Large-scale empirical evaluation finds that geometric augmentations such as affine, shift-scale-rotate, perspective, and random horizontal flip improve cross-dataset performance in echocardiography segmentation, while aggressive intensity augmentations degrade it.
Contrastive Learning for View Classification of Echocardiograms
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
EAGT: Echocardiography Augmentation for Generalisability and Transferability
Large-scale empirical evaluation finds that geometric augmentations such as affine, shift-scale-rotate, perspective, and random horizontal flip improve cross-dataset performance in echocardiography segmentation, while aggressive intensity augmentations degrade it.