ML models on EHR data achieve AUROC 0.719 for 3-month uncontrolled hypertension risk, outperforming a last-BP baseline of 0.634, with logistic regression matching RNN performance.
Scalable and accurate deep learning with electronic health records,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning to Identify Patients at Risk of Uncontrolled Hypertension Using Electronic Health Records Data
ML models on EHR data achieve AUROC 0.719 for 3-month uncontrolled hypertension risk, outperforming a last-BP baseline of 0.634, with logistic regression matching RNN performance.