Ω-Net (Omega-Net): Fully Automatic, Multi-View Cardiac MR Detection, Orientation, and Segmentation with Deep Neural Networks

Pixelwise segmentation of the left ventricular (LV) myocardium and the four cardiac chambers in 2-D steady state free precession (SSFP) cine sequences is an essential preprocessing step for volume estimation (e.g., ejection fraction, stroke volume, and cardiac output); morphological characterization (e.g., myocardial mass, regional wall thickness and thickening, and eccentricity); and strain analysis (Peng et  al., 2016). However, automatic cardiac segmentation remains a notoriously difficult problem, given:

Source: Medical Image Analysis - May 22, 2018 Category: Radiology Authors: Davis M. Vigneault, Weidi Xie, Carolyn Y. Ho, David A. Bluemke, J. Alison Noble Source Type: research