An AI-driven model using 30 years of EHR data enhances screening for primary aldosteronism, a frequently underdiagnosed hypertension cause.
Key Details
- 1Study presented at ENDO 2026 by Mayo Clinic researchers, using Mayo Clinic Platform data from 1986–2025.
- 2AI model analyzed clinical variables: age, gender, ICD codes for hypertension/hypokalemia, blood pressure, potassium levels, and prescriptions.
- 3Model was developed with over 22,000 patients, tested on 225,887 hypertensive adults.
- 4XGBoost architecture predicted primary aldosteronism risk 12 months before diagnosis.
- 5At low-risk threshold, the model flagged >90% of cases while missing <10%, identifying about two-thirds for further screening.
Why It Matters
This AI tool addresses a major gap in early detection of a serious but often-missed hypertension cause, enabling targeted intervention that can lower cardiovascular morbidity and healthcare costs. The study demonstrates powerful potential for machine-learning models in analyzing routine clinical data to guide large-scale, cost-effective diagnosis in at-risk populations.
Source
EurekAlert
Related News
•EurekAlert
Peking University Debuts LargePNet for Superior Fluorescence Image Restoration
Peking University's Xi Peng lab introduces LargePNet, a new AI for robust fluorescence image restoration, outperforming patch-based methods.
•EurekAlert
AI Detects Smuggled Marine Life in Airport CT Scans
Researchers developed an AI algorithm to identify smuggled marine wildlife in airport luggage using CT scans with high accuracy.
•EurekAlert
Broadband Optical Spectroscopy Enables Early NEC Detection in Preemies
Researchers successfully used a noninvasive broadband optical spectroscopy (BOS) device to detect necrotizing enterocolitis (NEC) early in premature infants.