A semiautonomous artificial intelligence algorithm is showing promise in reducing false positives, unnecessary procedures and medical expenses, according to new research published Wednesday.
Scientists developed the deep learning algorithm using a set of over 123,000 digital mammograms (including 6,100-plus cancer cases). They then put the system to work in a simulation study to determine whether AI could rule out low-risk cases, allowing radiologists to focus their attention on more-concerning scans.
The technology proved potent, reducing screening exams requiring radiologist interpretation by nearly 42%, experts detailed in Radiology: Artificial Intelligence [1].
“At the end of the day, we believe in a world where the doctor is the superhero who finds cancer and helps patients navigate their journey ahead,” study co-author Jason Su, PhD, co-founder and chief technology officer at Whiterabbit.ai, said in a statement. “This study demonstrates that AI can potentially be highly accurate in identifying negative exams,” he added later. “More importantly, the results showed that automating the detection of negatives may also lead to a tremendous benefit in the reduction of false positives without changing the cancer detection rate.”
After developing the algorithm, Su et al. then tested it on a nonoverlapping dataset of nearly 15,000 more mammograms, including 1,000-plus showing cancer. The exams spanned from 2008 to 2017 and came from two U.S. institutions and a third in the U.K. Using the larger U.S. dataset (11,592 mammograms/101 cancers), AI reduced the number of screening exams requiring radiologist interpretation by 42%, callbacks by 31%, and benign needle biopsies by 7%.
“In conclusion, rule-out devices promise to have several benefits. The elimination of incorrect follow-up examinations and biopsies, which constitute major limitations of breast cancer screening today, benefits patients directly and is the most critical advantage of cancer rule-out technology,” the authors wrote. “Quality assurance and monitoring systems must be devised to guarantee a safe operation of rule-out devices, and further investigations are required to substantiate the benefits to patients, radiologists and the healthcare system. With these in place, rule-out devices could offer a safer and more effective alternative to improving screening than restrictive nationwide guideline changes.”
Read much more about the results—including potential limitations and how the algorithm performed on the other two datasets—at the link below. Scientists at the Washington University School of Medicine in St. Louis also co-authored the study.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.