Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

This study aims to develop novel tools for quality assessment, quantification and outcome prediction from heart ultrasound images.


Heart ultrasound scanning (transthoracic echocardiography) is a technique used to diagnose cardiovascular disease. It uses ultrasound waves to create real-time images of the heart, from which measurements and interpretations about size, structure and function can be made. It is the most common method of imaging the heart in the clinical healthcare setting due to its rapid nature and non-invasive fashion.  

A major limitation of echocardiography is the subjective nature of interpretation of the images, with significant variation between different operators, and even variation between scans interpreted by the same individual at different times. In addition, the quality of image collection is dependent on operator skill and experience. Without methods to ensure that good quality images are collected consistently, it can be difficult to make accurate diagnoses or compare to a scan the patient previously had. 

The cardiac physiologist, Clinical Scientist or Doctor performing/interpreting an echocardiogram makes many routine measurements from the scan images (graphs, still frames and moving video clips). These only cover a small proportion of the possible information that is held within the images however. Cutting-edge analysis techniques allow us to extract further useful information from the images, which may allow for new ways of diagnosing heart disease or predicting patient outcomes, such as hospitalisation. 

A 12-lead electrocardiogram (ECG) is a 10 second recording of the electrical activity of the heart. It can be used to diagnose heart abnormalities and is also useful to determine the rhythm of the patient’s heartbeat. This information helps in the acquisition and interpretation of echocardiography images.


ECHOVision aims to develop tools that may be used to improve the clinical value of echocardiographic procedures, better diagnose heart problems and to predict patient outcomes. One objective is to use retrospective echocardiography scans to develop and/or test automated assessment of the quality of echocardiographic images.

 A second objective is to development and/or test quantification and automated extraction of diagnostic values using echocardiographic images. This will allow information about the patient's heart to be calculated automatically, rather than by an echocardiographer, which can be a time-consuming process. 

The third objective is to develop and/or test quantification and automated extraction of diagnostic values using ECG tracings. 

The final objective is to identify and validate features found in echocardiography scans and/or electrocardiogram tracings that can be used to predict both the risk of cardiovascular complications and outcomes.


Echocardiogram images, associated measurements, 12-lead electrocardiogram recordings, relevant clinical parameters and outcome data will be collected retrospectively from patients already seen at Oxford University Hospitals (OUH) NHS Foundation Trust as a part of their routine clinical care in the last 10 years. The data will be taken from the hospital electronic medical records and anonymised before leaving OUH, and as such explicit consent will not be obtained.


This research is funded by the Oxford Cardiovascular Clinical Research Facility, Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK.


This research has been approved by the UK Health Research Authority (reference 19/HRA/2068).