Study into AI's potential to save thousands from dying of lung cancer
A Nottingham respiratory consultant is co-leading a study into using artificial intelligence (AI) to diagnose lung cancer quicker, potentially saving thousands of lives each year.
Lung cancer is the deadliest type of cancer; every year in the UK, 47,000 people are diagnosed with it and more than 35,000 people die. The low survival rate is because it is often detected late (partly due to a lack of symptoms), so the chances of curing it are considerably lower.
GPs in England request around two million chest X-rays every year, most of which come back as normal but some show early indications of lung cancer. Once these signs are identified, the patient is referred for a CT scan, to help confirm or rule out lung cancer.
The qXR software – developed by Qure.ai – spots potential abnormalities on chest X-rays, and flags which need to be urgently reviewed; a patient may be immediately referred for a same-day CT scan if lung cancer is indicated.
The software’s potential is the subject of the LungIMPACT trial, jointly led by Professor David Baldwin, respiratory consultant at Nottingham University Hospitals NHS Trust, and Honorary Professor of Medicine at the University of Nottingham, and Dr Nick Woznitza, consultant radiographer at University College London Hospitals (ULCH) NHS Foundation Trust, and Clinical Academic at Canterbury Christ Church University.
In the study, the AI system will produce a secondary image for each chest X-ray with an overlay to highlight certain abnormalities. If the AI detects a problem, this will highlight the X-ray as a priority for immediate review.
The research team will evaluate whether use of this AI ‘triage’ can successfully reduce the time taken to reach a diagnosis.
The study is a collaboration between NUH, UCLH, Qure.ai and Canterbury Christ Church University. It has been funded by The Small Business Research Initiative (SBRI) Healthcare in partnership with the NHS Cancer Programme and Accelerated Access Collaborative to advance the earlier and faster diagnosis of cancer.
Prof Baldwin is Chair of the UK Clinical Expert Group for Lung Cancer and Mesothelioma, National Clinical Lead for Lung Cancer, NHSE/I, and Clinical Advisor to the UK National Screening Committee, Department of Health and Social Care.
He said: “Speed of diagnosis is critical to achieve the best outcomes in lung cancer and to reduce stress and worry for patients.
“Studies evaluating the clinical impact of AI are urgently needed to ensure the safe and effective implementation that is needed to help the NHS and our patients. Doing these studies is a significant challenge but a worthwhile one.”
Dr Woznitza said: “The quicker we pick up on any potential anomalies on a chest X-ray, the better. We review X-ray results as quickly as possible, but if this technology can help us prioritise who would most benefit from a rapid review of their X-rays, this would help improve outcomes for our patients.”
Darren Stephens, Senior Vice President and Commercial Head UK and Europe of Qure.ai, said: “Trust in healthcare AI as a tool for supporting clinical case prioritisation is growing. Real-world evidence from collaborative trials such as LungIMPACT is vital to help power confidence in digital health innovations and improve speed of cancer care for patients now and into the future.”