The West London Cardiac Ultrasound Development Programme
Research type
Research Study
Full title
The West London Cardiac Ultrasound Development Programme: A database of images to improve ultrasound scanning of the heart.
IRAS ID
272896
Contact name
Matthew Shun-Shin
Contact email
Sponsor organisation
NIHR Imperial Biomedical Research Centre
Duration of Study in the UK
1 years, 2 months, 1 days
Research summary
This study is aimed at improving the access and quality of cardiac ultrasound (echocardiography) using new technologies such as machine learning / artificial intelligence, and new acquisition techniques.
To support the development and testing of these new techniques we will need to acquire and interpret sets of images using both the traditional methods and the novel methods to compare the image quality and derived interpretation.
There are two strategies by which we aim to improve echocardiography.1) Increased access for patients in both the developed and developing world. The development of handheld echocardiography devices costing under £5,000 has expanded echocardiography availability in acutely unwell patients, yet such scans can still only be performed by cardiologists. Care would be improved if we could enable physicians to rapidly rule out cardiac conditions immediately at the point of care without a specialist being immediately available. We will apply machine learning / deep learning techniques to echocardiography images to make and aid in the diagnosis of cardiac disease.
2) Improving the quality of the acquired imaged. The availability of greatly increased processing power allows for “ultrafast” ultrasound acquisition. This technique, using standard ultrasound probes and currently available machines, allows for a substantial increase in frame-rate and resolution.
We will recruit patients in whom an echocardiography study has been requested. This may include a hand-held echocardiogram or a departmental study. We will ask the patients if we can acquire additional images (which will take up to 15 minutes) and use their clinically acquired images to support their study. We will not make any changes to their treatment or therapies.
We aim to recruit 700 patients at the hospitals within the Imperial College Healthcare NHS Trust. This study is funded by the Imperial Biomedical Research Centre.
All code, networks, and the anonymised HHE training data collected will be made freely available under a permissive license.
REC name
West of Scotland REC 4
REC reference
19/WS/0193
Date of REC Opinion
29 Jan 2020
REC opinion
Further Information Favourable Opinion