Back to Projects

Emerging Imaging, Affordable Imaging

Detecting heart failure with point-of-care ultrasound imaging

Project ID: 2020_013

1st supervisor: Peter Weinberg, Imperial College London
2nd supervisor: Jordi Alastruey, King’s College London
Clinical Champion: Jamil Mayet, Imperial College London

Aim of the PhD Project:

  • Develop low-cost echocardiography methods that will enable screening and monitoring patients for heart failure in a primary care environment using hand-held ultrasound scanners;
  • Test these methods in phantoms, rabbits and humans (healthy people and patients with heart failure).

Project Description / Background:

Heart failure (HF) is prevalent, increasing in incidence, and fatal within 1 year of diagnosis for 40% of cases. Part of the reason for the high mortality is that HF is often diagnosed late. Definitive diagnosis currently requires echocardiography and expert examination in a hospital setting. Here we propose a project to develop low-cost methods based on pulse wave analysis that will enable screening and monitoring in a primary care environment.

When the heart contracts during each beat, it increases the blood pressure, flow velocity, and cross-sectional area in the arteries immediately connected to the ventricles. Similarly, when the heart muscle relaxes, pressure, velocity and area just outside the ventricles all decrease. These two disturbances propagate as waves down our arteries; e.g. producing the pulse that can be felt in the radial artery of the wrist.

In HF, the intensity of these waves is altered; in simplistic terms, the intensity of the first wave is decreased when the heart fails to contract adequately (“systolic HF”), and the intensity of the second wave is decreased when the heart is stiffened and has difficulty relaxing (“diastolic HF”). Six published studies show that wave intensity can be used to assess HF.

Conventionally, wave intensity is calculated by multiplying together the changes in blood pressure and flow velocity occurring over a small-time interval at a single arterial site. In practical terms, however, changes in pressure occurring on such small timescales are difficult to assess accurately other than by inserting a pressure catheter into the artery. That is too invasive and risky to be used for screening/monitoring purposes.

A more recent derivation of wave intensity works instead by using the changes in arterial cross-sectional area (or, equivalently, diameter) and blood flow velocity. In principle, those should be easier to measure in a non-invasive fashion (e.g. diameter and velocity can be measured by M-mode and Doppler ultrasound, respectively). Although that has been used in some studies, it is not very accurate because M-mode and Doppler ideally require different ultrasound beam angles, yet need to be applied to the same arterial site.

We have recently demonstrated and patented a more accurate method based solely on B-mode ultrasound imaging, in which diameter is obtained by direct measurement from the images and velocity is obtained by a form of particle-image velocimetry as described in ‘Proposed plan of work.’ Blood flows quickly – 1 m/s or more at some sites – and this method therefore requires an ultrafast ultrasound machine. These are too costly to be used in widespread screening or monitoring programmes. We have therefore proposed and patented a second method, which should be more efficient in imaging and computational terms.

The purpose of the project is to develop, optimise and implement it on the new generation of handheld, phone-connected ultrasound scanners, some of which cost under $2k. That would revolutionise HF treatment.

The project requires a student with a background in engineering/physics/mathematics who knows or is unafraid to learn programming. An interest in entrepreneurship would be an advantage.

Back to Projects