A neural-net-transformed disease label is fed into causal discovery, and the resulting 'causal strength' ranks are compared with ML feature importance on heart failure EHR data, with the comparison likely inflated by the proxy's circularity.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Causal Explainability of Machine Learning in Heart Failure Prediction from Electronic Health Records
A neural-net-transformed disease label is fed into causal discovery, and the resulting 'causal strength' ranks are compared with ML feature importance on heart failure EHR data, with the comparison likely inflated by the proxy's circularity.