Multi-Stage Bi-Atrial Segmentation from 3D LGE MRI Using V-Net Models
A multi-stage framework for bi-atrial segmentation from 3D late gadolinium-enhanced MRI has been developed. The pipeline includes preprocessing with multidimensional contrast limited adaptive histogram equalization, coarse segmentation using a V-Net family model on down-sampled MRI, and fine segmentation with another V-Net model. Asymmetric loss optimizes model weights. The work is published on arXiv.
Key facts
- Multi-stage framework for bi-atrial segmentation from 3D LGE MRI
- Preprocessing step uses multidimensional contrast limited adaptive histogram equalization (MCLAHE)
- Coarse region segmentation from MCLAHE-enhanced and down-sampled MRI using a V-Net family model
- Fine segmentation from the coarse region using another V-Net model
- Asymmetric loss is adopted to optimize model weights
- Published on arXiv under Computer Science > Computer Vision and Pattern Recognition
Entities
Institutions
- arXiv