Covid imaging analysis [COVID-19]
Research type
Research Study
Full title
Computer analysis of novel coronavirus chest imaging
IRAS ID
282063
Contact name
Joseph Jacob
Contact email
Sponsor organisation
University College London
Duration of Study in the UK
5 years, 0 months, 2 days
Research summary
The UK is on the precipice of a major covid-19 outbreak threatening to overrun hospitals, schools and communities. We have moved from ‘contain’ to ‘delay’ in our approach[1]. Numbers of confirmed cases continue to rise, with Public Health England reporting 1950 positive cases and 71 deaths on March 17th 2020[2]. With reports of protective equipment such as facemasks for health care workers already in short supply, we may soon face a surge in COVID-19 infections that could overwhelm existing hospital resources[3]. At a time where early diagnosis is absolutely critical to containment, radiology services have found themselves at the centre of a debate on how to approach diagnosis of this novel coronavirus. \n\nProvided that RT-PCR test kits (currently used to diagnose COVID-19 from throat swabs) remain available in sufficient numbers - and laboratories processing the results are able to cope with the expected increased demand - CT scanning has only a limited role in COVID-19 diagnosis as outlined[4].However, should RT-PCR results continue to have a prolonged delay (24-48 hours) or be unavailable, the implementation of CT imaging as a frontline diagnostic tool, as rolled-out in China, requires urgent evaluation.\n\nOur analyses involves applying computer tools to CXR and CT imaging of patients with Covid-19 from China provided by Aladdin Healthcare Technologies Limited. We will identify features indicative of a poor prognosis. We will then test these features on a UK population of Covid-19 cases, acquired as part of the British Society of Thoracic Imaging Covid database with a control population of lung infection patients retrospectively identified from University College Hospital PACS records.\n\nThe ultimate aim of our analyses is to identify imaging features on an early CT that predict rapid disease progression. This would identify patients who need diverting to ITU at an early stage of the disease, therefore allowing date-driven triage of patients
REC name
Yorkshire & The Humber - Leeds East Research Ethics Committee
REC reference
20/YH/0120
Date of REC Opinion
26 Mar 2020
REC opinion
Further Information Favourable Opinion