MRC DiMeN Doctoral Training Partnership: Heart Motion Abnormality Detection Using Bayesian Deep Networks

Cardiovascular diseases (CVDs) is the second biggest killer in the UK and currently, more than 7 million people are living with CVD in the country. Early identification of individuals with significant risk is critical to improve the patient quality of life and reduce the financial burden on the social and healthcare systems. A large number of CVDs lead to aortic stiffness and reduced vascular distensibility, which can be diagnosed non-invasively by analysing the patient’s dynamic cardiac imaging data from aorta. Manual assessment of these images is subjective, non-reproducible, limited to the left ventricle, and time-consuming. Statistical atlases, describing the ‘average’ pattern of the aortic dispensability over a large healthy population, can be potentially useful to identify deviations from normality in individuals. However, the integration of the existing atlases into clinical practice is inhibited by three key limitations: (i) the derived stiffness statistics are often independent of the patient’s age, gender, weight, etc. (metadata) that are essential for precise diagnosis, (ii) Being non-probabilistic, these atlases fail to provide a measure of certainty in the extracted stiffness abnormalities thus their clinical reliability is seriously hampered, (iii) they are often derived using a small number of data sets, limiting their statistical power.

To alleviate these key limitations, this proposal aims, for the first time, to develop a full probabilistic atlas to accurately evaluate aortic dispensability abnormalities by holistically integrating imaging and metadata from a large population cardiac imaging study. This project will be a novel Bayesian approach extending the recent developments in deep recurrent neural networks (RNNs). These networks provide a natural mechanism to model sequential data such as 2D video. Yet, using RNNs to model the complex dynamics of the aortic wall motion is conceptually new and evidently powerful. The motion will be modelled as the spatiotemporal (2D+t) sequence of the aorta shapes across the full cardiac cycle, extracted from cine Cardiac Magnetic Resonance (CMR) images. The atlas will be a recurrent model that, given a sequence, it will predict a probabilistic distribution function (pdf) for the next status of the aorta. More importantly, the pdf will be conditioned on the patient’s metadata. Thus by measuring the spatial deviations from the expected shape at each phase, the atlas will allow very accurate quantification of aortic distensibility abnormalities (and variances showing uncertainties) specific to the patient’s age, gender, age, ethnicity, etc.

The project will entail extensive use of high-performance NVIDIA GPUs. The company is keen to see a wider impact on the industry with a provision that many of the computational developments achieved in this project will be general enough to promote similar research in other fields.

Requirements: The candidate is expected to have a very solid mathematical background, strong programming skills (in C++/Python), and interest in high-impact research work. These must be witnessed by the applicant’s transcripts of grades and GPA. These are in addition to the official requirements that must be satisfied (2 nd upper/above, English).