The Performance of Short-Term Heart Rate Variability in the Detection of Congestive Heart Failure

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Congestive heart failure (CHF) significantly reduces cardiac output and is a leading cause of global cardiac death.
• Current methods for CHF detection often rely on long-term electrocardiogram (ECG) or heart rate variability (HRV) recordings.
• There is a need for efficient and accurate methods to discriminate between CHF patients and healthy individuals.

Purpose of the Study:

• To develop and validate a novel framework for discriminating congestive heart failure (CHF) from healthy subjects using short-term heart rate variability (HRV) intervals.
• To explore the efficacy of a hybrid approach combining a matching pursuit algorithm with genetic algorithms and k-nearest neighbor classification for CHF detection.

Main Methods:

• Utilized short-term HRV data from 256 continuous RR samples.
• Employed a matching pursuit algorithm based on Gabor functions to extract features.
• Implemented a hybrid framework with a genetic algorithm and k-nearest neighbor classifier for feature selection and classification.

Main Results:

• The proposed framework achieved 100% overall accuracy in discriminating CHF from healthy subjects.
• Identified a subset of five key features from an initial set of 16 non-standard features that yielded optimal classification performance.
• Demonstrated superior performance compared to well-established classifier methods.

Conclusions:

• Short-term HRV analysis, combined with advanced signal processing and machine learning, offers a highly accurate method for CHF detection.
• Hybrid frameworks incorporating genetic algorithms show significant potential in outperforming traditional classification approaches for cardiac disease diagnosis.
• The developed framework provides a promising, efficient, and accurate tool for identifying congestive heart failure.