Viz.ai, the AI-powered stroke care coordination company, presented data at the American Heart Association’s International Stroke Conference in early 2026 showing a 44% reduction in door-in-door-out time, the interval clinicians use to measure how quickly a patient with a suspected large vessel occlusion stroke is evaluated, coordinated, and transferred to a comprehensive stroke center. The study, led by Caezar G. Jara at Adventist Health plus Rideout, evaluated the platform’s real-world impact at a regional primary stroke center inside a hub-and-spoke referral network.
Why door-in-door-out time matters
For patients with a large vessel occlusion, a blood clot blocking one of the brain’s major arteries, every minute of delay before mechanical clot removal is associated with worse odds of functional recovery. Regional and rural hospitals typically lack on-site neurointerventional teams, so patients diagnosed with an LVO have to be transferred to a comprehensive stroke center, and the handoff between the two facilities is one of the most time-sensitive and error-prone steps in stroke care. The American Heart Association’s Get With The Guidelines program has set specific DIDO benchmarks that hub-and-spoke networks are expected to hit, and missing them has been directly linked to worse 90-day disability outcomes in prior research cited by Viz.ai and Endovascular Today.
What the AI actually changed
According to the study results, the biggest gains came from two specific chokepoints in the transfer process. Time from CT angiography completion to LVO detection fell 84%, and the time it took for the care team to be notified once a clot was detected dropped from 45 minutes to just seven minutes. Combined, these workflow improvements produced a nearly 90-minute reduction in DIDO time relative to the hospital’s own historical performance, bringing it into alignment with AHA transfer benchmarks, according to Viz.ai’s announcement and coverage by Neuronews International.
Part of a broader push to prove AI pays for itself
The Adventist Health plus Rideout data followed a separate set of Viz.ai-sponsored studies released in early 2025 examining both patient outcomes and hospital economics, part of a deliberate company strategy to generate the kind of real-world evidence that skeptical hospital administrators and payers increasingly demand before adopting AI tools system-wide. That skepticism has grown alongside broader concerns, flagged by ECRI in its 2026 patient safety report, that many AI clinical tools have not been rigorously tested against actual outcomes despite widespread FDA clearance. Viz.ai’s emphasis on multicenter, AHA-conference-presented data is, in part, a direct response to that criticism, an attempt to show measurable transfer-time and outcome improvements rather than relying on marketing claims alone.
Competitive pressure in stroke AI
Viz.ai operates in an increasingly crowded stroke-detection AI market that includes RapidAI and Aidoc’s broader triage platforms, and hospital systems now frequently run head-to-head evaluations before selecting a vendor. Some stroke neurologists have cautioned publicly that AI detection tools are only as useful as the human workflow built around them, an algorithm that flags an LVO in seconds provides no benefit if the care team notification process remains slow, which is exactly the bottleneck Viz.ai says its platform targeted at Adventist Health plus Rideout. That nuance matters for hospitals evaluating vendors: the value proposition is coordination and notification speed as much as raw detection accuracy.
What’s next for stroke AI adoption
With reimbursement models increasingly tied to measurable outcomes, expect more hub-and-spoke hospital networks to demand DIDO and 90-day disability data before signing multi-year AI contracts, rather than accepting detection-accuracy statistics alone. Viz.ai has signaled it will continue publishing site-specific real-world results, and rival vendors are likely to follow with their own comparable data as health systems standardize around AHA benchmarks as the metric that actually determines contract renewals. For rural and regional hospitals without on-site neurointerventional capability, this generation of transfer-optimized AI tools may end up mattering as much for closing the urban-rural stroke-outcome gap as any single diagnostic breakthrough. Payers themselves are also starting to weigh in: several regional Medicaid managed-care plans have begun asking hospital networks applying for value-based stroke-care contracts to submit DIDO and 90-day disability figures as supporting evidence, a shift that would have been unthinkable just two or three years ago when AI vendor claims were rarely subjected to payer-level scrutiny. For Adventist Health plus Rideout specifically, the next test will be sustaining the 44% improvement outside the conditions of a closely monitored study period, since hospital quality-improvement initiatives often see gains erode somewhat once the novelty and extra attention of a formal evaluation fade.
