Key result
Random Forest model using ECG and clinical features detects hypertension with ~84% accuracy.
Why the study?
Hypertension is a major cardiovascular risk factor that often escapes diagnosis or requires confirmation by multiple office visits.
Does a machine learning algorithm using ECG and basic anthropometric features accurately detect arterial hypertension in patients without known cardiovascular disease?
Cross-Sectional (n=1,091)
Yes
Does a machine learning algorithm using ECG and basic anthropometric features accurately detect arterial hypertension in patients without known cardiovascular disease?
Effect estimate: AUC 0.86
A machine learning algorithm combining basic clinical parameters and ECG features can effectively screen for arterial hypertension with high sensitivity and accuracy.
No takes yet. Share an insight, caveat, or question.
May aid ECG-based hypertension screening; leaves open clinical adoption pending prospective validation.
Angelaki et al. (2022) conducted a cross-sectional in Arterial hypertension (n=1,091). Random Forest machine learning model using ECG and anthropometric features vs. Standard clinical diagnosis (normotensive reference) was evaluated on Detection of hypertension (AUC 0.86). A Random Forest machine learning model using basic clinical parameters and ECG-derived features distinguished hypertensive from normotensive patients with 84.2% accuracy, 91.4% sensitivity, and an AUC of 0.86.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: