3D Consistent Biventricular Myocardial Segmentation Using Deep Learning for Mesh Generation

Qiao Zheng (1) , Hervé Delingette (1) , Nicolas Duchateau (1) , Nicholas Ayache (1)

1 EPIONE - E-Patient : Images, données & mOdèles pour la médeciNe numériquE

We present a novel automated method to segment the my-ocardium of both left and right ventricles in MRI volumes. The segmen-tation is consistent in 3D across the slices such that it can be directly used for mesh generation. Two specific neural networks with multi-scale coarse-to-fine prediction structure are proposed to cope with the small training dataset and trained using an original loss function. The former segments a slice in the middle of the volume. Then the latter iteratively propagates the slice segmentations towards the base and the apex, in a spatially consistent way. We perform 5-fold cross-validation on the 15 cases from STACOM to validate the method. For training, we use real cases and their synthetic variants generated by combining motion simulation and image synthesis. Accurate and consistent testing results are obtained.

Mots clés: spatial consistency; convolutional neu- ral network; deep learning; myocardial segmentation; propagation

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Intelligence artificielle [cs.AI]; Imagerie médicale; Traitement des images [eess.IV]; Apprentissage [cs.LG]

Qiao Zheng, Hervé Delingette, Nicolas Duchateau, Nicholas Ayache. 3D Consistent Biventricular Myocardial Segmentation Using Deep Learning for Mesh Generation. 2018. ⟨hal-01755317⟩