New breast cancer hope as doctors use AI to spot tumours
Hi-tech system could improve detection rates
A PIONEERING research trial in Scotland is using artificial intelligence to improve detection rates for breast cancer.
Studies carried out in Aberdeen have shown the technology can slash the time it takes for women to receive a diagnosis – and might even spot more tumours.
It could soon be rolled out across Scotland, leading to faster and better detection rates.
More than 80,000 mammogram scans obtained from women over a four-year period were studied using an AI tool and a ‘human reader’.
Results showed the technology, when assisting a radiologist, improved detection rates and could even predict future cancers, allowing treatment to be targeted earlier.
The researchers believe a project named Gemini could usher in the standard use of AI in breast cancer screening programmes north of the Border within five years.
Dr Gerald Lip, clinical director of the North East Scotland Breast Screening programme, who led the research, said: ‘This is the first time that mammography AI has shown its effectiveness on Scottish women.
‘I think this will impact on every one of the 200,000 women a year we screen in Scotland.
‘First, we will be able to show that the AI is as good as radiologists, so we don’t expect the cancer detection rate to drop. We might even expect it to go up by another 10 to 20 per cent, reducing interval cancers, which are cancers that happen between screening rounds.
‘Secondly, we will be able to reduce the turnaround time. So, on average, in some places where it takes up to 14 days, you could get the results almost within three days, reducing anxiety for the women involved.’
Last night the results were welcomed by leading UK charity Breast Cancer Now.
Dr Kotryna Temcinaite, head of research communications, said: ‘Detecting breast cancer early is crucial as the sooner it is diagnosed the better the chance of treatment being successful.’
The AI system, named Mia, is a computer programme devised by Kheiron Medical Technologies. It is taught to examine breast scans in minute detail to look for distinctive patterns and clusters that could indicate the presence of a tumour.
Mia learns to identify tumours by studying earlier scans in which cancer was correctly detected. By comparing scans of known tumours with normal images, it constantly refines its performance. There are around 4,200 new breast cancer cases a year north of the Border. It accounts for 28 per cent of all cancers diagnosed, excluding nonmelanoma skin cancer.
One motivation for using AI is to support an ageing and underpressure radiologist workforce.
Mammograms are studied by two specialists who follow strict guidelines. Dr Lip said replacing a human reader with an AI tool could reduce the burden on a declining radiologist workforce by up to 40 per cent. He said that while broadly supportive of AI, patients stressed they ‘wanted a human in the loop somewhere’.
He said one of the specialists involved in reading scans could be replaced, adding: ‘Any disagreements between the technology and the human would be looked at by an extra person.
‘If we reach a scenario where we are almost 99.99 per cent sure their mammogram is normal, then we may not need a human in the loop, but it’s a stage-by-stage process. We’re going to have to show lots of levels of safety and monitoring.’
‘Faster results would reduce patient anxiety’