Cardiovascular diseases are the leading cause of death globally with 18 million deaths/year. Ultrasound is an inexpensive, non-invasive imaging technique without ionising radiation and so is widely used in many fields of medicine. In the diagnosis and assessment of cardiovascular diseases, ultrasound is used for imaging the blood flow and measuring the blood velocity. Unlike traditional Doppler ultrasound which can only provide the blood velocity component parallel to the ultrasound beam, novel methods such as echo-PIV can provide 2D, or even 3D, velocity vectors, therefore enabling more accurate assessments of disease. Of particular concern are the flows through diseased or prosthetic heart valves which produce jets into the ventricle. The jets can destabilise the ventricular diastolic vortex which is hypothesized as an indicator of poor prognosis with patients more likely to develop heart failure. The use of echo-PIV for accurately resolving intraventricular haemodynamics is then desirable; however, in many patients the depth of the heart results in significant signal attenuation and so poor velocity measurements of the blood there.
This project will explore the use of artificial intelligence (AI) to enhance ultrasonic velocity measurements to overcome poor signal-to-noise and produce spatiotemporal super-resolution imaging. The project will focus on one or more of the following ideas: solving inverse Navier-Stokes problems in which the domain boundary and boundary conditions are unknown; Physics Informed Neural Networks (PINNS); use of decorrelation in the velocity measurements to estimate turbulence intensity to select the governing equations; quantification of uncertainty in the blood velocity results.
This project is built on the principles of Accountable, Responsible and Transparent (ART) AI:
• Accountable: Maximising the use of ultrasound data using physics to achieve velocity imaging at previously impossible depths and accounting for the methods used, and the uncertainty in the results, to clinicians and ultimately patients.
• Responsible: The aim of the project is to develop algorithms for measuring blood velocity which can be used clinically for diagnosis of cardiovascular diseases and patient prognosis. Cardiac flows are particularly challenging but other applications include carotid artery stenosis, aortic aneurysms and flows within blood contacting medical devices.
• Transparent: Embedding physics in AI, provided it is done correctly and in a fully interpretable way, should be a transparent approach to image enhancement which minimises the use of problem specific fitting constants and other black-box requirements.
The student will become an expert on the use of AI in engineering, with a particular focus on the use of AI in medical imaging and fluid dynamics. Through the PhD programme the student will learn state-of-the-art AI techniques and will apply them to this worthy problem. In addition to the technical aspects the social implications of the work will be considered, including the diagnostic and prognostic potential of the methods.
This project is associated with the UKRI Centre for Doctoral Training (CDT) in Accountable, Responsible and Transparent AI (ART-AI). The ART-AI CDT aims at producing interdisciplinary graduates who can act as leaders and innovators with the knowledge to make the right decisions on what is possible, what is desirable, and how AI can be ethically, safely and effectively deployed. We value people from different life experiences with a passion for research. The CDT's mission is to graduate diverse specialists with perspectives who can go out in the world and make a difference.
Applicants should hold, or expect to receive, a master's degree or first or upper-second bachelor's degree in a relevant subject. Desirable qualities in candidates include intellectual curiosity, a strong background in maths and programming experience.
Informal enquiries about the project should be directed to Dr Kate Fraser on email address [Email Address Removed].
Formal applications should include a research proposal and be made via the University of Bath’s online application form. Enquiries about the application process should be sent to [Email Address Removed].
Start date: 2 October 2023.
Continue with Facebook
