| dc.description.abstract | In the course of stroke patient rehabilitation, electromyography (EMG) provides not only an indicator of muscle strength and endurance but also an observation on co-ordination of muscles. The purpose of this study is to develop a small-sized wireless EMG measurement system for patient’s convenience. In this study, commercial available data acquisition (NI9205) and wireless transmission (WLS9163) devices were first used for acquiring and transmitting EMG signal to test the function of the signal conditioning circuits. However, the noise interference induced by the long lead wire between the electrode and NI9205 contaminated the EMG signal. This study used Zigbee module and embedded system board to develop a better wireless electromyography measurement system with a Labview-based software tool to monitor, display and analyze the signal. The function of the software tool could be expanded with Labview. In order to verify the signal quality of sensor, intra- and inter-subject variability of EMG signals were measured on four healthy subjects (3 males and 1 female, age 23-25) with three different wireless EMG measurement systems ( i.e., Shimmer research Inc., our Zigbee-based and NI-based measurement systems) in terms of signal-to-noise ratio (SNR) and spectral fatigue index. The results show that both Zigbee-based system (22.7 1.38 dB) and Shimmer system (23.6 1.98 dB) were with similar inter-subject SNRs and variation (6% vs. 8.3%). Spectral fatigue indices of both systems also indicated obvious appearance of muscle fatigue. Before multi-channel EMG measurement, whether the signal quality of sensors is identical has been examined. The results show that sensors’ signal SNRs are all about 24 1.5dB. We then conducted the multi-channel EMG measurement on the same four healthy subjects with three different movements (i.e., shoulder flexion, abduction, and extention). By calculating Co-contraction index (CI), characteristics in different movements were identified. Finally, the EMG signals of three stroke patients were measured and analyzed. Based on the CI calculation, the EMG signals before and after rehabilitation could show muscular recovery from abnormal waveform before rehabilitation and muscular power enforced after rehabilitation. These results imply that EMG signals could be used as an index of efficacy for the rehabilitation. | en_US |