Improving Localization and Measurement of M-Waves Using High-Density Surface Electromyography

Academic Background

Surface electromyography (sEMG) is an important tool for studying muscle function and controlling prosthetics. However, its effectiveness is often limited by cross talk from nearby muscles, especially in densely packed areas like the forearm. To address this issue, researchers have introduced high-density surface electromyography (HD-sEMG) technology, which improves spatial resolution and better isolates M-waves (muscle action potentials) from target muscles. This study aims to evaluate the performance of HD-sEMG in localizing M-waves and explore the role of spatial filters in reducing cross talk.

Source of the Paper

This paper was co-authored by Ernesto H. Bedoy, Efrain A. Guirola Diaz, and other researchers from institutions such as the University of Pittsburgh and Carnegie Mellon University. It was first published on December 20, 2024, in the Journal of Neurophysiology, with the DOI 10.1152/jn.00354.2024.

Research Process

1. Study Participants and Experimental Design

The study recruited five right-handed healthy volunteers (3 males, aged 24-40 years). All participants provided informed consent and met the inclusion criteria. The main objective was to precisely measure M-waves in forearm muscles using HD-sEMG combined with ultrasound imaging, and to evaluate the effectiveness of different spatial filters (monopolar, bipolar, and tripolar) in reducing cross talk.

2. HD-sEMG Electrode Placement and Recording

Researchers placed a 64-electrode HD-sEMG grid (SAGA 64+, TMSi) on the right forearm of each participant, with an inter-electrode distance of 8.75 mm. The grid covered multiple forearm muscles, and ultrasound imaging was used to confirm muscle locations and boundaries. During the experiment, electrical stimulation was applied to the median, ulnar, and radial nerves to selectively activate target muscles.

3. Signal Processing and Analysis

Custom MATLAB scripts were used to process HD-sEMG data. Stimulation artifacts were removed, and a 10 Hz high-pass filter was applied. M-wave amplitudes were quantified using peak-to-peak measurements, and recruitment curves were generated at different stimulation intensities. To detect cross talk, linear correlation coefficients between electrode pairs were calculated, and ultrasound imaging confirmed muscle activation regions.

4. Application of Spatial Filters

The study compared the effectiveness of monopolar, bipolar, and tripolar spatial filters in reducing cross talk. Results showed that the tripolar filter performed best, significantly reducing cross talk from adjacent muscles, especially at higher stimulation intensities.

Key Findings

1. Localization of M-Waves and Cross Talk

The study demonstrated that HD-sEMG successfully isolated M-waves in forearm muscles, with minimal cross talk at low stimulation intensities. However, cross talk increased significantly at higher intensities. Bipolar and tripolar spatial filters effectively reduced cross talk, with the tripolar filter performing exceptionally well, reducing cross talk to negligible levels.

2. Role of Ultrasound Imaging

Ultrasound imaging played a crucial role in confirming muscle boundaries and locations. It allowed researchers to accurately identify muscles beneath the HD-sEMG grid and distinguish true muscle activation from cross talk.

3. Effectiveness of Spatial Filters

Results showed high correlations between electrode pairs in monopolar HD-sEMG (transverse r=0.97, longitudinal r=0.95), while bipolar and tripolar filters exhibited lower correlations (bipolar: transverse r=0.41, longitudinal r=0.19; tripolar: transverse r=0.17, longitudinal r=0.01). The tripolar filter significantly reduced cross talk, with a 51.10% amplitude decay one electrode away, compared to only 10.32% for the monopolar filter.

Conclusions and Significance

By combining HD-sEMG with ultrasound imaging, this study successfully improved the precision of M-wave localization and effectively reduced cross talk. The tripolar spatial filter performed best in reducing cross talk, significantly enhancing the accuracy of muscle activation measurements. This research provides new tools for neurophysiological studies, particularly in assessing motor pathway integrity and developing neurorehabilitation tools.

Research Highlights

1. Innovative Approach: This study is the first to combine HD-sEMG with ultrasound imaging, offering a novel method for precise M-wave measurement and localization.
2. Effective Cross Talk Reduction: The application of the tripolar spatial filter successfully reduced cross talk to negligible levels, significantly improving measurement accuracy.
3. Clinical Application Potential: This technology can be used to develop more accurate prosthetic control systems and provide new assessment tools for patients with neurological injuries.

Additional Valuable Information

The study also found that cross talk is particularly pronounced at forearm muscle boundaries, and ultrasound imaging played a key role in distinguishing muscle activation regions from cross talk. Additionally, the study data and analysis code have been made publicly available for further validation and application by other researchers.

---

Through this study, researchers not only addressed the issue of cross talk in HD-sEMG technology but also provided new insights and methods for future neurophysiological research and clinical applications.