Explainable machine learning for hypertension prevalence classification: a cross-sectional study in Hainan Province, China - Takeaways - MDSpire

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

  • By

  • Yuewei Wu

  • Shan Huang

  • Miaomiao Qi

  • Yuanyuan Zhang

  • Mei Lu

  • Tianfa Li

  • Yueqiong Kong

  • July 17, 2026

Share

  • 1

    The study analyzed hypertension prevalence using explainable machine learning models on data from 4,606 residents in Hainan Province, China.

  • 2

    The random forest algorithm identified the top 10 significant features for hypertension classification, including age, smoking, and diabetes.

  • 3

    The eXtreme Gradient Boosting (XGBoost) model achieved the highest performance with an AUC of 0.8461 for classifying hypertension cases.

  • 4

    SHAP and LIME algorithms were utilized to explain model predictions, enhancing the interpretability of machine learning outputs.

  • 5

    A web page was developed to facilitate hypertension prevalence screening based on the study's findings, supporting clinical applications.

Original Source(s)

Explainable machine learning for hypertension prevalence classification: a cross-sectional study in Hainan Province, China

  • by Yuewei Wu, Shan Huang, Miaomiao Qi, Yuanyuan Zhang, Mei Lu, Tianfa Li, Yueqiong Kong

  • July 17, 2026

Related Content

ACP Ranks Obesity Drug Options

A living clinical guideline outlines a treatment hierarchy for selected pharmacologic therapies in patients with obesity and selected patients with overweight.

Emerging insights into inflammation-driven atherosclerosis: immune cell mechanisms

Behavior Change Content and Implementation of Large Language Model–Driven Conversational Agents in Cardiometabolic Care: Scoping Review