Key result
Hybrid XGBSVM model predicts 3-year hypertensive heart disease with an AUC of ~0.94, outperforming traditional ML.
Why the study?
The prevalence of hypertensive heart disease has increased annually, creating a need for effective prediction methods to reduce its burden.
Does the XGBSVM hybrid model improve the prediction of hypertensive heart disease within three years in hypertensive patients compared to traditional machine learning models?
Population
Hypertensive patients at risk of hypertensive heart disease
Comparison
XGBSVM hybrid model vs Random Forest, GBDT, and XGBoost models
Design
Other machine learning prediction study
Follow-up
3 years
Authors
Loading...
May aid hypertension risk stratification; leaves open prospective validation before clinical adoption.
Does the XGBSVM hybrid model improve the prediction of hypertensive heart disease within three years in hypertensive patients compared to traditional machine learning models?
Absolute Event Rate: 0.939% vs 0.885%
The proposed XGBSVM hybrid machine learning model demonstrates high accuracy in predicting the 3-year risk of hypertensive heart disease, potentially aiding in targeted preventive treatment.
Chang et al. (2019) studied Hypertension (n=372). XGBSVM hybrid model vs. XGBoost, Random Forest, and GBDT models was evaluated on Area Under the Curve (AUC) for predicting hypertensive heart disease. The XGBSVM hybrid model predicted the development of hypertensive heart disease within three years with an AUC of 0.939, outperforming traditional machine learning models like XGBoost.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: