CNN-type myocardial infarction prediction based on cardiac cycle determination

Abstract

This paper proposes the cardiac infarction prediction based on CNN-type myocardial infarction prediction using evaluate of myocardial dynamics. The proposed algorithm (i) remove noise using masking image for the ultrasound images into frame-by-frame, (ii) determine the cardiac cycle using optical flow analysis [1] for the characteristic dynamics of each cardiac anatomical region, and then, (iii) predict the myocardial infarction using CNN from time-frequency analysis of myocardial dynamics.