Doctors vs. AI : Who Can Diagnose Breast Cancer Better?

Artificial intelligence (AI) has outperformed doctors in a significant test to detect breast cancer. Researchers in the United Kingdom have found that the technology could better read and provide more accurate diagnoses from mammograms.

The NHS currently relies on two radiologists to analyze a woman's X-rays. In some cases, a single patient may need a third doctor to see the images when the first two disagree on their findings.

The new study, published in the journal Nature, looked into how AI could help improve and speed up the process to diagnose breast cancer. Researchers, including representatives from Google Health and Imperial College London, built their own algorithm that could assess X-ray images.

The team took mammograms from nearly 29,000 women provided by the Cancer Research U.K. The AI model received only random images without the identity of women unlike doctors that check patient history when identifying breast cancer. 

Results showed that the AI provided more accurate diagnoses compared to six individual radiologists. The technology performed as good as two doctors working together in reading mammograms. 

Researchers said their AI model reduced false positives, or a mammogram incorrectly diagnosed as abnormal, by 1.2 percent compared to one radiologist. The tool also helped cut false negatives, where a cancer is missed, by 2.7 percent, the BBC reported Thursday.

"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, said. 

Researchers hope AI would speed up cancer diagnosis by analyzing images by only seconds. The tool is also expected to help reduce doctors’ workload by providing continuous observation and interpreting mammograms 24/7. 

"This went far beyond my expectations,” Ara Darzi, study co-author and director of the Cancer Research U.K. Imperial Center, told the BBC. “It will have a significant impact on improving the quality of reporting, and also free up radiologists to do even more important things." 

However, researchers noted AI may not replace human experts completely in diagnosing cancer. The technology will still need one radiologist to monitor its functions. 

Artificial Intelligence Scientists continue to explore artificial intelligence and how it could improve common processes, such as diagnosing diseases. Pixabay