A biomechanics-informed registration model extracts strain-derived features that, together with volumetric features from ground-truth segmentations, classify ACDC cardiac diseases with 100% reported test accuracy.
Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Transactions on Medical Imaging 2018;37(11):2514--2525
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
From Motion to Meaning: Biomechanics-Informed Neural Network for Explainable Cardiovascular Disease Identification
A biomechanics-informed registration model extracts strain-derived features that, together with volumetric features from ground-truth segmentations, classify ACDC cardiac diseases with 100% reported test accuracy.