การพัฒนาเซนเซอร์วัดสัญญาณไฟฟ้ากล้ามเนื้อชนิดไร้สายเพื่องานวิจัย

Abstract

Electromyography (EMG) is the method of observing the activities of muscles in terms of electrical signals. This type of biopotential measurement has been applied as a tool for neuromuscular diagnostic assessment and as well as control signals for related innovative technologies, e.g., wheelchairs and limb prostheses. A specific device is necessary to amplify this type of biopotential signal from its tiny voltage level to an appropriate level for its applications. The EMG devices are generally expensive and bulky. They require an amount of electrode cable for signal acquisition. Therefore, the devices are unsuitable for use in high mobility movement activities. This engineering project aimed to design and develop a portable wireless EMG sensor prototype. The designed prototype consists of 2 main parts; an analog biopotential amplifier and a microcontroller. The first part consists of 4 stages; a preamplifier, a lowpass filter, a high pass filter, and a non-inverting amplifier. These circuits amplify the EMG signal according to the adjustable amplification gain. Then, the analog signals are converted into digital data by the analog-to-digital converter (ADC) on the ESP32 microcontroller development board. The digital EMG data are recorded to an SD card and visualized on the computer screen. The ESP32 microcontroller was programmed to be able to adjust the data sampling rate of the ADC at specific values to acquire the EMG signal data at a higher resolution if necessary for neuromuscular research. The efficiencies of the developed prototype were evaluated in terms of battery consumption, signal-to-noise ratio (SNR), and frequency response in comparison with a low-cost muscle sensor, MyoWare. Additionally, the programmed firmware was evaluated in terms of its error in data acquisition rate at specified sampling rates of 2000, 4000, 8000, and 10,000 samples-per-second respectively. The evaluations showed that the prototype device can be used with a 10,000 mAh battery for up to 5 hours with a signal-to-noise ratio of 22.2715 dB. The signal-to-noise ratio should be approximately 100 dB or more depending on the amplification rate of the signal within the circuit as well and the frequency response from the Myoware sensor and prototype device has a frequency range of 0-500HZ, which is the normal frequency range of the EMG signal. เท addition, the sampling rate of 2KHz, 4KHz, 8KHz, and lOKHz frequencies showed a small error of 1.42% of the Myoware sensor and 1.51% of the prototype, which was considered a low error value and indicates the stable performance of the firmware.