AI might point out atrial fibrillation

News

According to the latest study, researched have established a machine-learning technique that can point out atrial fibrillation, a possibly deadly heart condition.

The research, published in The Lancet on Friday, stated how researchers succeeded in detecting atrial fibrillation by utilizing artificial intelligence guided electrocardiography with a whooping accuracy of 90%. The technique as a success even when the patients failed to display any symptoms.

The chair at Mayo Clinic’s Dept. of Cardiovascular Medicine as well as the senior research author, Paul Friedman, stated in a press release how when patients came to them with a stroke, they wanted to know whether they had atrial fibrillation prior to the stroke as it helps them direct the treatment. He said how blood thinners work for people with atrial fibrillation in averting another stroke, but for patients without atrial fibrillation, the utilization of blood thinners would raise the threat of bleeding for no significant advantage.

The standard electrocardiography tests take just 10 seconds, however, atrial fibrillation tends to come and go without people noticing, making it notice with the EKG test.

In most scenarios, patients having atrial fibrillation must be tested using a loop recorder. This consists of monitoring, over a longer period of time, so as to identify the condition when it strikes, however, this is an expensive method.

The research consisted of about 36,000 patients, 3,051 of whom had been diagnosed with atrial fibrillation prior to the research.

According to the study, the Artificial Intelligence-guided electrocardiography technique is compatible with smart phones/watches.

The most notable form of heart arrhythmia is a trial fibrillation. It takes place when the heart beats erratically; too fast, or too slow. According to the Centers for Disease Control and Prevention, nearly 2.7 to 6.1 million people in the U.S. are suffering from atrial fibrillation.