Related Experiment Video
Updated: Jul 9, 2026
Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide
Published on: January 15, 2017
Impact of Artificial Intelligence-Based Triage on Stroke Workflow Metrics: A Systematic Review and Meta-Analysis
Juan Carlos Barrera Gutierrez1, Elaina Vivian2, Jimmy Shah2
1Methodist Digestive Institute, Methodist Dallas Medical Center, Dallas, TX, USA.
Background:
Artificial intelligence is an increasingly valuable tool in ischemic stroke management. This systematic review and meta-analysis evaluated the effect of artificial intelligence implementation on stroke workflow metrics.
Methods:
PubMed, EMBASE, OpenEvidence, and the Cochrane Central Register of Controlled Trials were searched for studies (2015-2025) evaluating automated large vessel occlusion detection. Pooled mean differences with 95% CIs were calculated for door-to-groin puncture, door-to-first pass, door-to-revascularization, door-to-needle, and door-in-door-out times.
Results:
Twelve studies met the inclusion criteria: one clinical trial and eleven observational studies. Artificial intelligence was associated with reductions in multiple workflow intervals, including door-to-groin puncture (-17.12 minutes), door-to-first pass (-26.55 minutes), door-to-revascularizationon (-14.55 minutes), door-to-needle (-4.44 minutes), and door-in-door-out time (-36.8 minutes).
Conclusion:
Overall, AI-based platforms appear to contribute meaningfully to stroke workflow optimization, although randomized controlled trials are still needed to confirm their effectiveness.
Level Of Evidence:
Level 2c, Systematic review of randomized clinical trials, and observational studies.