Quality assessment of echocardiographic cine using recurrent neural networks: Feasibility on five standard view planes

Abstract

Echocardiography (echo) is a clinical imaging technique which is highly dependent on operator experience. We aim to reduce operator variability in data acquisition by automatically computing an echo quality score for real-time feedback. We achieve this with a deep neural network model, with convolutional layers to extract hierarchical features from the input echo cine and recurrent layers to leverage the sequential information in the echo cine loop. Using data from 509 separate patient studies, containing 2,450 echo cines across five standard echo imaging planes, we achieved a mean quality score accuracy of 85 %compared to the gold-standard score assigned by experienced echosonographers. The proposed approach calculates the quality of a given 20 frame echo sequence within 10 ms, sufficient for real-time deployment.