.

When life gives you lemons, make lemonade. Let's all stay positive and do some science. (Cave Johnson)

Shape-changing food restaurant experience


This project is about transforming food in collaboration with the world-class restaurant Alchemist in Copenhagen. Edible flower shapes are made from carrots and beetroot, on which a reactive gel is printed. When served in front of the guests, the gel swells and the petals of the flower start to close.

If you want to make a change as a designer within a restaurant, it is important to understand the chefs’ values and their daily practices.

Deep reinforcement learning for solving a multi-objective online order batching problem


On-time delivery and low service costs are two important metrics in e-commerce warehousing operations. These are heavily influenced by the method used to combine orders in batches and pick them simultaneously. This work proposes a novel artificial intelligence-based approach that optimizes both objectives and gives insights in the learned method. 

AI is the new electricity; it has the potential to transform every industry and create huge economic value. The question now is, how can we leverage this technology to do good?

Organizing Climate Justice: Activist Research of a Social Innovation Initiative in the Dutch Climate Movement


The climate crisis is too complex to invent ourselves out of it. In my thesis I explore social innovation processes at Milieudefensie. I discuss paradoxes, conflicts and tensions that occur in these processes. Looking beyond technological innovation allows for a broader strategic repertoire to address the climate crisis.

Action on behalf of life transforms. Because the relationship between self and the world is reciprocal, it is not a question of first getting enlightened or saved and then acting. As we work to heal the earth, the earth heals us. (Robin Wall Kimmerer)


Patient-ventilator asynchrony detection and classification in mechanical ventilation


Mechanical ventilators are crucial for patients that cannot breathe sufficiently on their own but can lead to patient-ventilator asynchrony. In this research, we developed an algorithm that automatically is able to detects different types of ventilator asynchrony, which can be used by clinicians to improve patient treatment and reduce the length of hospital stay.

Developing a mechanical ventilator that improves patient comfort and allows the hospital staff to catch their breath.