A Multi-Class Approach for the Automatic Detection of Congestive Heart Failure in Windowed ECG

Area of Science:

• Cardiology
• Biomedical Engineering
• Artificial Intelligence in Healthcare

Background:

• Congestive heart failure (CHF) is a chronic condition with significant mortality and morbidity.
• Early detection of CHF is crucial for improving patient outcomes.
• Current diagnostic methods may lack real-time capabilities or require extensive data processing.

Purpose of the Study:

• To introduce HARPER, a novel, automatic detector for congestive heart failure (CHF) episodes.
• To evaluate HARPER's ability to distinguish between Normal Sinus Rhythm (NSR), CHF, and no-CHF states.
• To demonstrate HARPER's reliability, early diagnosis capability, and independence from ECG annotation.

Main Methods:

• HARPER utilizes real-time features extracted from brief ECG signal segments.
• The system operates as an independent tool, not requiring ECG annotation or segmentation algorithms.
• Validation involved both intra-patient and inter-patient experimental schemes.

Main Results:

• HARPER accurately distinguishes between NSR, CHF, and no-CHF.
• The detector provides reliable and early CHF diagnosis.
• Performance is comparable to state-of-the-art methods.

Conclusions:

• HARPER is a fast, highly accurate, multi-class detector suitable for modern IoMT systems.
• The tool offers a reliable approach for early CHF detection.
• Guidelines for temporal window configuration in automatic CHF detection are provided.