Hybrid-loss whole-heart CT segmentation (MMWHS)

CUHK (Yang et al.)

CT angiography

Filter catalog by Modality:
Cardiac CT

Cardiac chamber segmentation

Filter catalog by Disease / Trait:
Structural Heart & Cardiomyopathy

Segmentation

Filter catalog by Task Type:
Segmentation & Detection

CNN (3D)

Filter catalog by Architecture:
Convolutional (CNN)

code View code

Model weights not public. Contact creators for more information.

3D CNN that segments seven cardiac substructures - both ventricles, both atria, the LV myocardium, ascending aorta, and pulmonary artery trunk - from cardiac CT angiography. Trained with a hybrid loss function combining multiple segmentation objectives. Developed at CUHK for the MICCAI 2017 Multi-Modality Whole Heart Segmentation (MM-WHS) challenge.

memory Specifications

description Publication

Hybrid Loss Guided Convolutional Networks for Whole Heart Parsing open_in_new

2017 · original paper

database Training & evaluation data

MM-WHS 2017 open_in_new

train

public 60 subjects

Multi-Modality Whole Heart Segmentation challenge: 60 cardiac CT/CTA and 60 cardiac MRI volumes from multiple clinical sites, anonymized.

science Capabilities & performance