Inferring cardiac contractility using non-invasive methods

Cardiac contractility is a key predictor of cardiac function and a prognostic indicator of heart disease outcomes. Current methodologies for estimating cardiac contractility are, at best, unreliable, and at worst, flawed. Fundamentally, cardiac contractility is an intrinsic property of cardiac tissues and is underpinned by mechanisms that enhance the ability of the muscle to generate force. The slope of the end-systolic pressure-volume relation (ESPVR) is a commonly used metric for quantifying cardiac contractility. Our recent work has demonstrated that this metric is fraught with error because the slope of this relation depends on the loading experienced by the heart or its tissues.

In this project, a method for inferring cardiac contractility will be developed that is focused on specific components of a work-loop. The theoretical framework for the method will be developed using experimental data on ex vivo isolated cardiac tissues using our own purpose-built devices. The method will then be applied to the in vivo whole heart using pressure and 3D echocardiogram data. Machine learning algorithms will be employed to identify key timing events in the cardiac cycle predicted by the theoretical framework. The performance of the new method will be compared to current methods using the ESPVR.

• Skills in computational and mathematical modelling
• Interest in heart tissue experiments
• Motivation to learn and be challenged
  

Page expires: 4 May 2025