Breast cancer screenings are an important preventative health measure. However, up to 50% of the time these tests can produce a false positive.
Now, scientists at the Houston Methodist Research Institute have developed artificial intelligence (A.I.) software that can predict women's breast cancer risk faster and more reliably than other tests.
The software boasts 99% accuracy, meaning the risk of false-positive results (and the unnecessary anxiety that follows) would be significantly lowered.
30 times human speed
The cancer-detecting AI, which scientists have not yet named, works by reviewing patient charts and converting them into diagnostic information.
In addition to its impressive rate of accuracy, the software also produces quick results. Manual physician evaluations can take up to 60 hours. The AI, on the other hand, runs at 30 times human speed.
Beyond shortening the time it takes to effectively predict cancer risk, the AI could help reduce the number of unnecessary biopsies, said lead researcher Dr. Stephen T. Wong of the Department of Systems Medicine and Bioengineering, per Breast Cancer News.
"This software intelligently reviews millions of records in a short amount of time, enabling us to determine breast cancer risk more efficiently using a patient's mammogram," Wong said in release.
The AI software successfully and efficiently evaluated the more than 500 in just a few hours, which saved doctors over 500 hours. It also helped researchers rate patients’ probability of being at risk for breast cancer.
"Accurate review of this many charts would be practically impossible without AI," said Wong.