A clinical trial found that artificial intelligence (AI) breast screening solution, Mia, could assist radiologists with cancer detection in mammograms.
The study by British machine learning startup Kheiron Medical Technologies indicated that using its Mia solution alongside a human radiologist did not make a significant impact on the performance of breast cancer screening services.
In Europe, double reading screening is considered the gold standard in cancer detection. This means screenings are read by two human radiologists who each decide independently whether a woman should be recalled for further assessment.
Mia was used to replace one of two radiologists in 40,588 mammograms from three NHS breast screening centres between 2012-2019. Of these cases, 40,230 had a normal outcome and 358 were biopsy proven cancers.
The study found that when one human radiologist was replaced with Mia, the overall double reading recall rate was 4-5% and the cancer detection rate was 8.4 per 1,000 women, compared to a recall rate of 4-5% and cancer detection rate of 8.5 per 1,000 women when two human radiologists worked together on the same case sample.
Mia and the independent human readers agreed whether to recall women in 81.9% of cases.
Results of the study were presented at the Radiological Society of North America’s (RSNA) annual conference this week.
WHY IT MATTERS
Researchers concluded that using Mia together with one human reader could help to address the radiology workforce crisis in breast screening, and provide a solution for dealing with the backlog of cases due to COVID-19. Research organisation Breast Cancer Now says there are more than one million women in the UK who have not been screened because of the pandemic.
THE LARGER CONTEXT
A larger clinical study of Mia being led by an independent research organisation will be published shortly, consisting of more than 250k mammograms from seven sites in two countries.
ON THE RECORD
Peter Kecsekmethy, cofounder and CEO of Kheiron Medical said: “It is important that Mia is not beating a reader in our study, Mia is supporting, being combined with the radiologists to create a solution that maintains the standard of care for women who are being screened.”
Dr Nisha Sharma, director of breast screening and clinical lead for breast imaging, Leeds Teaching Hospital NHS Trust said: “What we have presented today represents the next step in what we need to see for AI - how it will actually impact the real world.”
Dr Jonathan James, consultant breast radiologist, Nottingham NHS Trust said: “We know that double reading breast screening means better outcomes for women, but we also have a growing workforce crisis, so we need to find a way to solve that problem and keep offering women the best service possible when it comes to screening for breast cancer.”