GridNet with automatic shape prior registration for automatic MRI cardiac segmentation

05/24/2017
by   Clement Zotti, et al.
0

In this paper, we propose a fully automatic MRI cardiac segmentation method based on a novel deep convolutional neural network (CNN) designed for the 2017 ACDC MICCAI challenge. The novelty of our network comes with its embedded shape prior and its loss function tailored to the cardiac anatomy. Our model includes a cardiac centerof-mass regression module which allows for an automatic shape prior registration. Also, since our method processes raw MR images without any manual preprocessing and/or image cropping, our CNN learns both high-level features (useful to distinguish the heart from other organs with a similar shape) and low-level features (useful to get accurate segmentation results). Those features are learned with a multi-resolution conv-deconv "grid" architecture which can be seen as an extension of the U-Net. Experimental results reveal that our method can segment the left and right ventricles as well as the myocardium from a 3D MRI cardiac volume in 0.4 second with an average Dice coefficient of 0.90 and an average Hausdorff distance of 10.4 mm.

READ FULL TEXT

page 7

page 10

page 11

page 12

page 14

research
08/03/2017

Automatic Segmentation and Disease Classification Using Cardiac Cine MR Images

Segmentation of the heart in cardiac cine MR is clinically used to quant...
research
06/14/2017

ν-net: Deep Learning for Generalized Biventricular Cardiac Mass and Function Parameters

Background: Cardiac MRI derived biventricular mass and function paramete...
research
08/06/2018

V-FCNN: Volumetric Fully Convolution Neural Network For Automatic Atrial Segmentation

Atrial Fibrillation (AF) is a common electro-physiological cardiac disor...
research
06/18/2019

Cardiac Segmentation from LGE MRI Using Deep Neural Network Incorporating Shape and Spatial Priors

Cardiac segmentation from late gadolinium enhancement MRI is an importan...
research
08/18/2020

PC-U Net: Learning to Jointly Reconstruct and Segment the Cardiac Walls in 3D from CT Data

The 3D volumetric shape of the heart's left ventricle (LV) myocardium (M...
research
09/05/2023

High-resolution 3D Maps of Left Atrial Displacements using an Unsupervised Image Registration Neural Network

Functional analysis of the left atrium (LA) plays an increasingly import...
research
03/01/2022

Tempera: Spatial Transformer Feature Pyramid Network for Cardiac MRI Segmentation

Assessing the structure and function of the right ventricle (RV) is impo...

Please sign up or login with your details

Forgot password? Click here to reset