Because of AI-assisted decision support, 10 fewer stroke patients per 1,000 had another stroke, heart attack, or vascular death within three months.
26% fewer vascular eventsAchievedJan 2021 – Jun 2023Publishedpublished 2 yr 9 mo after the benefit ended
GOLDEN BRIDGE II (BMJ, March 2026) cluster-randomized 77 Chinese hospitals and 21,603 acute ischemic stroke patients; new vascular events at three months were 2.9% with the AI-imaging decision-support system vs 3.9% with usual care (adjusted HR 0.71), with the benefit holding through 12 months.
Hospitals, not patients, were randomized, and the intervention bundled AI image analysis with guideline-based treatment prompts, so the AI's isolated contribution isn't separable; enrollment ran January 2021 to June 2023.
- Compared with
- Hospitals with AI-CDSS vs usual care; primary endpoint composite vascular events at 3 months
- Effect
- 26% relative reduction (HR 0.71) (3.9% → 2.9%: 10 fewer events per 1,000 patients)
- Reach
- 21,603 patients, 77 hospitals, 23 provinces
- Domain
- Medicine · Classic ML
- Conflicts
- verify: CDSS vendor and funding disclosures in the BMJ paper.
- What would move the grade
- A factorial design separating AI imaging from guideline prompts would resolve attribution.
Sources
- primaryBMJ 2026;392:e085810 (press release with trial figures) (eurekalert.org)
- primaryClinicalTrials.gov NCT04524624 (clinicaltrials.gov)
- secondaryUrban Health Today summary (enrollment dates) (urbanhealthtoday.com)
Editor's notes
Worked example on /method for the achieved-vs-published rule.
Related
History
- Added.
Cite this entry
Permalink https://becauseofai.org/entry/golden-bridge-ii-stroke
“Because of AI-assisted decision support, 10 fewer stroke patients per 1,000 had another stroke, heart attack, or vascular death within three months.” Because of AI, updated Oct 22, 2026. https://becauseofai.org/entry/golden-bridge-ii-stroke