Core Concepts
A wearable and stretchable high-density electromyography (EMG) array is introduced, with a fabrication methodology that enables the manufacturing of stretchable sleeves with consistent and standardized coverage across subjects, eliminating the need for time-consuming skin preparation.
Abstract
The paper presents the design, fabrication, and evaluation of a stretchable high-density electromyography (HD-EMG) array. The key highlights are:
Fabrication Methodology:
The stretchable array is fabricated from a flexible printed circuit board (PCB) substrate, which is then encased in a stretchable silicone rubber substrate.
The fabrication process allows for the manufacturing of stretchable sleeves with consistent and standardized coverage across subjects, eliminating the need for time-consuming skin preparation.
The fabrication can be executed within a lab/maker-space setting using readily available materials.
Characterization and Validation:
Baseline noise characterization shows the dry-electrode configuration performs comparably to the wet-electrode grid, with a 7.2% decrease in baseline RMS for the wet-electrode configuration.
Electrochemical characterization of the electrode sites demonstrates the suitability of the gold-coated electrodes for high-quality EMG signal acquisition.
Validation experiments on gesture recognition and EMG signal decomposition show the stretchable array matches or outperforms traditional EMG grids, with over 95.9% accuracy in gesture classification tasks.
Advantages and Applications:
The stretchable array design overcomes practical challenges associated with traditional EMG grids, such as the need for skin preparation and the lack of compact electrode interfaces.
The fabrication method can be generalized to other wearable modalities, including ultrasound, inertial measurement units, vibration motors, and functional electrical stimulation electrodes.
The open-source nature of the design and fabrication process enables researchers and makers to fabricate customized stretchable human-machine interfaces.
Stats
The average RMS noise values were 14.55 μV and 13.5 μV for the dry and wet-electrode configurations, respectively.
The decomposition analysis detected 12 motor units using the dry-electrode grid and 13 motor units using the wet-electrode grid.
Quotes
"The proposed fabrication method allows the manufacturing of stretch-able sleeves, with consistent and standardised coverage across subjects."
"The results of our study showed that the developed stretchable array matches or outperforms traditional EMG grids and holds promise in furthering the real-world translation of high-density EMG for human-machine interfaces."