New machine learning methods developed by Paul Shekelle, MD, PhD, and his colleagues at Santa Monica, Calif.-based Rand Corp. may help physicians review medical research more quickly, according to a Reuters report.
With new medical research coming out each day, clinical practice guidelines have the potential to become outdated, Dr. Shekelle told Reuters. To save physicians' time, Dr. Shekelle and his team created a machine-learning method that identifies relevant research studies, so that physicians can decrease the amount of time they spend manually screening new research.
Dr. Shekelle and his team provided the machine learning model with article titles and their summaries, to determine whether it could identify relevant literature. The program had a 96 percent accuracy rate when compared to humans who screened the articles for relevance, according to the study results, published in Annals of Internal Medicine.
"Machine learning methods are very promising as a way to reduce the amount of time and effort for the literature search, which in turn should make it easier to update the systematic review, which in turn can facilitate keeping clinical practice guidelines up to date," Dr. Shekelle told Reuters.
More articles on health IT:
GAO: VA pharmacy system needs data exchange capabilities
Why the perks of working in Silicon Valley may not extend to your health
Health Catalyst names former Allina Health CFO to board: 4 notes