ADAPTS [COVID-19]
Research type
Research Study
Full title
AI Diagnostic And Prognostic Tools for Sonography in Covid19 and other lung conditions at point-of-care
IRAS ID
284236
Contact name
Claudia Wheeler-Kingshott
Contact email
Sponsor organisation
University College London
Eudract number
2017-003008-30
Clinicaltrials.gov Identifier
Z6364106/2020/06/67, UCL Data Protection Registration Number
Duration of Study in the UK
1 years, 6 months, 31 days
Research summary
COVID-19 is currently diagnosed using lab tests. There are indications that ultrasound scans (US) can detect lung involvement in COVID-19 patients, but it is rarely used in the UK because it needs specialists to interpret. We propose to record lung US (LUS) of COVID-19 patients in UK hospitals and demonstrate that these can really help doctors make their decisions. We will develop a computerised method (ADAPTS) using the latest artificial intelligence approach to recognise whether somebody has COVID-19 pneumonia, its severity and predict the risk of clinical deterioration. Finally, we will show that the ADAPTS software is capable of differentiating COVID-19 from other forms of lung diseases. The ADAPTS software will make it possible for emergency clinicians and non-experts like GPs and ambulance paramedics to use this safe technique to help make immediate decisions such as deciding whether to admit to hospital, transfer to intensive care units or discharge.
REC name
North East - Tyne & Wear South Research Ethics Committee
REC reference
20/NE/0179
Date of REC Opinion
1 Jul 2020
REC opinion
Further Information Favourable Opinion