AI system from Google is more accurate than doctors in diagnosing breast cancer from mammograms, study from UK and US finds
An Artificial Intelligence (AI) system from Google is reportedly more accurate than doctors in diagnosing breast cancer from mammograms.
The study in the journal Nature reported that researchers from the UK and US, including Google Health and Imperial College London, designed and trained a computer model on X-ray images from nearly 29,000 women.
It found that the AI system was still as good as two radiologists working together. At the moment, the NHS uses two radiologists to analyse X-rays. In rare cases where they disagree, a third doctor assesses the images. In the United States by comparison, only one radiologist reads the result.
AI cancer detection
But according to the study, the AI algorithm outperformed six radiologists in reading mammograms.
Breast cancer affects one in eight women globally, and the hope is that AI has the potential to improve the accuracy of screening for the disease.
At the moment, radiologists miss about 20 percent of breast cancers in mammograms, the American Cancer Society was quoted by Reuters as saying.
Half of all women who get the screenings over a 10-year period also have a false positive result, which can be highly distressing for the patient.
As part of the study, the UK and US researchers trained the AI system to identify breast cancers on nearly 29,000 mammograms. Unlike the doctors who have access to patient data, the AI system was only given anonymised images (with no case history), so that the women could not be identified.
They then compared the AI system’s performance with the actual results from a set of 25,856 mammograms in the United Kingdom, and 3,097 from the United States.
UK vs US
The study showed the AI system could identify cancers with a similar degree of accuracy to expert radiologists, while reducing the number of false positive results by 5.7 percent in the US-based group and by 1.2 percent in the British-based group (which seems to suggest the British method of using two radiologists is a better system).
The AI system also cut the number of false negatives, where tests are wrongly classified as normal, by 9.4 percent in the US group, and by 2.7 percent in the British group.
This suggests that the AI system was as good as the UK current double-reading system of two doctors. And it is superior at spotting cancer compared to the United States single doctor method.
“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,” Dominic King from Google Health was quoted by the BBC as saying.
Put your knowledge of artificial intelligence to the test. Try our quiz!