Robots as good as radiologists at spotting breast cancer, Lancet study finds

AI could halve the screening workload and relieve the national shortage of radiologists, Lancet study finds

Robots might be as good as radiologists at spotting signs of breast cancer, a Lancet study suggests.

Researchers found that “AI-supported” mammography screenings – where machine learning replaces one medic – is as effective as two radiologists in identifying disease.

Experts said the use of artificial intelligence (AI) had the potential to almost halve the screening workload.

Currently, “double reading” of such scans by two experts is mandatory in the NHS.

However, with a national shortage of radiologists and almost one-third of posts vacant, the health service is exploring ways to make use of machine learning to check scans.

The randomised control trial, published in the journal Lancet Oncology, involved more than 80,000 women from Sweden with an average age of 54.

Researchers found that computer-aided detection could spot cancer in mammograms – X-ray pictures of the breast – at a “similar rate” to two radiologists.

Previous studies examining whether AI can accurately diagnose breast cancer in mammograms have been carried out retrospectively, where the technology assesses scans that have already been looked at by doctors.

But in the interim study, AI-supported screening was put head-to-head with standard care.

Half of the scans were assessed by two radiologists, known as standard care.

The other half were assessed by the AI-supported screening tool, followed by interpretation by at least one radiologist.

In total, 244 women from AI-supported screening were found to have cancer, compared with 203 women recalled from standard screening.

The use of AI did not generate more false positives, where a scan is incorrectly diagnosed as abnormal. The false positive rate was 1.5 per cent in the AI group and the group assessed by radiologists.

There were 36,886 fewer screen readings by radiologists in the AI-supported group compared with the group that received standard care – resulting in a 44 per cent reduction in the screen-reading workload of radiologists, said the authors.

Dr Kristina Lang, from Lund University, in Sweden, and lead author for the study, said: “We still need to understand the implications on patients’ outcomes, especially whether combining radiologists’ expertise with AI can help detect interval cancers that are often missed by traditional screening.”

An NHS spokesman said: “The NHS is already exploring how AI could help in breast screening by enabling complicated image analysis very quickly and at scale, which, if proven effective, could in future help speed up diagnosis for many women, detect cancers at an earlier stage, and ultimately save more lives.”

Dr Katharine Halliday, president of the Royal College of Radiologists, said: “AI holds huge promise and could save clinicians time by maximising our efficiency, supporting our decision-making and helping identify and prioritise the most urgent cases.

“There is a great deal of research interest in how AI could support reporting for mammograms because they are complex, requiring significant oversight and interpretation by clinical radiologists.

“The UK’s shortfall in radiologists, at 29 per cent, makes this challenging.”

Prof Fiona Gilbert, professor of radiology and head of department, University of Cambridge, said: “There are considerable manpower savings which will translate favourably to the UK to help address our workforce issues.

“These findings will help plan the testing and implementation of AI into the UK national breast screening programme.”

Steve Barclay, the Health Secretary, said: “These encouraging results again demonstrate the potential artificial intelligence has, if safely deployed, to transform health care. That’s why I recently pledged £21 million of funding to back the swift rollout of AI across the NHS.

“We’re already seeing the significant benefits of AI in action across the NHS, delivering quicker diagnosis and treatment.

“I’ll continue to explore ways we can harness the power of AI to provide better, more efficient care and ultimately bring down waiting times - one of the Government’s top priorities.”