黑料福利网 MSc academic award nominees
Development of NMR Methods for the Study of Marine Sediments
APPLIED PHYSICS
Each year, Europeans spend billions dredging harbors and rivers, dumping tens of millions of tons of sludge as waste. This research combines material science, fluid dynamics, and MRI technology to develop new methods for measuring sludge properties鈥攑aving the way to re-use this material as a cost-effective and sustainable building resource.
Tackling today's challenges requires engineers who can integrate perspectives beyond their own field of expertise.
Self-Learning Fuel Path Control for Heavy-Duty RCCI Engines
AUTOMOTIVE TECHNOLOGY
This thesis develops a smart calibration system for heavy-duty RCCI engines, increasing fuel efficiency while ensuring safe engine operation. It enables much faster calibration times compared to traditional methods, reducing time and costs, and supports cleaner, sustainable transport by cutting emissions and fuel use.
My goal is to power industries with engines that are as kind to the planet as they are robust.
Understanding CXCL9 regulation in pancreatic cancer cell lines: combining in silico and in vitro experiments for precision oncology
BIOMEDICAL ENGINEERING
Attracting immune cells to tumors via the expression of chemokines like CXCL9 is a promising approach to treat aggressive tumors such as pancreatic cancer. In this thesis, I investigated regulatory mechanisms of CXCL9 in pancreatic cancer cell lines to guide the development of new anti-cancer treatments with immunomodulatory effects.
It鈥檚 amazing to live in an era where we can study complex systems like the tumor microenvironment, while previously all we knew was the pain and damage they cause
Comparing Human, Human-AI, and AI-Automated Structural - Design Processes and Products for Zoning and Stabilization
BUILT ENVIRONMENT
This thesis investigates structural building design performed either manually, fully via AI, or through human-AI collaboration. An interactive AI-enabled software tool is built, and its use is evaluated with graduate students (protocol analysis). Human-AI collaboration shows most potential as AI streamlines human processes (efficiency, complexity) without diminishing creative freedom.
We demonstrate that combining the strengths of AI agents with those of humans can unlock solutions to complex design tasks - that a human or AI agent alone cannot.
From Siloed Centralized Information Systems Towards Federated Data Ecosystems: Developing AECO Data Spaces
BUILT ENVIRONMENT
This project developed a first prototype of a data space for the construction industry, enabling secure, interoperable and decentralized data sharing. By combining state-of-the-art web technologies with socio-technical system analysis, it highlights the need for standard-aligned, value-driven and nested data spaces to accelerate the transition toward a circular built environment.
To reuse materials, we must reuse knowledge: data spaces make that possible
Multi-component polymer mixtures at an interface
CHEMICAL ENGINEERING
My research improves our ability to predict the behavior of mixtures of polymers and colloidal particles, critical for food technology, drug delivery, and sustainable coatings. By understanding how these mixtures behave, we can design more stable, efficient, and eco-friendly materials used in everyday life.
Interactions are all around us.
Improving TSN Traffic Shaping with Constant Bandwidth Server Scheduling
COMPUTER SCIENCE AND ENGINEERING
Ethernet lacks support for time-sensitive applications, which prompted development of the TSN standard. TSN defines techniques that provide timing and quality-of-service guarantees for Ethernet traffic. My research showed how specific TSN components cause wasted bandwidth and higher latency. I demonstrated an alternative that preserves TSN guarantees while reducing communication delays.
A good way to have good ideas is by being unoriginal - Bram Cohen
Leveraging Contrastive Learning and Spatial Encoding for Traffic Network Prediction with Expanding Infrastructure
DATA SCIENCE AND ARTIFICIAL INTELLIGENCE
My research presents a new way to predict traffic in growing cities, even when new sensors are added to the roads. By combining clever techniques in machine learning, this approach makes predictions more accurate. This means cities can manage traffic better, making daily travel smoother for everyone.
Innovation thrives in tight corners鈥攂end, re-route, and rise stronger
Dropout Neural Network Training Viewed from a Percolation Perspective
INDUSTRIAL AND APPLIED MATHEMATICS
AI is prevalent across many vital industries and developing solid mathematical theory behind AI is imperative to tackle challenges in safety, explainability, and performance. I showed potential mathematical issues with dropout, a near ubiquitous neural network training algorithm, using the same mathematics that describes water flow over coffee grounds.
AI has potential to do enormous good, but it must be developed responsibly
Designing Equitable Healthcare: how do we design for women鈥檚 health?
INDUSTRIAL DESIGN
Cardiovascular disease is the leading cause of death among women, since women are often misdiagnosed because our healthcare systems are designed around the male body. Through a feminist design lens, I redesigned the care pathway for women with heart failure. Making healthcare more equitable, and responsive to women鈥檚 lives.
Women are being failed by the healthcare system but designers can challenge that. It鈥檚 time we redesign care pathways that truly acknowledge and improve women鈥檚 health.
Physics-Informed Learning of Neural Activity in a Brain-on-Chip Device
MECHANICAL ENGINEERING
The human brain is incredibly efficient, it even outperforms conventional computers. Brain-on-Chip (BoC) devices aim to utilize this by using living neurononal cell cultures, but they鈥檙e hard to build and study. This project develops a surrogate model that mimics BoC behavior, enabling faster, software-based development of next-generation energy efficient computing hardware.
Brains, not bytes: modelling the next generation of energy efficient computing hardware
A new organizational approach at a circular construction company
OPERATIONS MANAGEMENT AND LOGISTICS
This research helps companies transition to a circular economy by aligning their organizational structure with circular goals. It introduces practical tools to measure performance and improve communication between departments. The result is a framework that supports better decision-making and helps modular construction companies turn strategy into action.
A circular future does not start with the product, it starts with how we organize ourselves.