A closed-loop Neural ODE ROM with CNN-based feedback from growth features stabilizes long-horizon predictions of tissue growth and remodeling, reaching 90.3% clinical tolerance versus 43.7% for open-loop baselines.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CE 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Stable Long-Horizon Neural ODE Reduced-Order Models via Learned Feedback for Biological Growth and Remodeling
A closed-loop Neural ODE ROM with CNN-based feedback from growth features stabilizes long-horizon predictions of tissue growth and remodeling, reaching 90.3% clinical tolerance versus 43.7% for open-loop baselines.