Heart attacks are widely considered to be impossible to predict. But artificial intelligence might be changing that.
What’s happening:
- A new study conducted at Oxford University has successfully validated that artificial intelligence can help predict the probability of a heart attack
- Funded by the British Heart Foundation, the study used artificial intelligence to analyze data from over 40,000 individuals to identify commonalities that could predict heart attacks
How it works:
- The AI enabled technology focused on arterial narrowings as one of the primary indicators of risk for a heart attack
- The conclusion of the study was that individuals with no significant arterial narrowings were much more likely to experience a heart attack when compared to individuals who did have arterial narrowing
- By measuring artery fat and other risk factors, the screening technology was able to accurately predict many of the risk factors common in cardiac events
Why it matters:
- While many individuals take a CT scan as a preventive measure, there is an inability to discover small, undetectable narrowings that are happening in the arteries
- If artificial intelligence can enable new insights into CT scans, that could unlock entirely new possibilities for predictive medicine and treatments for heart attacks
By the numbers:
- The Oxford study back tested the AI enabled screening on 3,393 patients who had had CT scans in the past 8 years and found that it was accurately able to predict their likelihood of having a heart attack
- More than 45% of individuals involved in the study had their treatment plans changed due to AI generated predictive health data
- The initial outcomes from the study suggest that artificial intelligence could help prevent a heart attack up to 10 years before it happens