Artificial intelligence taketh away, but so doth it giveth. Yes, the technology exacerbates challenges in education, information, national security, and mental health. But, boy, was I reminded this week—by an obscure article detailing a cardiology conference in Madrid—what a boon it is to the medical field. In England, it has tripled the proportion of stroke patients who fully recover.
By Emma Varvaloucas
Here’s how it works:
There are three main types of strokes, all of which involve the interruption of blood flow to the brain but otherwise require different treatment—from blood thinners for lighter blockages to emergency surgery for more severe ones. When a patient arrives at a hospital, CT scans that help diagnose the type of stroke determine the care team’s next steps.
Time is crucial. Patients with strokes caused by blood clots lose 1.9 million neurons, 14 billion synapses, and 7.5 miles (12km) of brain signal-transmitting nerve fibers each minute a stroke is left untreated.
Fortunately, speed is the AI program’s superpower. It reads the scans and produces a “profusion map” report, which shows clinicians the areas of the brain that are receiving little to no blood—in only a minute.
This report reaches the entire clinical team immediately and simultaneously, streamlining a consultation process that used to progress in stages. With the team in quick lockstep, a stroke patient can begin treatment much faster—in England, a full hour faster.
In early-stage analysis of pilots run by the National Health Service (NHS), this has led to a tripling of recovery rates. Nearly half of stroke patients now recover to the point of functional independence, up from 16%.
NHS began their pilots of the AI system in 2019 and as of summer 2024 had rolled out the technology to all 107 stroke centers in England. While similar systems are also available in Europe and a number of hospitals in the United States, England, from what I can tell, is the first to reach nationwide coverage.
And this is not the end of AI’s impact on stroke care, either. Researchers are hoping that the technology could soon make it easier to predict strokes, as well as recover from them.
(Many thanks to Alexis V. for sharing)
Skip to content
