A Novel ECG Noise Reduction Technique Employing the Chaotic Adaptive Fish Migration Optimization Algorithm

Abstract

Optimization problems are ubiquitous, and obtaining ideal solutions to optimization problems is a challenging task. In terms of denoising the electrocardiogram (ECG) signal, the weight parameters of the adaptive filtering algorithm determine the quality of the output ECG signal to a large extent. However, adaptive filters need to adjust too many parameters, which is a challenging problem. Heuristic algorithm is a powerful tool for solving various optimization problems, and it is very suitable for solving such complex problems. In this paper, a novel ECG denoising method is proposed, which combines a heuristic algorithm with an adaptive filtering algorithm to adjust the weight parameters of the filter. In addition, a new heuristic algorithm, Chaotic Adaptive Fish Migration Optimization (CAFMO), is proposed to introduce the chaotic strategy into the Adaptive Fish Migration Optimization (AFMO) algorithm. The efficiency of a novel denoising method is validated through the use of synthetic data generated by the FECGSYN toolbox. The CAFMO algorithm exhibits superior performance in noise mitigation in ECG data, outperforming other algorithms such as PSO, ABC, BH, GWO, SO and AFMO. The combination of CAFMO algorithm and adaptive filter produces a significant 28% improvement over traditional LMS adaptive filter, with another 20% improvement over other heuristic algorithms combined with adaptive filter.

Keywords

• Intelligence computing
• mate-heuristic algorithm
• fish migration algorithm
• evolutionary computing
• ECG signal processing