Fast Parabolic Fitting: An R-Peak Detection Algorithm for Wearable ECG Devices.

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Heart diseases are a leading global health concern, necessitating early detection and continuous monitoring.
• Wearable electrocardiogram (ECG) devices require algorithms with low power and memory footprints for prolonged use.
• Accurate R-peak detection is crucial for analyzing ECG signals and determining heartbeat intervals.

Purpose of the Study:

• To develop a novel, computationally efficient algorithm for R-peak detection in ECG signals.
• To enhance the robustness of ECG analysis for wearable devices, particularly in noisy conditions.
• To provide a reliable method for continuous heart rate monitoring in clinical and personal health applications.

Main Methods:

• A novel algorithm fitting a least-squares parabola to the ECG signal and adaptively shaping it.
• Elimination of band-pass filters to prevent R-peak smoothing and improve detection accuracy.
• Testing and validation using the QT database and the BIH-MIT database.

Main Results:

• The proposed algorithm demonstrated superior performance over the Pan-Tompkins algorithm on the QT database.
• The algorithm showed competitive results compared to state-of-the-art methods on the QT database.
• Performance on the BIH-MIT database confirmed the algorithm's practical utility in clinical settings.

Conclusions:

• The novel algorithm offers a robust and computationally efficient solution for R-peak detection in wearable ECG devices.
• Its ability to function without precise isoelectric line localization contributes to its low complexity and broad applicability.
• The findings support the use of this algorithm for improved early diagnosis and continuous monitoring of heart conditions.