MADRID, July 28 (EUROPA PRESS) -Artificial intelligence (AI) may soon predict the onset of atrial fibrillation, or so a study led by Massachusetts General Hospital (MGH) cardiologist and Harvard Medical School professor of medicine has suggested, Jagmeet Singh, and published in the 'European Heart Journal-Digital Health'.The paper examined a deep neural network learning model from 'Cardiologs', part of Philips' cardiac diagnostics and monitoring offering, to identify patients at risk of atrial fibrillation within two weeks of a heart attack. 24-hour ambulatory ECG, without previously documented atrial fibrillation.The study consisted of training the deep neural network to predict the short-term presence or absence of atrial fibrillation, using only the first 24 hours of an extended Holter recording.The results showed that the network was able to predict whether atrial fibrillation would occur in the near future with an area under the receiver operating curve, sensitivity and specificity of 79.4 percent, 76 percent and 69 percent. percent, respectively, and exceeded the predictive power of the ECG for atrial fibrillation.In addition, they showed a ten-point improvement compared to a reference model that used age and gender."Getting patients to early care and preventing potentially more serious outcomes could help save lives. The study demonstrates that 24-hour Holter data could be used to improve current monitoring capabilities, and brings hope to patients of high risk, as they would benefit from proactive treatment and atrial fibrillation mitigation strategies," Singh said.Therefore, the work shows the possibilities offered by artificial intelligence to predict atrial fibrillation in the short term, using the 24-hour Holter, compared to the 12-lead ECG.While the 12-lead ECG gives access to a broader view of the heart's activity over a short period of time, the 24-hour Holter provides signals of longer duration, thus extracting additional data for predictive models.Developed in collaboration with leading specialists, 'Cardiologs' technology speeds up diagnostic reporting, reduces reporting errors, and streamlines workflow for medical professionals and patient care.This allows professionals to deliver advanced cardiac care "faster and more efficiently."© 2022 Europe Press.The redistribution and redistribution of all or part of the contents of this website without your prior and express consent is expressly prohibited.