In an industry first, an AI algorithm has been used to instantly and accurately assess patients’ blood flow, acting as a risk prediction tool for major adverse CV events.
James Moon and colleagues at University College London and Barts Health NHS Trust wrote in Circulation that while blood flow is a known predictor of CV risk, historically it’s been a difficult metric to measure. Cardiovascular magnetic resonance (CMR) is one of the more popular noninvasive assessments, but even high-tech imaging fails to capture scans that are accurate enough for a prognosis.
“Artificial intelligence is moving out of the computer labs and into the real world of healthcare, carrying out some tasks better than doctors could do alone,” Moon said in a release from UCL. “We have tried to measure blood flow manually before, but it is tedious and time-consuming, taking doctors away from where they are needed most.”
Moon and his team, whose research was funded by the British Heart Foundation, developed an AI model from more than 1,000 routine CMR scans of patients who visited St. Bartholomew’s Hospital or the Royal Free Hospital. The researchers used a new automated AI technique to analyze the images, allowing them to more quickly quantify blood flow to the heart muscle.
Comparing AI-generated blood flow results to each patient’s health outcomes, Moon et al. found that patients with reduced blood flow as measured by the algorithm were more likely to experience adverse health outcomes including heart attack, stroke, heart failure and death.
“The predictive power and reliability of the AI was impressive and easy to implement within a patient’s routine care,” co-author Kristopher Knott said in the release. “The calculations were happening as the patients were being scanned, and the results were immediately delivered to doctors. As poor blood flow is treatable, these better predictions ultimately lead to better patient care, as well as giving us new insights into how the heart works.”