The paper is about the diagnosis of predictors of sudden cardiac death (Long QT interval, Short QT interval, left ventricular hypertrophy, ventricular tachycardia, coronary artery disease, Brugada syndrome, hypertrophic cardiomyopathy) based on the analysis of electrocardiograms. It is proposed to consider the electrocardiosignal as a spatiotemporal process and recognize it using a method used for automatic video annotation, a hybrid neural network consisting of convolutional and recurrent neural networks. The high efficiency of the proposed approach was shown.

Marat Robertovich Bogdanov, PhD in Biology, is an Associate Professor of Computational Mathematics and Cybernetics at Ufa University of Science and Technology. He also serves as Associate Professor at the Institute of Physics, Mathematics, Digital and Nanotechnology, Bashkir State Pedagogical University named after M. Akmulla. His research interests include computational biology and digital modeling. He is based in Ufa, Russia