A Novel and Effective Method for Congestive Heart Failure Detection and Quantification Using Dynamic Heart Rate...

Area of Science:

• Cardiology
• Biomedical Engineering
• Data Science

Background:

• Congestive heart failure (CHF) risk assessment is crucial for patient management and decision-making.
• Existing studies primarily focus on CHF detection using heart rate variability (HRV) rather than quantification.
• Accurate risk stratification aids in personalized treatment strategies for CHF patients.

Purpose of the Study:

• To develop and validate a novel multistage classification approach for CHF risk assessment and quantification.
• To utilize dynamic indices of HRV for capturing autonomic activity changes in CHF patients.
• To establish a clinically meaningful outcome for early CHF assessment and prognosis.

Main Methods:

• Analysis of 116 24-hour RR interval records from the MIT/BIH database (72 normal, 44 CHF patients).
• Implementation of a 4-level risk assessment model (N, P1, P2, P3) based on New York Heart Association (NYHA) classes.
• Application of a non-equilibrium decision-tree-based support vector machine classifier with extracted classical and dynamic HRV indices.

Main Results:

• The multistage risk assessment model achieved a total accuracy of 96.61% for CHF detection and quantification.
• The model demonstrated a powerful predictive capability between assessed and actual CHF risk ratings.
• Dynamic HRV indices effectively captured autonomic activity changes associated with CHF severity.

Conclusions:

• The proposed multistage model offers a robust method for early CHF detection and risk stratification.
• This approach provides a clinically meaningful prognostic marker for CHF patients.
• The dynamic HRV quantification enhances understanding of autonomic dysfunction in CHF.