Cancellation of artifacts in ECG signals using block adaptive filtering techniques

Area of Science:

• Biomedical Engineering
• Signal Processing
• Medical Informatics

Background:

• Electrocardiogram (ECG) signals are vital for diagnosing cardiac conditions.
• Noise and artifacts in ECG recordings can obscure critical diagnostic information.
• Accurate signal processing is essential for reliable ECG interpretation.

Purpose of the Study:

• To present block-based adaptive filter structures for ECG signal processing.
• To evaluate the efficacy of Block LMS (BLMS) and Fast Block LMS (FBLMS) algorithms in noise removal.
• To compare the performance of block-based algorithms against the conventional LMS algorithm for ECG artifact reduction.

Main Methods:

• Implementation of block-based adaptive filter structures, including BLMS and FBLMS algorithms.
• Application of these algorithms to real ECG signals from the MIT-BIH database.
• Comparative analysis of the proposed algorithms with the conventional Least Mean Squares (LMS) algorithm.

Main Results:

• Block-based adaptive filters successfully estimate deterministic components and remove noise from ECG signals.
• BLMS and FBLMS algorithms effectively remove artifacts while preserving low-frequency components and subtle ECG features.
• The proposed block-based algorithms demonstrated superior performance compared to the conventional LMS algorithm.

Conclusions:

• Block-based adaptive filtering, particularly BLMS and FBLMS, offers a superior approach for ECG noise and artifact removal.
• These algorithms are suitable for applications demanding high signal-to-noise ratios and rapid convergence.
• The findings support the use of block-based adaptive filters for enhanced ECG signal quality and diagnostic reliability.