An international team, including researchers from Google Health and Imperial College London, designed and trained a computer model on X-ray images from nearly 29,000 women. AI was still as good as two doctors working together. Unlike humans, AI is tireless. Experts say it could improve detection.
The current system in the NHS uses two radiologists to analyse each woman’s X-rays. In rare cases where they disagree, a third doctor assesses the images. In the research study, an AI model was given anonymised images, so that the women could not be identified. Unlike the human experts, who had access to the patient’s history, AI had only the mammograms to go on.
The results showed that the AI model was as good as the current double-reading system of two doctors. And it was actually superior at spotting cancer than a single doctor. Compared to one radiologist, there was a reduction of 1.2% in false positives, when a mammogram is incorrectly diagnosed as abnormal. There was also a reduction of 2.7% in false negatives, where a cancer is missed.
Dominic King from Google Health said: “Our team is really proud of these research findings, which suggest that we are on our way to developing a tool that can help clinicians spot breast cancer with greater accuracy.” Most of the mammograms came from Cancer Research UK’s OPTIMAM dataset collected from St George’s Hospital London, the Jarvis Breast Centre in Guildford and Addenbrooke’s Hospital, Cambridge. It takes over a decade of training as a doctor and specialist to become a radiologist, capable of interpreting mammograms.
Reading X-rays is vital but time-consuming work, and there is an estimated shortage of more than 1,000 radiologists across the UK.