Automatic Myocardial Segmentation by Using A Deep Learning Network in Cardiac MRI

TL;DR

This paper introduces a deep learning-based method for automatic myocardial segmentation in cardiac MRI, utilizing residual learning, batch normalization, and Jaccard distance optimization, achieving faster and more accurate results than previous methods.

Contribution

The study presents novel improvements including the use of Jaccard distance as a loss function, residual learning, and batch normalization in a fully convolutional network for myocardial segmentation.

Findings

Outperforms previous segmentation approaches.

Segmentation takes less than 22 seconds for a typical volume.

Provides accurate and efficient myocardial segmentation.

Abstract

Cardiac function is of paramount importance for both prognosis and treatment of different pathologies such as mitral regurgitation, ischemia, dyssynchrony and myocarditis. Cardiac behavior is determined by structural and functional features. In both cases, the analysis of medical imaging studies requires to detect and segment the myocardium. Nowadays, magnetic resonance imaging (MRI) is one of the most relevant and accurate non-invasive diagnostic tools for cardiac structure and function. In this work we propose to use a deep learning technique to assist the automatization of myocardial segmentation in cardiac MRI. We present several improvements to previous works in this paper: we propose to use the Jaccard distance as optimization objective function, we integrate a residual learning strategy into the code, and we introduce a batch normalization layer to train the fully convolutional neural network. Our results demonstrate that this architecture outperforms previous approaches based on a similar network architecture, and that provides a suitable approach for myocardial segmentation. Our benchmark shows that the automatic myocardial segmentation takes less than 22 seg. for a volume of 128~x~128~x~13 pixels in a 3.1 GHz intel core i7.