Input-space standardization via lung cropping and density-based slice sampling reduces inter-source feature variance by 75% and improves COVID-19 CT classification F1 by roughly 24 points across architectures.
Deep learning for the harmonization of structural mri scans: a survey
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
eess.IV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Taming Domain Shift in Multi-source CT-Scan Classification via Input-Space Standardization
Input-space standardization via lung cropping and density-based slice sampling reduces inter-source feature variance by 75% and improves COVID-19 CT classification F1 by roughly 24 points across architectures.