The abstract is read at the publisher; the summary is JClub's.
Active learning–assisted title/abstract screening for EULAR SLRs can reach ≥95% recall after screening 35% of records, skipping about two-thirds.
- Why it matters: EULAR evidence synthesis is slowed by the large number of records to screen, creating a need for stopping rules that preserve sensitivity while cutting workload.
- What they did: Simulation on 40 development and 5 validation datasets (202,177 records) from EULAR SLRs used ASReview active learning models (h3, u3, u4) and time-to-event analyses to derive stopping criteria.
- The result: The u4 model was optimal, with recall targets met in validation; workload dropped by 67.1%, and an alternative rule used 26% screened plus a 10% irrelevant streak.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.