One of the biggest killers in the world is breast cancer, it doesn’t only affect women men and even some children have been diagnosed and sometimes the news comes too late for treatment.
Like any cancer, the earlier the diagnosis the better and with this in mind research and testing have been completed using an AI system that can spot tumors that even some well-trained doctors missed.
The computer algorithm has the potential to save thousands of lives by slashing the number of ‘false negatives’ from a staggering 9.4 percent to 2.7 percent.
The machine was designed and trained using mammography images to teach the machine what to look for. In the main, the images used were from women who live in the UK.
The new algorithm will also prevent ‘false positives’ during the routine screening – sparing women unnecessary surgery and great stress. It correctly identified cancers better than experts.
In its infancy and construction the machine tested 25,856 unidentified women at three hospitals in the UK, and 3,097 at Northwestern Medicine, the machine proved to show far fewer cases of missed or incorrectly identified cancers.
The results published in Nature demonstrates how AI could “potentially be applied in clinical settings around the world,” said the researchers.
Dr. Mozziyar Etemadi, an assistant professor of anesthesiology at Northwestern University, Chicago, who co-authored the study, said: “This is a huge advance in the potential for early cancer detection.
“Breast cancer is one of the highest causes of cancer mortality in women. Finding cancer earlier means it can be smaller and easier to treat.
“We hope this will ultimately save a lot of lives.”
The breakthrough has been likened to “a spell-check for writing email”.
The trial found the rate of cancer was reduced from about one in eighteen to nearly one in a hundred. Breast cancer is the most common type of cancer in women globally, occurring in about one in eight women.
Currently, Mammography is the most widely used breast cancer screening tool, and not the most comfortable experience to undergo and even then, diagnosing cancer from these images is a challenge.
Overall, one in five cases are missed by radiologists – twice the rate of that in the study, which is simply horrifying.
What is more, about half of women who undergo screening over a ten-year period will experience a false positive in which cancer is wrongly suspected.
Co-lead author Scott McKinney, a software engineer at Google Health in California, said: “Computers are really good at these tasks.
“We hope someday this tool for radiologists becomes as ubiquitous as spell-checking for writing an e-mail.”
AI tools could support clinical decision-making in the future and relieve pressure on healthcare systems by reducing workload.
It is estimated more than two million women across the world developed the disease last year – with more than 600,000 deaths.