CUH Logo

Mobile menu open

Artificial Intelligence Covid breakthrough will save lives

Cambridge University Hospitals NHS Foundation Trust (CUH) and the University of Cambridge are spearheading a new way of rapidly diagnosing Covid-19 in hospital, which speeds up the correct medical intervention and helps save lives.

A team from the University of Cambridge is developing artificial intelligence (AI) algorithms to support doctors to decide what action to take on the front-line, such as giving oxygen and medications, before patients reach a critical stage.

Dr Effrossyni Gkrania-Klotsas, CUH consultant in infectious diseases, suggests that this will ultimately lead to individualised treatment plans. She said:

I firmly believe that this will result in improved patient triage and opportunities for research into personalised medicine, with the right treatment given at the right moment for each patient.

Dr Effrossyni Gkrania-Klotsas

In addition to assisting with diagnosis and prognostication, the AI will assist in predicting Intensive Care Unit (ICU) capacity and staffing requirements.

Many of the team have been working on the project in addition to their usual duties and everyone involved is quick to stress the importance of the unique working relationship between partners.

Professor Carola-Bibiane Schönlieb
Professor Carola-Bibiane Schönlieb

Carola-Bibiane Schönlieb, professor of applied mathematics, head of the Cambridge Image Analysis group and co-director of the Cambridge Mathematics of Information in Healthcare Hub (opens in a new tab) (CMIH) at the University of Cambridge, said:

A shared altruistic desire to support efforts during the pandemic has brought together AI specialists and clinicians at a scale that is truly unique for the benefit of patients.

Professor Carola-Bibiane Schönlieb

Dr James Rudd, honorary consultant cardiologist and senior lecturer said:

This exciting collaboration should help us to identify patients most at risk in this pandemic but also provides a template for use against other common diseases, like heart attack and stroke.

Dr James Rudd

The algorithm draws on the National COVID-19 Chest Imaging Database (NCCID) containing CT scans and X-rays from more than 10,000 patients from across the UK during the course of the pandemic.

The database was brought together by NHSX, a unit tasked with driving the digital transformation of care in the NHS, and access to it has been extended to hospitals and universities across the country. CUH consultant cardiothoracic radiologist, Dr Judith Babar, instigated the collaboration’s involvement in the NCCID project, ensuring the group was among the first to be able to utilise the data.

Professor Evis Sala
Professor Evis Sala

CUH consultant radiologist and professor of oncological Imaging at the University of Cambridge School of Clinical Medicine, Professor Evis Sala, said it was another example of clinicians utilising AI for the benefit of patients.

Professor Sala said: “Access to the database has been given to hospitals and universities across the country which are using the images to track patterns and markers of illness.

“In our case, we are developing an algorithm which gives us early insight into whether a patient is likely to deteriorate and the opportunity for early intervention to change the course of their disease. She added:

This has the potential to save lives and is precisely the initiative we need to ensure we are better prepared and more responsive for future pandemics

Professor Evis Sala
Collaboration team on Zoom call
Collaboration team has daily virtual meetings

The collaboration is supported by a number of partners including DRAGON (a new Innovative Medicines Initiative (IMI) project), Intel, AstraZeneca, and CMIH. Further information may be found on their website (opens in a new tab).

The NCCID is a collaborative effort between NHSX, the British Society of Thoracic Imaging (BSTI), Royal Surrey NHS Foundation Trust and Faculty (a London-based AI company). Scans in the library are provided by 18 trusts, including CUH, and are stripped of any identifying patient details before they are submitted. More information can be found here (opens in a new tab).