###Ultrafast analysis of atherosclerotic plaque stress using in vivo imaging, computational modelling and machine learning for more accurate coronary artery disease risk stratification
Project ID: 2228bd1010 (You will need this ID for your application)
Research Theme: Healthcare Technologies
UCL Lead department: Mechanical Engineering
Lead Supervisor: Ryo Torii
Project Summary:
Coronary heart disease is associated with more than 10% of mortality worldwide. It is typically caused by build-up of an atherosclerotic plaque in the vessel wall which may rupture, leading to flow obstruction and myocardial ischemia. Ruptured plaques have a common phenotype consisting of a lipid pool that is covered by an inflamed thin fibrous cap, called Thin Cap FibroAtheroma (TCFA). However, only 5% of TFCAs (typically detected by intravascular imaging) rupture, causing a heart attack. Therefore, accurate identification of plaques at risk is of utmost importance as this can determine patient management and justify aggressive therapies using endovascular devices (e.g. stents) or novel pharmacotherapies.
Computational solid mechanics modelling is shown to provide additional prognostic information that enables more accurate risk stratification. However, plaque stress and rupture risk estimation has been realised only in limited cases because of the computational-resource-intensive nature of solid mechanics computations, in association with the geometrical complexity of the model. A technological breakthrough is therefore needed to accelerate the process to a clinically-viable level.
The overarching aim of this project is therefore to develop an ultra-fast method to analyse solid mechanical stress in coronary atherosclerotic plaques towards patient-specific prediction of plaque rupture and clinical events such as myocardial infarction or even death.
We seek highly motivated students wanting to make a significant contribution to the society through healthcare research, with the knowledge and expertise in computational modelling and machine-learning.
In a close collaboration with Barts Heart Centre, one of the largest European centres treating coronary heart diseases, the student will streamline the process of conventional solid mechanical plaque stress analysis, followed by the generation of large-scale training dataset and acceleration of computational process using machine learning. The developed method will then be verified using existing clinical trial data set of approximately 300 patients with atherosclerotic disease.