DistillerAI promises to change the way systematic reviews are done, forever.
OTTAWA, Canada (PRWEB) February 12, 2018
Researchers who have bemoaned the time and resources required to produce a high quality systematic review can now take advantage of a groundbreaking artificial intelligence-based screening assistant when using market leading DistillerSR systematic review software.
The screening assistant, called DistillerAI, was launched this week as part of the latest update to DistillerSR.
“DistillerAI promises to change the way systematic reviews are done, forever,” said Peter O’Blenis, President of Evidence Partners. “The integration of a robust AI platform represents a sea change in review methodology that will allow researchers to produce important evidence faster and with higher quality.”
DistillerAI offers several ways to improve speed and accuracy in the systematic review process. It can:
- Sort and rank references by their likelihood of inclusion
- Be the “second screener” after one pass by a human reviewer
- Conduct post-review audits to flag possible inclusion or exclusion errors
- Complete the screening process based on learning from how humans have reviewed a subset of references
To maximize efficiency, DistillerAI learns to make accurate inclusion or exclusion decisions by monitoring human reviewers in the background as they work through the screening process. Once DistillerAI has learned enough to make confident decisions, it becomes available to assist in the screening process.
To test the power of DistillerAI, Evidence Partners is currently working with a number of research groups to benchmark the performance of the tool against that of human reviewers.
"Given the time and cost associated with the review process today, we need more efficient ways to conduct systematic reviews,” said Dr. Annette O’Connor, Professor of Epidemiology at the College of Veterinary Medicine, Iowa State University. “AI is going to play an important role in the evolution of how reviews are conducted. It will help us speed up the translation of research to end-users, so we can realize the value of research findings sooner.”