On a 360-patient NHL trial dataset, a transfer-learned ARX-GRU network predicts platelet nadirs better than mechanistic and hybrid UDE models when at least three dense training cycles are available, while mechanistic models remain competitive under sparse data.
Title resolution pending
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.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Developing hybrid mechanistic and data-driven personalized prediction models for platelet dynamics
On a 360-patient NHL trial dataset, a transfer-learned ARX-GRU network predicts platelet nadirs better than mechanistic and hybrid UDE models when at least three dense training cycles are available, while mechanistic models remain competitive under sparse data.