A new approach to early diagnosis of congestive heart failure disease by using Hilbert-Huang transform

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Congestive heart failure (CHF) is a significant cardiac condition often caused by coronary artery disease (CAD).
• Accurate and early diagnosis of CHF and CAD is crucial for patient outcomes.
• Non-linear and non-stationary signal analysis offers potential for improved diagnostic capabilities.

Purpose of the Study:

• To develop and validate a system for the simultaneous diagnosis of CHF and CAD.
• To investigate the feasibility of early CHF diagnosis by analyzing CAD indicators.
• To leverage advanced signal processing techniques for enhanced cardiac disease detection.

Main Methods:

• Utilized the Hilbert-Huang transform (HHT) to analyze R-R intervals from electrocardiogram (ECG) data.
• Extracted statistical features from intrinsic mode functions derived via HHT.
• Employed a multilayer perceptron neural network for classification of CHF, CAD, and control groups.

Main Results:

• The system achieved high diagnostic performance: 97.83% accuracy, 93.79% specificity, and 100% sensitivity for CHF and CAD.
• Early diagnosis of CHF through CAD interpretation showed excellent results with 97.53% accuracy, 98.18% specificity, and 97.13% sensitivity.
• Demonstrated the integration of CAD diagnosis into a unified system for both diagnosis and early detection of CHF.

Conclusions:

• A novel integrated system effectively diagnoses both CHF and CAD using ECG R-R interval analysis.
• The proposed method enables early detection of CHF by accurately interpreting associated CAD.
• This approach offers a promising tool for comprehensive cardiac disease management and patient care.