SWIFT-Active Screener helps researchers identify important studies faster while maintaining high accuracy
A new independent study published in Annals of the Rheumatic Diseases found that SWIFT-Active Screener, an AI-powered research tool developed by Sciome, can significantly reduce the time required to review scientific literature while still identifying relevant studies.
Researchers evaluated several AI tools as part of a project supporting the European Alliance of Associations for Rheumatology (EULAR). Their goal was to determine whether AI could help streamline the process of reviewing thousands of scientific papers used to develop medical recommendations.
Across four literature reviews related to rheumatoid arthritis treatment, SWIFT-Active Screener reduced the number of articles requiring manual review by approximately 60% to 70% while identifying the vast majority of studies ultimately included in the final analyses. In one review, researchers screened about 1,200 articles instead of nearly 3,900, saving substantial time and effort while reaching the same overall conclusions as the traditional review process.
Addressing a Growing Challenge
As the volume of scientific research continues to expand, keeping up with the latest evidence has become increasingly difficult. Systematic literature reviews remain the gold standard for informing healthcare decisions, but they often require researchers to manually assess thousands of articles.
The study found that AI-assisted screening can help by quickly prioritizing the most relevant studies, allowing researchers to focus their efforts more efficiently.
Human Expertise Remains Essential
While the findings highlight the promise of AI in evidence review, the researchers emphasized that these tools are designed to support – not replace – human experts.
The authors concluded that machine learning can improve efficiency and help researchers manage the growing volume of scientific information, while stressing the need for ongoing oversight, training, and validation.
To learn more about SWIFT-Active Screener, contact us at swift-activescreener@sciome.com.
