Causal Inference for Hypertension Prediction.

Area of Science:

• Cardiology
• Biomedical Engineering
• Data Science

Background:

• Hypertension is a major global health issue, increasing cardiovascular disease and mortality.
• Accurate hypertension detection is crucial for effective healthcare management.
• Current methods often rely on signal correlations, which can be unreliable.

Purpose of the Study:

• To develop a more reliable method for hypertension prediction using noninvasive cardiac signals.
• To differentiate between correlation and causation in feature selection for hypertension diagnosis.
• To leverage causal inference for improved hypertension risk assessment.

Main Methods:

• Utilized electrocardiogram (ECG) and photoplethysmogram (PPG) signals.
• Employed greedy equivalence search to construct a causal graph linking signal features to hypertension.
• Applied machine learning models, including random forest, for hypertension classification based on causal features.

Main Results:

• The causal inference approach effectively identified features causally related to hypertension.
• Machine learning models demonstrated high classification performance.
• The random forest model achieved an accuracy of 0.987, precision of 0.990, recall of 0.981, and F1-score of 0.985.

Conclusions:

• Causal inference provides a more reliable basis for hypertension prediction than correlation-based methods.
• This novel approach enhances the accuracy and reliability of diagnosing hypertension from ECG and PPG signals.
• The findings support the clinical relevance of causal inference in cardiovascular risk prediction.