Recent Research for Unobtrusive Atrial Fibrillation Detection Methods Based on Cardiac Dynamics Signals: A Survey

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia, leading to severe conditions like stroke and heart failure.
• Increasing AF morbidity and mortality necessitate unobtrusive, long-term detection methods for routine life.
• Current non-invasive methods include electrocardiogram (ECG) and cardiac dynamics signals (BCG, SCG, PPG).

Purpose of the Study:

• To review current unobtrusive atrial fibrillation detection methods using cardiac dynamics signals.
• To summarize data acquisition, preprocessing, feature extraction, and classification techniques.
• To analyze limitations and discuss future research challenges in AF detection.

Main Methods:

• Review of existing literature on non-invasive AF detection using BCG, SCG, and PPG signals.
• Analysis of data acquisition and preprocessing strategies for cardiac dynamics signals.
• Examination of feature extraction, selection, and classification algorithms for AF diagnosis.

Main Results:

• Cardiac dynamics signals offer a viable alternative for unobtrusive, long-term AF monitoring.
• Methods encompass signal acquisition, preprocessing, feature engineering, and machine learning classification.
• Existing approaches show potential but face limitations requiring further investigation.

Conclusions:

• Unobtrusive AF detection via cardiac dynamics signals is crucial for managing increasing AF prevalence.
• Further research is needed to overcome limitations in current methods for robust AF diagnosis.
• Future work should focus on improving signal quality, feature robustness, and diagnostic accuracy.