Cardiovascular risk analysis by means of pulse morphology and clustering methodologies

Area of Science:

• Cardiology
• Biomedical Engineering
• Data Science

Background:

• Cardiovascular diseases remain a leading cause of mortality worldwide.
• Accurate cardiovascular risk assessment is crucial for timely intervention.
• Current risk assessment tools have limitations in identifying all at-risk individuals.

Purpose of the Study:

• To develop and validate a novel clustering methodology for cardiovascular risk assessment.
• To utilize arterial pressure waveform (APW) parameters within this methodology.
• To compare the clustering results with established cardiovascular risk scores.

Main Methods:

• Clustering analysis using Expectation Maximization (EM) and k-means algorithms.
• Inclusion of 116 subjects: 23 hypertensive and 93 healthy.
• Analysis of APW parameters and biochemical data (hypertensive group).
• Validation against Framingham Risk Score (FRS), SCORE, ASSIGN, and PROCAM.

Main Results:

• The clustering methodology demonstrated a very significant correlation with the ASSIGN risk score (r=0.582, p<0.01).
• A significant correlation was found with the Framingham Risk Score (FRS) (r=0.458, p<0.05).
• The analysis successfully identified a cluster with higher cardiovascular risk within the healthy subject group.

Conclusions:

• The developed clustering methodology using APW parameters shows promise for cardiovascular risk assessment.
• This approach offers new insights and potential for integration into future risk scoring trials.
• The methodology can aid in identifying high-risk individuals, including within seemingly healthy populations.