From the Journals

AI for Stroke: What the Data Show

Researchers evaluated artificial intelligence stroke imaging across England's health service stroke network.

Share

7 Key Takeaways

  • 1

    AI imaging software increased thrombectomy rates from 2% to 5% at evaluation sites.

  • 2

    Interhospital transfer times reduced by a median of 64 minutes with AI use.

  • 3

    A 69% adoption of AI software occurred at non-evaluation sites during the study.

  • 4

    AI use led to higher rates of intravenous thrombolysis.

  • 5

    No link between AI use and in-hospital mortality was found.

  • 6

    Residual confounding may have affected results.

  • 7

    Long-term outcomes remain unclear due to limited follow-up.

Original Source(s)

AI for Stroke: What the Data Show

Researchers evaluated artificial intelligence stroke imaging across England's health service stroke network.

  • by Kathryn Wighton

  • January 14, 2026

  • 3 min

Related Content

Patient Safety and Comfort Drive Shift in Arterial Access

For neurointerventional radiology procedures, arterial access is critical. The femoral artery has long been the go-to access point. But, say doctors at Marcus Neuroscience Institute at Boca Raton Regional Hospital, a part of Baptist Health South Florida, radial access may be better for many patients.

  • by

  • February 1, 2022

  • 3 min
How Safe Are FDA-Cleared AI Tools?

“These findings indicate that standardized efficacy, safety, and risk assessment reporting remains inadequate for FDA-cleared AI/ML devices, underscoring the need for dedicated regulatory pathways and robust postmarket surveillance to ensure patient safety."

Crisis Imaging: Mass Casualty Lessons Learned

Radiology department adaptations during unprecedented patient surge reveal preparedness strategies.