• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1485-1488
• /
• 2003

This paper proposes a wireless EMG-based human-computer interface (HCI) for persons with disabilities. For the HCI, four interaction commands are defined by combining three elevation motions of shoulders such as left, right and both elevations. The motions are recognized by comparing EMG signals on the Levator scapulae muscles with double thresholds. A real-time EMG processing hardware is implemented for acquiring EMG signals and recognizing the motions. To achieve real-time processing, filters such as high- and low-pass filter and band-pass and -rejection filter, and a full rectifier and a mean absolute value circuit are embedded on a board with a high speed microprocessor. The recognized results are transferred to a wireless client system such as a mobile robot via a Bluetooth module. From experimental results using the implemented real-time EMG processing hardware, the proposed wireless EMG-based HCI is feasible for the disabled.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.1062-1065
• /
• 2002

We have evaluated the ability of a time-delayed artificial neural network (TDANN) to predict muscle forces using only eletromyographic(EMG) signals. To achieve this goal, tendon forces and EMG signals were measured simultaneously in the gastrocnemius muscle of a dog while walking on a motor-driven treadmill. Direct measurements of tendon forces were performed using an implantable force transducer and EMG signals were recorded using surface electrodes. Under dynamic conditions, the relationship between muscle force and EMG signal is nonlinear and time-dependent. Thus, we adopted EMG amplitude estimation with adaptive smoothing window length. This approach improved the prediction ability of muscle force in the TDANN training. The experimental results indicated that dynamic tendon forces from EMG signals could be predicted using the TDANN, in vivo.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.57-67
• /
• 2024

This paper explores the potential of electromyography (EMG) as a means of gesture recognition for user input in gesture-based interaction. EMG utilizes small electrodes within muscles to detect and interpret user movements, presenting a viable input method. To classify user gestures based on EMG data, machine learning techniques are employed, necessitating the preprocessing of raw EMG data to extract relevant features. EMG characteristics can be expressed through formulas such as Integrated EMG (IEMG), Mean Absolute Value (MAV), Simple Square Integral (SSI), Variance (VAR), and Root Mean Square (RMS). Additionally, determining the suitable time for gesture classification is crucial, considering the perceptual, cognitive, and response times required for user input. To address this, segment sizes ranging from a minimum of 100ms to a maximum of 1,000ms are varied, and feature extraction is performed to identify the optimal segment size for gesture classification. Notably, data learning employs overlapped segmentation to reduce the interval between data points, thereby increasing the quantity of training data. Using this approach, the paper employs four machine learning models (KNN, SVC, RF, XGBoost) to train and evaluate the system, achieving accuracy rates exceeding 96% for all models in real-time gesture input scenarios with a maximum segment size of 200ms.

• Journal of Biomedical Engineering Research
• /
• v.27 no.5
• /
• pp.237-244
• /
• 2006

This study proposes an approach to the performance improvement of EMG(Electromyogram) pattern recognition. MFCC(Mel-Frequency Cepstral Coefficients)'s approach is molded after the characteristics of the human hearing organ. While it supplies the most typical feature in frequency domain, it should be reorganized to detect the features in EMG signal. And the dynamic aspects of EMG are important for a task, such as a continuous prosthetic control or various time length EMG signal recognition, which have not been successfully mastered by the most approaches. Thus, this paper proposes reorganized MFCC and HMM-GMM, which is adaptable for the dynamic features of the signal. Moreover, it requires an analysis on the most suitable system setting fur EMG pattern recognition. To meet the requirement, this study balanced the recognition-rate against the error-rates produced by the various settings when loaming based on the EMG data for each motion.

• Korean Journal of Applied Biomechanics
• /
• v.18 no.2
• /
• pp.105-114
• /
• 2008

The purpose of this study was to compare of muscles in clean and jerk techniques between 69kg class(n=3), 58kg class(n=3) in high school female weight lifters using EMG(electromyographic) system. EMG analysis were executed on 6 major muscles and dividing clean and jerk techniques into 6 phases. In that result, in the difference by weight, it was shown that EMG value increased gradually as the weight is raised of all muscles group & phases in 58kg class. In EMG signal scale by classes, it was shown that EMG signal scale didn't increase according to class & weight. In the result of this study, that EMG value was inconsistent in 69kg class is showing that the consideration of the technical factor together with muscle power has positive affect more on the performance improvement in the heavy class.

