A new algorithm for fetal heart rate detection: Fractional order calculus approach

Area of Science:

• Biomedical Signal Processing
• Cardiology
• Maternal-Fetal Medicine

Background:

• Fetal heart rate (FHR) monitoring is crucial for assessing fetal well-being.
• Surface abdominal electrophysiological recordings often suffer from low signal-to-noise ratios (SNR).
• Existing methods may struggle with accurate FHR estimation in noisy conditions.

Purpose of the Study:

• To introduce a novel modified Pan-Tompkins (mPT) algorithm for enhanced FHR detection.
• To validate the mPT method's efficacy on abdominal electrophysiological signals.
• To identify optimal parameters for the mPT algorithm for improved FHR estimation.

Main Methods:

• The mPT algorithm utilizes optimal fractional order derivative and moving average filter window width.
• Algorithm performance was evaluated on PhysioNet Computing in Cardiology Challenge database signals.
• FHR detection was performed on segments selected based on SNR estimation and manual selection.

Main Results:

• The mPT method achieved high accuracy, with an average sensitivity of 97% and positive predictive value of 97%.
• An error rate of approximately 3.5% and an F1 score of 97% were recorded.
• Optimal parameters identified were a fractional order of 0.51 and a moving average filter window width of 24.5 ms.

Conclusions:

• The proposed mPT algorithm demonstrates satisfactory performance for FHR detection.
• The mPT method shows potential for broader applications in noisy biomedical signal analysis.
• Further adaptations could enhance peak detection in various challenging signal environments.