Principal Investigator: Professor Daniel Rueckert
Institution: Imperial College London
Collaborating lead –
Dr Giacomo Tarroni, City, University of LondonTags: 40119, cardiovascular, featured, Imaging, Machine Learning
Collaborator: Dr Jinming Duan, University of Birmingham, UK
The research proposed here is funded by an EPSRC programme grant which aims to develop machine learning approaches for the improved acquisition, reconstruction, analysis and interpretation of cardiovascular MR images (cMRI). CVD causes more than a quarter of all deaths in the UK (155,000 deaths pa). The cost to the UK of premature death, lost productivity, hospital treatment and prescriptions relating to CVD is estimated at 19bn GBP each year, with healthcare costs alone totalling an estimated 8bn GBP (source: BHF). cMRI offers the unique advantage of non-invasive and non-ionizing multi-parametric assessment of cardiac structure and function. This opens an exciting and timely opportunity to use engineering and computer science innovations to transform MRI into a smart and diagnosis-aware imaging modality, allowing the direct reconstruction of much richer, quantitative information. Furthermore, recent breakthroughs in machine learning, including the success of techniques such as deep learning, have shown the potential human-like performance in high-level cognitive tasks that have been previously deemed too complex for automated solutions. Coupled with the emergence of large-scale clinical datasets (e.g. UKBB) it is now becoming possible to fully leverage the benefits of machine learning techniques. The combination of these technologies offers the potential for the development of smart, integrated image acquisition, analysis and interpretation approaches for cMRI for a far more accurate, reproducible and objective quantification of CVD which can significantly boost human performance in clinical decision making.
Last updated Feb 25, 2020