Artificial intelligence was first used to accurately measure blood flow in a study that determined the risk of heart attack and the likelihood of death.
A study from University College London emphasized that a reduction in blood flow, which is often treatable, is a common symptom of many heart conditions.
Available non-invasive (without skin punctures) blood flow assessments, including imaging of cardiovascular magnetic resonance (CMR), so far it has been very difficult to analyze. A new type of artificial intelligence will be able to study these analyzes.
As part of the artificial intelligence training, scientists took CMR images of more than 1 thousand people. After studying artificial intelligence, doctors compared his prognosis with the opinion of qualified doctors. In addition, based on these data, the neural network has learned to predict the possibility of a heart attack or stroke even better than doctors.
Comparing the results of the blood flow study with the health results of each patient, the team found that patients with reduced blood flow had more likely adverse health effects, including death, heart attack, stroke, and heart failure.
Thus, for the first time, artificial intelligence technology was demonstrated, which allows one to better predict which patients can die or suffer serious complications than using traditional approaches.
Professor James Moon of University College London said: “Artificial intelligence is moving from computer labs to the real world of healthcare, performing some tasks better than doctors alone. “We tried to measure blood flow manually before, but it is tiring and time-consuming.”
“The predictive power and reliability of AI was impressive and easy to implement as part of everyday patient care. Calculations were performed during the scan of patients, and the results were immediately transmitted to doctors. Since poor blood flow is treatable, these predictions ultimately lead to better patient care and also give us a new understanding of how the heart works, ”the study authors emphasized.