A two-stage nnUNet framework with patient-specific signed distance maps and wall-masked loss achieves 61.1% Dice and 1.711 mm ASSD for left atrial scar segmentation on the LAScarQS 2022 dataset.
A fully auto- matic deep learning method for atrial scarring segmentation from late gadolinium-enhanced mri images
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Two Stage Pipeline for Left Atrial Wall Constrained Scar Segmentation and Localization from LGE-MR Images
A two-stage nnUNet framework with patient-specific signed distance maps and wall-masked loss achieves 61.1% Dice and 1.711 mm ASSD for left atrial scar segmentation on the LAScarQS 2022 dataset.