Enhancing Hypertension Risk Diagnosis Using a Hybrid Machine Learning Framework: Leveraging Body Composition Data.

Area of Science:

• Biomedical Informatics
• Machine Learning in Healthcare
• Cardiovascular Disease Research

Background:

• Hypertension is a major global health issue, leading to severe complications.
• Current prediction methods may lack interpretability and reliability.
• Noninvasive body composition features offer a promising avenue for risk assessment.

Purpose of the Study:

• To develop and evaluate a dual-scenario hybrid machine learning framework for hypertension risk prediction.
• To enhance both the interpretability and predictive reliability of hypertension risk assessment.
• To explore the utility of noninvasive body composition features in personalized hypertension risk prediction.

Main Methods:

• Implemented a hybrid machine learning framework with unsupervised clustering (K-Means) and supervised classification (SVM, ExtraTrees, etc.).
• Scenario 1: Identified physiological subgroups within hypertensive individuals.
• Scenario 2: Performed binary classification between healthy and hypertensive subjects using cluster-augmented data.

Main Results:

• Scenario 1 identified five distinct hypertensive subgroups with significant intercluster variability (p < 0.001).
• Scenario 2, using ExtraTrees classifier on cluster-augmented data, achieved high accuracy (98.23%) and AUC (99.87%).
• Both clustering and feature selection improved model generalization, especially for ensemble methods.

Conclusions:

• Integrating unsupervised clustering with supervised classification provides a robust and explainable framework for hypertension risk prediction.
• The proposed hybrid model enhances early detection and supports precision healthcare initiatives.
• Noninvasive body composition features, when analyzed with advanced ML, are valuable for personalized risk assessment.