Researchers at Oxford have developed a groundbreaking AI system that can forecast heart failure up to five years prior to the onset of any symptoms. This innovative technology analyzes CT scans of the heart, focusing on alterations in the surrounding fatty tissue. These subtle indicators may reveal early signs of inflammation that are not detectable by the naked eye. The AI model was trained and validated using data from over 70,000 patients. Initially, it learned to identify specific changes in the tissue and subsequently correlated these markers with the occurrence of heart failure in individuals. As a result, the algorithm has demonstrated an impressive ability to recognize early risk patterns and assess the likelihood of heart failure in undiagnosed patients, achieving an accuracy rate of 86%. Notably, the findings are particularly critical for high-risk patients, who are identified as being 20 times more likely to experience heart failure compared to those classified as low-risk. Over a five-year period, this translates to an estimated probability of 1 in 4 for developing the condition.
Informational material. 18+.
