การศึกษาเชิงเปรียบเทียบเวฟเลตแม่แบบในการขจัดสัญญาณรบกวนจากคลื่นไฟฟ้าหัวใจในสัญญาณกล้ามเนื้อกระบังลมระหว่างการฝึกหายใจ

Abstract

The diaphragm is the primary muscle responsible for inspiration. Recently, surface electromyography (sEMG) has been used to monitor diaphragmatic activity as a noninvasive technique by placing electrodes at the torso area. However, this area is close to the heart, resulting in sEMG signals that are often contaminated by electrocardiogram (ECG) also known as ECG Artifact. This can lead to a misinterpretation of the analysis of diaphragmatic muscle activity. Various methods have been proposed to remove ECG artifacts from sEMG, one of which is Wavelet Transform, a technique that has garnered considerable attention. The results of wavelet transform heavily depend on selecting appropriate parameters, particularly the other wavelet, which influences the characteristics of signal separation. However, there is currently no clear guideline or standard for selecting a suitable mother wavelet for ECG artifact removal from diaphragmatic sEMG signals. Therefore, this study aims to identify and compare suitable mother wavelets for this purpose using a single-level Discrete Wavelet Transform (DWT) approach. By applying signal processing techniques alongside statistical analysis. The Data were collected from six volunteers using bilaterally long electrode placement. The results identified five mother wavelets—Daubechies4, Symlet5, Coiflect3, Biorthogonal3.1, and Reverse Biorthogonal1.3—as particularly effective. These wavelets demonstrated excellent signal preservation, suitable frequency characteristics, and effective extraction of respiration correlated signals. Thus, they are considered suitable for applications in monitoring diaphragmatic muscle activity.