About the project
This project aims to investigate how quantum technologies can transform MRI based dementia research, laying foundations for novel diagnostics. You'll work at the intersection of quantum engineering, neuroscience, and clinical analysis, and lead an interdisciplinary project with the potential to shape future dementia care and beyond.
Quantum technologies offer fundamentally new ways to represent and process information, enabling algorithms that may surpass classical capabilities in high-dimensional tasks such as dementia magnetic resonance imaging (MRI) analysis. This novel project investigates variational quantum circuits, quantum enhanced feature maps, and hybrid quantum/classical architectures to capture subtle neuroanatomical correlations.
Leveraging entanglement, interference, and non-classical feature interactions, the research explores algorithmic and approaches in quantum machine learning and their applicability to real-world medical imaging for early detection of dementia. By improving classification of dementia related changes in MRI, this work could lead to more accurate prognostic tools, earlier intervention strategies, and enhanced monitoring of disease progression, ultimately supporting better patient outcomes and informing clinical decision-making.
This project will design and implement cutting-edge quantum algorithms integrated with classical convolutional neural network (CNN) feature extractors. Key research components include developing quantum kernels for high-dimensional embeddings, optimising variational quantum circuits, evaluating performance under realistic noise, and benchmarking against classical models. Simulations will be performed using University of Southampton high-performance computing (HPC) resources and existing software frameworks, with select circuits deployed on real quantum devices. The project will systematically analyse scalability, robustness, and hybrid architecture performance, providing insights into novel quantum algorithmic strategies for medical image processing.
The project benefits from a rich internal MRI dataset covering healthy, at-risk, and clinical dementia cohorts, supplemented by open-source datasets for validation. The interdisciplinary supervisory team provides combines expertise in quantum engineering, algorithm development, and MRI-based dementia research.