• Korean Journal of Applied Biomechanics
• /
• v.15 no.1
• /
• pp.197-206
• /
• 2005

The purpose of this study was to quantify the maximum EMG levels and determine if there are differences in these EMG levels with respect to different knee flexion angles. Eight university students with no known musculoskeletal disorders were recruited as the participants. The maximum voluntary isometric knee extensions and flexions were taken from each participant sat on the isokinetic exercise machine (Cybex 340) at five different knee flexion angles ($10^{\circ}$, $30^{\circ}$, $50^{\circ}$, $70^{\circ}$, $90^{\circ}$) After surface electrodes were attached to rectus femoris, vastus medialis, vastus laterlis, biceps femoris, and semitendinosus, maximum EMG levels at five different knee flexion angles were measured. The results showed that there was no significant difference in maximum EMG levels among five different knee flexion angles. Although there was no significant difference in EMG levels and were some variations among different knee flexion angles, the EMG signals of quadriceps in extension and biceps femoris in flexion were the greatest at $30^{\circ}$. It seems that different joint angles or relative locations of body segments might affect the magnitude of EMG levels. Because the maximum EMG levels could change with a different knee flexion angle, an attempt should be made to more accurately measure these values. If then, %MVIC measure provides more reliable data and is most appropriate for EMG normalization.

• Physical Therapy Korea
• /
• v.10 no.1
• /
• pp.37-50
• /
• 2003

The purpose of this study was to compare visual analogue scale (VAS), pain threshold (PT), $%RMS_{RVC}$, and EMG gaps before and after applying transcutaneous electrical nerve stimulation (TENS) on the upper trapezius muscle at the patients with myofascial pain syndrome (MPS). The subjects were 4 men and 10 women composed of both the inpatients and outpatients who were diagnosed as MPS at Wonju Medical Center. VAS and PT measurements were performed to assess the subjective pain level. The reference voluntary contraction (RVC) test was performed for 15 seconds for normalization on the bilateral trapezius muscle using surface electromyography (sEMG). After 3-minute resting time, the EMG signal was recorded while performing a typing activity for 2 minutes and then TENS was applicated with a comfortable intensity for 10 minutes. The EMG activity of the upper trapezius muscle was recorded during typing for 2 minutes. The results of study were as follows: 1) VAS score was significantly decreased on the more painful side after treatment, however, it was not significantly different on the less painful side. 2) PT was increased after treatment on both sides, however, it was not significantly different between before and after the TENS application. 3) The EMG activity during typing was significantly decreased after treatment, and 4) The EMG gaps were significantly increased after TENS treatment compared to before it. Consequently, the study showed that TENS was effective in decreasing VAS, $%RMS_{RVC}$, and in increasing EMG gaps. The EMG gap analysis could be a useful method to measure pain in patients with MPS in the upper trapezius.

• Science of Emotion and Sensibility
• /
• v.6 no.1
• /
• pp.27-32
• /
• 2003

The objectives of this study the effects of fabric noise from active wear on electrodiagnosis(EMG), to examine the effects on subjective sensation, and to relate the EMG results and the subjective sensation.. Five nylon water repellent taffeta fabrics were rubbed each other and the fabric noise were recorded. EMG was done from 10 female university students and electrodes were attached on each participant's upper arm and lower am. The subjective sensation was measured by FMME(Free Modulus Magnitude Estimation). The EMG values from upper arm showed higher voltage than those from lower arm, and the differences between values with fabric sound and without were larger at upper arm than those at lower am. EMG decreased when fabric sound was evaluated soft and pleasant, however It increased in proportion as fabric sound was evaluated loud and sharp. The predicted models for subjective sensation using physical sound properties and EMG results were well explained except roughness. Pleasantness was well predicted by EMG at upper am and EMG at lower arm, as the result, it was explained that the lower the EMG, the more pleasant the participant.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.6-8
• /
• 2004

The objective of this paper is to develop a powered wheelchair controller based on EMG for users with high-level spinal cord injury using a proportional control scheme. An advantage of EMG is relative convenience of acquisition by a surface electrode to users. Direction information can be easily extracted from two EMG channels and force information can be acquired by proportional relationship between the amplitude of EMG and user's power, respectively. Pattern classification algorithm is a threshold method with a supervised learning process. Furthermore, the emergency situation can be avoided using an interrupt function. We evaluated the performance of powered wheelchair controller by navigating a pre-defined path with three non-handicapped people. The results show the feasibility of EMG as an input interface for powered wheelchair and other devices for the seriously disabled.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.2
• /
• pp.291-296
• /
• 2014

Most of the statistical signal analysis processed in the time domain and the frequency domain are based on the assumption that the signal is weakly stationary(wide sense stationary). Therefore, it is necessary to know whether the surface EMG signals processed in the statistical basis satisfy the condition of weak stationarity. The purpose of this study is to analyze the accuracy of the Run-test, modified Run-test, RA(reverse arrangement)-test, and modified RA-test for assessing surface EMG signal stationarity. Six stationary and three non-stationary signals were simulated by using sine wave, AR(autoregressive) modeling, and real surface EMG. The simulated signals were tested for stationarity using nine different methods of Run-test and RA-test. The results showed that the modified Run-test method2 (mRT2) classified exactly the surface EMG signals by stationarity with 100% accuracy. This finding indicates that the mRT2 may be the best way for assessing stationarity in surface EMG signals.