Brainomix, an AI-powered medical imaging company, has published a large-scale real-world evaluation showing that its artificial intelligence (AI) imaging platform significantly increased the use of a life-saving stroke procedure and reduced critical transfer times within the UK National Health Service (NHS).

The company boasted that the prospective observational study published in the Lancet Digital Health, represents the largest assessment of stroke AI imaging to date. It analyzed data from over 450,000 patients admitted to 107 NHS hospitals over a five-year period.

Study Design & Scale

Researchers evaluated the impact of Brainomix 360 Stroke, an AI-powered imaging decision-support software, across four regional stroke networks encompassing 26 hospitals. The study compared treatment metrics before and after the software’s implementation, using data from the national Sentinel Stroke National Audit Programme (SSNAP). The evaluation sites included both primary stroke centers (PSCs), which lack on-site thrombectomy capabilities, and comprehensive stroke centers (CSCs).

The results, detailed in the Lancet publication, showed a pronounced positive effect associated with the AI software:

• Hospitals using Brainomix 360 Stroke saw a 100% increase in rates of endovascular thrombectomy (EVT), a minimally invasive procedure to remove blood clots, compared to a 63% increase at non-evaluation sites over the same period. At the individual patient level, the use of the AI software was associated with a 57% increased likelihood of receiving EVT.
• The increase in EVT was most marked in primary stroke centers. This suggests the AI tool provides the greatest benefit in hospitals with fewer stroke specialists, where it supports interpretation of commonly available scans like non-contrast CT and CT angiography.
• At these primary centers, the use of the AI software was associated with a notable 64-minute reduction in median “door-in-door-out” (DIDO) time—the critical window between a patient arriving and being transferred to a specialized center for thrombectomy.
• The study also found a higher rate of intravenous thrombolysis (IVT) at evaluation sites, with patients assessed with AI support being twice as likely to receive this drug treatment.

Implications for Stroke Care Pathways

The study indicates that AI decision-support can help standardize and accelerate stroke care pathways, particularly in the early “hub-and-spoke” transfer model. By providing rapid, consistent interpretation of brain scans, the technology appears to empower clinicians at non-specialist centers to identify EVT-eligible patients more confidently and initiate transfers sooner.

“This landmark study confirms what stroke physicians have been witnessing on their wards every day—that this AI technology is changing lives,” said Dr. David Hargroves, NHS National Clinical Director for Stroke and a study co-author. He highlighted the role of strategic UK government investment in AI research for the NHS.