Weekly PhD Newsletter | SIGN UP NOW Weekly PhD Newsletter | SIGN UP NOW

Augmenting Cardiovascular Magnetic Resonance Imaging with Artificial Intelligence to Automatically Characterise Heart Structure and Function in Complex Patients


   School of Computer Science

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Dr Jinming Duan, Prof Dipak Kotecha  No more applications being accepted  Competition Funded PhD Project (Students Worldwide)

About the Project

Cardiovascular magnetic resonance (CMR) imaging is a powerful diagnostic tool used in routine practice, but has inherent limitations especially in complex and coexisting cardiac diseases such as atrial fibrillation (AF) and heart failure [1]. Inter- and intra-study variability, increases in clinical demand, and a shortage of skilled staff all threaten the ability of CMR to meet the current needs of the NHS for high-quality and cost-efficient tests that can help to improve patient outcomes. Automation of manual tasks and augmentation of human expertise using artificial intelligence (AI) has the potential to drive both improved accuracy and efficiency in CMR imaging. In this project, we will utilise AI pipelines to develop advanced and fully automated approaches for the fast and accurate characterisation of heart structure and function from CMR data. We aim to improve upon traditional analysis which relies on cardiologists and radiologists to labour over each individual scan. With access to data from UK Biobank, UHB and Ultromics Ltd (a world-leading AI company specialising in using cardiac imaging for cardiovascular disease diagnosis), we will be able to train cutting-edge machine learning algorithms which enable rapid diagnosis of a wide range of heart diseases with a high level of accuracy. Such algorithms have the potential to be used not only in the NHS but globally.

 More specifically, we aim to develop end-to-end deep learning approaches for whole-heart segmentation [2] and co-registration [3,4], which will allow efficient and high-throughput quantitative analysis on cardiac structure and function from CMR images of patients with AF and heart failure, giving the ability to quantify cardiac motion over time, derive cardiac displacement, strain and strain velocity rate [5]. We will investigate feature tracking, by developing novel diffeomorphic registration [6] with realistic physical constraints as well as and Long Short-Term Memory Networks and Transformers. The proposed methods will reliably output chamber volumes, ejection fraction, mass and motion traits for diagnostic purposes. The postgraduate researcher will work in our vibrant multi-disciplinary cardAIc team [1] that encompasses machine learning and clinical staff in a world-leading and supportive research environment. The postgraduate researcher will have the opportunity to liaise with an extensive network of collaborators, attend scientific conferences, contribute to advances in clinical medical science, and carry out a pre-arranged internship in Ultromics Ltd for a period of three month time.

How to apply

Informal enquiries should be directed to Jinming Duan [Email Address Removed]

To apply, please download the application form and complete all documentation available at https://more.bham.ac.uk/mrc-aim/phd-opportunities/ and send completed applications to [Email Address Removed]

The deadline for submitting applications is 09:00 GMT Monday 16 January 2023. Please ensure that your application is submitted with all required documentation as incomplete applications will not be considered. 


References

[1] Kotecha, D. and Piccini, J.P. Atrial fibrillation in heart failure: what should we do?. European Heart Journal, 36(46), pp.3250-3257, 2015
[2] Kotecha, D., Lam, C.S., Van Veldhuisen, D.J., Van Gelder, I.C., Voors, A.A. and Rienstra, M. Heart failure with preserved ejection fraction and atrial fibrillation: vicious twins. Journal of the American College of Cardiology, 68(20), pp.2217-2228, 2016
[3] Bello, G.A., Dawes, T.J., Duan, J., Biffi, C., De Marvao, A., Howard, L.S., Gibbs, J.S.R.,Wilkins, M.R., Cook, S.A., Rueckert, D. and O’regan, D.P. Deep-learning cardiac motion analysis for human survival prediction. Nature Machine Intelligence, 1(2), pp.95-104, 2019
[4] Duan, J., Bello, G., Schlemper, J., Bai, W., Dawes, T.J., Biffi, C., de Marvao, A., Doumoud, G., O’Regan, D.P. and Rueckert, D. Automatic 3D bi-ventricular segmentation of cardiac images by a shape-refined multi-task deep learning approach. IEEE Transactions on Medical Imaging, 38(9), pp.2151-2164, 2019
[5] Stoll, V.M., Bunting, K.V., Liu, B., Townend, J., Lip, G.Y., Kirchhof, P., Steeds, R. and Kotecha, D. Improving validity of cardiac magnetic resonance imaging in patients with atrial fibrillation. EUROPACE-LONDON, 20(4), pp.iv47-iv47, 2018
[6] Karwath, A., Bunting, K.V., Gill, S.K., Tica, O., Pendleton, S., Aziz, F., Barsky, A.D.,Chernbumroong, S., Duan, J., Mobley, A.R., Cardoso, V.R. & Kotecha, D. Redefining β-
blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis. The Lancet, https://doi.org/10.1016/S0140-6736(21)01638-X, 2021

How good is research at University of Birmingham in Computer Science and Informatics?


Research output data provided by the Research Excellence Framework (REF)

Click here to see the results for all UK universities
PhD saved successfully
View saved PhDs