Non-invasive fetal ECG extraction using discrete wavelet transform recursive inverse adaptive algorithm.

Area of Science:

• Biomedical Signal Processing
• Fetal Health Monitoring
• Cardiology

Background:

• Fetal heart activity is crucial for assessing fetal well-being.
• Early detection of fetal cardiac issues enables timely intervention and improved outcomes.

Purpose of the Study:

• To present a novel adaptive filtering algorithm for fetal electrocardiogram (FECG) extraction from abdominal electrocardiogram (AECG) signals.
• To accurately identify and extract fetal QRS complex waves for enhanced fetal healthcare.

Main Methods:

• Utilized a discrete wavelet transform recursive inverse (DWT-RI) adaptive filtering algorithm.
• Employed maternal electrocardiogram (MECG) as a reference to suppress MECG components in AECG.
• Compared the DWT-RI algorithm against Least Mean Squares (LMS), Recursive Least Squares (RLS), and Recursive Inverse (RI) algorithms.

Main Results:

• Successfully identified and extracted fetal QRS waveforms from AECG signals.
• Validated the algorithm's performance using both synthetic and real clinical data.
• Demonstrated superior accuracy and positive predictivity compared to conventional adaptive filtering methods.

Conclusions:

• The proposed DWT-RI adaptive filtering algorithm effectively extracts fetal QRS waveforms from AECG.
• The algorithm exhibits superior performance over existing adaptive filtering techniques in accuracy and positive predictivity for fetal monitoring.