CardiacNAS uses resource-aware evolutionary NAS on a cardiac-specific search space to produce a UNet variant with 93.22% DSC and 4.73 mm HD95 using only 3.58 M parameters and 14.56 GFLOPs on the ACDC dataset.
Hidiff: Hybrid dif- fusion framework for medical image segmentation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Resource-Aware Evolutionary Neural Architecture Search for Cardiac MRI Segmentation
CardiacNAS uses resource-aware evolutionary NAS on a cardiac-specific search space to produce a UNet variant with 93.22% DSC and 4.73 mm HD95 using only 3.58 M parameters and 14.56 GFLOPs on the ACDC dataset.