Researchers have used data collected from patient electronic health records (EHRs) to train a machine learning algorithm to predict the onset of hypertension in individuals in the next year, according to a study published in the Journal of Medical Internet Research.
Hypertension prediction models could assist in managing the disease. In this study, researchers outlined the development and validation of a risk prediction model in forecasting the onset of hypertension.
Data was collected from 823,627 patient EHRs from the Maine Health Information Exchange network. The machine learning algorithm, XGBoost, was tested on an additional 680,810 EHRs. The algorithm was evaluated on its accuracy in sorting patients into one of five risk factors based on their probability of developing hypertension within one year.
In addition to predicting patient risk scores for hypertension, the algorithm was able to recognize Type 2 diabetes, lipid disorders, cardiovascular disease, mental illness, clinical utilization indicators and socioeconomic determinants as features associated with incidents of hypertension.
“With statewide EHR datasets, our study prospectively validated an accurate one-year risk prediction model for incident essential hypertension,” concluded first author Chengyin Ye, PhD and colleagues. “Our real-time predictive analytic model has been deployed in the state of Maine, providing implications in interventions for hypertension and related diseases and hopefully enhancing hypertension care.”