Clinical Scorecard: Utilizing Explainable Machine Learning to Classify Hypertension Prevalence: A Cross-Sectional Analysis in Hainan Province, China

At a Glance

Key Highlights

• 32.5% prevalence of hypertension identified in the study population.
• eXtreme Gradient Boosting (XGBoost) model achieved an AUC of 0.8461.
• Top predictors included age, smoking, urine microalbumin, and diabetes.
• SHAP and LIME algorithms used for model interpretability.
• Study supports enhanced hypertension screening and prevention strategies.

Guideline-Based Recommendations

Diagnosis

• Utilize machine learning models to classify hypertension status.

Management

• Implement screening programs based on machine learning findings.

Monitoring & Follow-up

• Regularly assess hypertension prevalence in community settings.

Risks

• Hypertension is a major risk factor for cardiovascular disease and stroke.

Patient & Prescribing Data

General population in Hainan Province, China.

Focus on early identification and preventive interventions for high-risk individuals.

Clinical Best Practices

• Incorporate explainable machine learning techniques in clinical practice.
• Enhance awareness programs for hypertension management.
• Utilize identified predictors for targeted screening efforts.

Related Resources & Content

• World Health Organization
• Cardiovascular Disease Surveillance and Risk Factors in the Chinese Population