Artificial intelligence may be used to foretell a person’s chance of suffering a heart attack up to ten years in the future, according to research conducted at the University of Oxford. Researchers at the University of Oxford believe the technology has the potential to save thousands of lives while also enhancing therapy for almost half of patients. Cardiac CT scans are used to find blockages or narrowing in the arteries, and this study, financed by the British Heart Foundation (BHF), investigated whether or not AI might increase their accuracy.
Providing an accurate picture of risk to clinicians can change and potentially improve the course of treatment for many heart patients. Although over 350,000 people in the UK have a CT scan annually, many of these scans miss fatally narrowed arteries, leading to the deaths of many patients.
Over 40,000 patients with cardiac CT scans done regularly at eight different hospitals in the UK were analyzed, with a median follow-up duration of 2.7 years. After testing the AI system on an additional 3,393 patients over the course of nearly eight years, researchers found that it correctly predicted the likelihood of a heart attack. Medical professionals were then given AI-generated risk rankings for 744 patients, with 45% subsequently having their treatment plans modified.
The findings “highlight the crucial role AI-based technology may play” in predicting who would suffer a heart attack in the near future. Researchers utilized over 150,000 chest X-rays to train an artificial intelligence software to detect patterns in the pictures related to risk from severe cardiovascular disease events. A different set of roughly 11,000 people was used to test the software, and a “significant correlation” was discovered between the risk level predicted by the AI and the actual incidence of a severe cardiovascular disease event.