In a first-of-its-kind trial in England, artificial intelligence (AI) software is to be used within breast cancer screening services at Leeds Teaching Hospitals NHS Trust.
The Leeds Investigation of Breast screening AI (LIBRA) study aims to generate evidence for the safe use of AI and investigate if it could increase cancer detection rates, reduce unnecessary patient recalls, and ease workforce pressures.
The NHS is currently facing a 30% shortfall in clinical radiologists, leading to longer waiting times and potentially-worse health outcomes, according to the Royal College of Radiologists.
To help tackle the problem, the trust will deploy an AI reader – Mammography Intelligent Assessment (MIA) – alongside two human readers to analyse mammograms for tell-tale signs of cancer.
Check and check again
Under the current standard of care, every mammogram is seen by two human readers. If they disagree on their readings, a third reader or readers will review the mammogram film in a process known as arbitration before a decision is made whether to recall the patient or not.
In the LIBRA study, if the two humans and the AI reader agree a mammogram is normal, the patient will receive the all-clear. If any of the human or AI readers disagree, clinicians will carry out the arbitration before making a decision on recall.
The LIBRA study is designed to be safe and non-invasive and, if successful, could lead to an AI reader replacing one of the two human readers in the standard model of care. This could free up clinicians to spend more time with patients, speed up screening, and increase the numbers of women seen by the service.
Dr Nisha Sharma is a consultant breast radiologist at the trust, director of the breast-screening programme for Leeds and Wakefield – which screens up to 50,000 women every year – and chief investigator for the prospective study.
The power of technology
She said: “The LIBRA study is important because AI will help us to sustain a high quality service in the future.
“Breast screening isn’t perfect and it is hoped that the LIBRA study will lead to improvements in reviewing mammograms.
“Cancers that present between screening rounds are often quite aggressive and the reason these can be overlooked is because they are not clearly visible on a mammogram.
“AI works differently to humans and by combining human expertise with technology advances this could lead to earlier detection of cancers and impact positively on women’s lives.
“We will be the first breast screening unit in England to do a prospective study using AI. This is another exciting example of Leeds Teaching Hospitals NHS Trust leading the way in innovation.”
An earlier retrospective study using large anonymised datasets found the AI reader showed ‘non inferiority’ to human readers and could help reduce workloads by a significant amount.
A final decision
Kheiron Medical Technologies, the UK company behind the AI technology, has trained Mia to review mammograms for markers such as masses, architectural distortions or calcification and suggest a simple ‘yes or no’ to recall. The final decision rests with the radiologists as they carry the clinical risk.
Simon Harris, senior project manager at Kheiron, said: “The NHS does not have enough radiologists, which is leading to delays in readings and diagnoses and workforce burnout.
“To solve this problem, we are looking at how AI software could replace the second human reader and release that person to do something more clinically urgent.
“We want to demystify what AI is. It is just a piece of software that’s very good at picking up certain patterns that it recognises in images and coming back with an output.”
Leeds Teaching Hospitals NHS Trust is recruiting nearly 7,000 women for the study, which is being funded as part of NHS England’s £100m AI in Health and Care Award to accelerate the most-promising AI technologies for health and social care. It is set to begin later this year.