15/01/2024 | Press releases

Research study uses AI to ‘read’ chest x-rays

Research study uses AI to ‘read’ chest x-rays and help diagnose lung cancer faster

Radiologists are using revolutionary AI software to help detect abnormalities on chest x-rays as part of a new research study happening at Ipswich Hospital.

When a patient is referred by their GP for a chest x-ray, consultant radiologists will assess the image to see whether there are any abnormalities such as lesion that needs to be investigated.

As part of the LungIMPACT study, patients’ x-rays will have the added benefit of being reviewed by AI software to check within minutes if it detects any issues.

man looking at camera with computer screen of x-ray by him
Dr James Hathorn

Dr James Hathorn, consultant radiologist and principal investigator for the study at East Suffolk and North Essex NHS Foundation Trust (ESNEFT), said: “There aren’t many big clinical research studies focused on artificial intelligence to prove its real worth, so we’re really excited to be part of this study to help find clinical evidence for the benefits. We want all AI products to be properly researched and evidenced so this is an important study for the future of healthcare.”

The hope for the study – and using AI – is that it will help speed up any detection of an abnormality and then prioritise this for radiologists.

x ray of a chest
Mocked up screen of an x-ray showing where the Qure.ai programme has detected a nodule in someone’s chest.

Dr Hathorn added: “Hopefully using this technology will speed up the time to diagnosis and ensure patients who need treatment can have it as fast as possible.”

Every x-ray will still be viewed by a consultant radiologist, with AI being used in addition to the existing clinical processes for patients.

Dr Hathorn added: “This is a fantastic support tool in addition to radiologists viewing x-rays. The final decision will still always be the radiologist. The main point of the study is to show that AI could help prioritise which x-rays to report first, leaving the normal ones for later and letting the investigation of potential cancers happen more quickly.”

The study is being run by Nottingham University Hospitals NHS Trust with Ipswich Hospital one of a few national sites.

The LungIMPACT study is using Qure.ai’s software and will run until the end of July 2024 – so may view 9,000 chest x-rays taken from GP referrals.

Darren Stephens, senior vice president and commercial head of UK & Europe at Qure.ai, said: “We are delighted to welcome Ipswich Hospital to the LungIMPACT trial. It joins other NHS hospitals gathering real-world evidence of how AI can assist in detecting abnormalities on GP-referred chest x-rays to reduce the time taken to diagnose lung cancer.

“Our qXR software works by highlighting any abnormalities and prioritising the case to the radiologist. The quicker abnormalities are found, the quicker patients can gain treatment, which in the case of cancer has the potential to save lives.”

All GP-referred patients’ chest x-rays at Ipswich Hospital will go through the LungIMPACT study. However there is an opt-out option which patients can do by talking to the team when they come to hospital.

Frances Farnworth, assistant director of research and development at ESNEFT, said: “Real-world research in routine environments is really important, as it comes with all the necessary regulatory checks and approvals necessary in the NHS. This assures our patients their data stays within a secure environment.”

The study is a collaborative piece of work between the research and development support team, ICT and innovation at ESNEFT.

The LungIMPACT trial was commissioned and funded by the NHS Cancer Programme, with the support of SBRI Healthcare and the NHS Accelerated Access Collaborative.

Back to top