Key result
PU-Boost cuts false-negative hypertension screening results ~41% compared to standard XGBoost.
Why the study?
In hypertension screening, previously diagnosed cases provide reliable positive labels while individuals without a prior diagnosis comprise both undiagnosed hypertensive and non-hypertensive individuals, creating a positive-unlabeled learning challenge under incomplete diagnostic labeling.
Does the PU-Boost two-stage positive-unlabeled learning framework improve sensitivity and reduce false negatives when screening for undiagnosed hypertension compared to conventional supervised machine learning models in community-dwelling adults aged ≥45 years?
Cross-Sectional (n=43,105)
Yes
Does the PU-Boost two-stage positive-unlabeled learning framework improve sensitivity and reduce false negatives when screening for undiagnosed hypertension compared to conventional supervised machine learning models in community-dwelling adults aged ≥45 years?
The PU-Boost machine learning framework reduces missed cases of undiagnosed hypertension by 41.1% compared to standard XGBoost using only non-invasive, readily obtainable demographic and anthropometric variables.
May enhance undiagnosed hypertension detection with routine data; hypothesis-generating and requires prospective validation before clinical use.
In hypertension screening, previously diagnosed cases provide reliable positive labels, whereas individuals without a prior diagnosis comprise both undiagnosed hypertensive and non-hypertensive individuals, creating a canonical positive-unlabeled (PU) learning setting. We propose PU-Boost, a two-stage reliable-sample reconstruction framework that uses only readily obtainable demographic, behavioral, medical-history, and anthropometric variables. During training, only previously diagnosed hypertension cases are treated as labeled positives (P), while all remaining individuals are treated as unlabeled (U); no additional reference labels are supplied to the two-stage reconstruction procedure. Elastic Net first provides a global ranking of P versus U, after which a random forest refines the uncertain region and highly ambiguous observations are rejected. XGBoost is then trained on the reconstructed class-balanced pseudo-labeled sample. In the independent test set, PU-Boost achieved a sensitivity of 0.817 and a balanced accuracy of 0.682, the highest observed values among the compared models. It produced 729 false negatives, 41.1% fewer than standard XGBoost, although specificity decreased to 0.547. SHAP analysis indicated that established hypertension-related features, including age, body mass index, waist circumference, and family history, contributed strongly to model output. PU-Boost therefore provides a statistically motivated framework for reducing missed cases when screening for undiagnosed hypertension under incomplete diagnostic labeling.
No takes yet. Share an insight, caveat, or question.
Wu et al. (2026) conducted a cross-sectional in Undiagnosed hypertension screening (n=43,105). PU-Boost (two-stage reliable-sample reconstruction positive-unlabeled learning framework combining Elastic Net, random forest, and XGBoost) vs. Conventional supervised-learning baselines (standard XGBoost, logistic regression, single-hidden-layer neural network) was evaluated on Sensitivity for screening undiagnosed and reference hypertension in the independent test set. PU-Boost achieved a sensitivity of 0.817 and balanced accuracy of 0.682 for hypertension screening, reducing false negatives by 41.1% compared to standard XGBoost.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: